`
`Joseph L Rau PhD RRT FAARC
`
`Introduction
`Compliance or Adherence?
`Defining Adherence
`Types of Nonadherence
`Measurement of Adherence With Aerosol Regimens
`General Studies of Adherence With Aerosol Therapy
`Correct Aerosol Device Technique
`Complexity of Inhalation Regimen
`Dosing Frequency
`Combination Formulations of Inhaled Drugs
`Route of Administration: Oral vs Aerosol
`Type of Inhaled Medication: Inhaled Corticosteroids vs 2 Agonists
`Patient Awareness of Monitoring and Effect of Feedback on Monitoring
`Patient Beliefs, Sociocultural, and Psychological Factors
`Summary: Improving Adherence With Aerosol Therapy
`
`Patient adherence with prescribed inhaled therapy is related to morbidity and mortality. The terms
`“compliance” and “adherence” are used in the literature to describe agreement between prescribed
`medication and patient practice, with “adherence” implying active patient participation. Patient
`adherence with inhaled medication can be perfect, good, adequate, poor, or nonexistent, although
`criteria for such levels are not standardized and may vary from one study to another. Generally,
`nonadherence can be classified into unintentional (not understood) or intentional (understood but
`not followed). Failing to understand correct use of an inhaler exemplifies unintentional nonadher-
`ence, while refusing to take medication for fear of adverse effects constitutes intentional nonad-
`herence. There are various measures of adherence, including biochemical monitoring of subjects,
`electronic or mechanical device monitors, direct observation of patients, medical/pharmacy records,
`counting remaining doses, clinician judgment, and patient self-report or diaries. The methods cited
`are in order of more to less objective, although even electronic monitoring can be prone to patient
`deception. Adherence is notoriously higher when determined by patient self-report, compared to
`electronic monitors. A general lack of adherence with inhaled medications has been documented in
`studies, and adherence declines over time, even with return clinic visits. Lack of correct aerosol-
`device use is a particular type of nonadherence, and clinician knowledge of correct use has been
`shown to be imperfect. Other factors related to patient adherence include the complexity of the
`inhalation regimen (dosing frequency, number of drugs), route of administration (oral vs inhaled),
`type of inhaled agent (corticosteroid adherence is worse than with short-acting 2 agonists), patient
`awareness of monitoring, as well as a variety of patient beliefs and sociocultural and psychological
`factors. Good communication skills among clinicians and patient education about inhaled medica-
`tions are central to improving adherence. Key words: compliance, adherence, aerosol, metered-dose
`inhaler, MDI, dry powder inhaler, DPI.
`[Respir Care 2005;50(10):1346 –1356. © 2005 Daedalus En-
`terprises]
`
`1346
`
`RESPIRATORY CARE • OCTOBER 2005 VOL 50 NO 10
`
`Liquidia's Exhibit 1033
`Page 1
`
`
`
`DETERMINANTS OF PATIENT ADHERENCE TO AN AEROSOL REGIMEN
`
`Introduction
`
`The importance of patient adherence to prescribed med-
`ication therapy lies in the documented relationship of poor
`adherence to increased morbidity and even mortality.1–3
`Bauman et al found significantly worse asthma morbidity
`among children when they or their caregivers scored high
`on measures of nonadherence with therapy.2 Williams et al
`found that adherence to inhaled corticosteroid therapy,
`based on medical/pharmacy records, was approximately
`50% in a large group of asthmatics, and negatively corre-
`lated with the number of emergency department visits.3
`They also reported that each 25% increase in the propor-
`tion of time without inhaled corticosteroid medication re-
`sulted in a doubling of the rate of asthma-related hospi-
`talization. Milgrom et al found that median compliance
`with inhaled corticosteroids among asthmatic children was
`13.7% for those having exacerbations and 68.2% for those
`who did not.4
`
`Compliance or Adherence?
`
`There are 2 terms used in the literature to refer to how
`well a patient follows a prescribed regimen of drug dosing
`or any prescribed therapy: adherence and compliance. The
`latter term seems to be favored more recently in the liter-
`ature, and this may be because of differences in the exact
`meaning of the 2 terms. While both terms describe agree-
`ment between a patient’s actions and prescribed therapy,
`“compliance” has the connotation of giving in to a request
`or demand; “adherence” on the other hand connotes stay-
`ing attached or staying firm in supporting or approving,
`based on definitions in a standard Webster’s dictionary.5
`“Adherence” thereby seems to imply a patient’s choice to
`follow prescribed therapy, while “compliance” implies a
`certain passivity to another’s request. In fact a synonym
`for “compliant” in one dictionary consulted is “obedient.”5
`In a 1995 publication, Tashkin defined compliance “sim-
`ply as following the instructions of the health-care provid-
`er.”6 As a result, “compliance” conjures a view of the
`patient as a passive participant following orders. In con-
`trast, “adherence” describes an active patient who is an
`empowered partner in his or her care.7 Aside from political
`correctness, it seems to make sense to have a patient who
`
`Joseph L Rau PhD RRT FAARC is Professor Emeritus, Cardiopulmo-
`nary Care Sciences, Georgia State University, Atlanta, Georgia.
`
`Joseph L Rau PhD RRT FAARC presented a version of this article at the
`36th RESPIRATORY CARE Journal Conference, Metered-Dose Inhalers and
`Dry Powder Inhalers in Aerosol Therapy, held April 29 through May 1,
`2005, in Los Cabos, Mexico.
`
`Correspondence: Joseph L Rau PhD RRT FAARC, 2734 Livsey Trail,
`Tucker GA 30084. E-mail: joerau@comcast.net.
`
`actively desires to work with a health-care provider in-
`stead of one who follows directions with little interest in
`taking responsibility for the process. In an editorial ac-
`companying a study on patient compliance, Mellins and
`associates commented that “there is a growing recognition
`that to improve significantly the way in which they use
`medicines and otherwise manage disease, patients must be
`actively involved in the process of determining the thera-
`peutic plan.”8 Throughout this review, the terms “compli-
`ance” and “adherence” will correspond to those used in the
`particular studies described. Otherwise the term “adher-
`ence” will be used to describe agreement between pre-
`scription and practice.
`
`Defining Adherence
`
`Rand and Wise define “adherence” as “the degree to
`which patient behaviors coincide with the clinical recom-
`mendations of health-care providers.”9 They note that this
`definition is too broad and call for adherence to be situ-
`ationally defined, with good adherence explicitly delin-
`eated. They also note that there is no gold standard for
`“good” or “acceptable” adherence. For example, adequate
`adherence may describe asthma-clinic patients who use
`40% of the prescribed medication and are symptom-free
`and controlled. However, a subject in a research study who
`takes 60% of prescribed doses may be considered nonad-
`herent.9 An example of the type of definition of adherence
`called for by Rand and Wise can be found in the context
`of a study by Tashkin et al, who used metered-dose inhaler
`(MDI) canister-weight criteria to define compliance rat-
`ings.10 For example, using calculated grams of medication
`per day, ⬎ 0.45 g/d might be “over-compliance,” 0.35–
`0.45 g/d “good compliance,” and so forth. Such a method
`gives a specific criterion (g/d) to rate degrees of compli-
`ance.
`
`Types of Nonadherence
`
`Nonadherence with therapy takes multiple forms, rang-
`ing from incomplete to total nonuse. The various types of
`nonadherence with prescribed therapy can be broadly cat-
`egorized into 2 types: unintentional (not understood), and
`intentional (understood but not followed).11 Table 1 gives
`a more detailed outline of potential factors that can pre-
`dispose to these types of nonadherence.11–13 Unintentional
`nonadherence includes misunderstanding the prescribed
`regimen, incorrect aerosol device technique, or language
`barriers. Intentional nonadherence can be caused by pa-
`tient beliefs (eg, that drug therapy is ineffective, unneces-
`sary, or dangerous), forgetfulness, stress, busy lifestyle, or
`complex, demanding aerosol regimens. Of the two, unin-
`tentional nonadherence may be easier to remedy.
`
`RESPIRATORY CARE • OCTOBER 2005 VOL 50 NO 10
`
`1347
`
`Liquidia's Exhibit 1033
`Page 2
`
`
`
`DETERMINANTS OF PATIENT ADHERENCE TO AN AEROSOL REGIMEN
`
`Table 1. General Types of Nonadherence to Prescribed Aerosol
`Therapy and Potential Factors That Can Predispose to Each
`Type*
`
`Unintentional: Patient does not understand therapy correctly
`Misunderstanding prescribed drug regimen (poor doctor-patient
`communication)12
`Incorrect aerosol device technique
`Language barriers
`Intentional: Patient understands therapy but does not adhere correctly
`Patient beliefs
`I do not really require regular medication
`I am not really sick
`I gain attention from parents, am kept at home (children)
`The medication is too expensive
`I have concern about adverse effects
`I do not perceive effect from the medication
`Forgetfulness
`Stress and busy lifestyle
`Complex, demanding aerosol regimens
`Psychological factors (eg, depression)13
`
`*Two general categories of nonadherence are based on Reference 11.
`
`Measurement of Adherence With Aerosol Regimens
`
`There are a number of methods for measuring congru-
`ence of patient behavior with prescribed aerosol therapy,
`which are listed in Table 2.9,11 These methods differ sub-
`stantially in the degree of accuracy and objectivity with
`which patient adherence can be determined. In general,
`direct measures of patient behavior, such as direct obser-
`vation or electronic inhaler monitors, give more accurate,
`valid measures than indirect methods such as patient dia-
`ries, self-report, or clinician’s judgment.9,11,14 There are
`several electronic monitors that have been reported in the
`literature for use with MDIs or dry powder inhalers (DPIs).
`The “nebulizer chronolog” device and the “Doser Clin-
`ical Trials” device have been used with MDIs.15–17 The
`nebulizer chronolog is a microprocessor device built into
`the sleeve housing an MDI; it records the date and time of
`each inhaler actuation, by activation of a microswitch.4,15
`The Doser Clinical Trials device is described as an inex-
`pensive pressure-activated device, also used with MDIs.17
`It is a round, flat device secured to the top of the MDI
`canister, and it records only the number of daily uses over
`a period of 45 days.18 A similar MDI electromechanical
`counter was reported by Yeung et al.19 The Electronic
`Diskhaler allows monitoring of the Diskhaler DPI, by re-
`cording drug blister piercing and airflow through the in-
`haler.20 A similar device, the Turbohaler Inhalation Com-
`puter has been used with the Turbohaler DPI, known as the
`Turbuhaler in the United States.13 An electronic adherence
`monitor has also been reported for the Diskus DPI.21 It
`should be noted that not all electronic monitors guarantee
`
`actual inhalation of medication by patients. With the neb-
`ulizer chronolog, medication can be sprayed into the air, or
`the switch flicked manually. The Electronic Diskhaler
`records both blister perforation and airflow, which gives
`some indication that inhalation occurred following DPI
`loading.20
`Tashkin et al investigated adherence with aerosol ther-
`apy, using the nebulizer chronolog, in comparison with
`canister weighing and patient self-report with a group of
`patients with chronic obstructive pulmonary disease
`(COPD).10 Their study found that both canister weights
`and self-report overestimated adherence with prescribed
`therapy among patients who were not informed of the
`nebulizer chronolog’s recording ability (Fig. 1).
`Rand et al also used the nebulizer chronolog to compare
`adherence to a 3-times-daily use of 2 MDI inhalations of
`ipratropium or placebo by patient self-report at follow-up
`and canister-weight-change over a 4-month period.15 Both
`self-report and canister-weighing overestimated correct in-
`haler use, compared to nebulizer chronolog measures. Neb-
`ulizer chronolog data showed that only 15% of the sub-
`jects used the MDI an average of 2.5 or more times per
`day, as prescribed. In contrast, 73% of subjects self-re-
`ported correct daily inhaler use. Canister-weighing over-
`estimated correct inhaler use as prescribed for 61% of
`participants, correctly estimated use for 39% (although not
`always as prescribed), and underestimated use for 0%.
`Nebulizer chronolog data also showed that 14% of sub-
`jects actuated their inhalers more than 100 times in a 3-hour
`interval, often before clinic visits, a practice known as
`“dumping,” or the “parking lot phenomenon.”9,15 Canister
`weighing cannot differentiate correct use from wasted med-
`ication.
`Milgrom et al also looked at patient compliance to both
` agonists and inhaled corticosteroids, using the nebulizer
`chronolog versus patient diaries.4 Figure 2 shows a sum-
`mary of the compliance data for both inhaled medications
`over 13 weeks. Diary reports claimed a median use of 
`agonists of 78.2% of prescribed dose, and a steroid use of
`95.4%. Data from the nebulizer chronolog giving time-
`corrected compliance (doses taken within the correct time
`window) showed 48% for  agonists and 32% for inhaled
`steroids. Only 
`2 agonists taken on a fixed schedule (2 or
`3 times a day or every 6 hours) were included in the
`analysis. Similar results for electronic monitors in com-
`parison with patient reports, canister weight, and remain-
`ing dose counts have been reported in other studies.18,22–23
`A study by Burrows et al showed that patient self-report-
`ing also overestimated adherence when compared to data
`from pharmacy-dispensing records for nebulized dornase
`alfa in cystic fibrosis patients.24 Based on the comparisons
`cited, it is relevant to note that results of different studies
`can depend at least partly on which measure of aerosol
`adherence is employed.
`
`1348
`
`RESPIRATORY CARE • OCTOBER 2005 VOL 50 NO 10
`
`Liquidia's Exhibit 1033
`Page 3
`
`
`
`DETERMINANTS OF PATIENT ADHERENCE TO AN AEROSOL REGIMEN
`
`Table 2. Methods of Measuring Adherence With Prescribed Aerosol Drug Therapy, Based On Measures Noted in the Literature*
`
`Method
`
`Example
`
`Strengths
`
`Limitations
`
`Biochemical measures
`
`Analysis of blood, urine, or
`secretions to measure drug level
`
`Accurate
`Objective
`
`Medication/device monitors
`
`Electronic monitor records date
`and time of inhaler use
`
`Accurate
`Objective
`
`Expensive
`Intrusive
`Limited drug tests
`Limited to recent drug therapy
`
`Cannot tell if patient actually
`received dose
`Expensive
`Possible alteration of patient habits?
`
`Observation of device
`technique
`
`Direct review of patient
`performance with aerosol
`device, usually periodic
`
`Accurate with training of
`observer
`Simple
`Objectively based
`
`Limited to time of observation
`Limited to device-use only, not
`dose schedule
`Requires staff time
`
`Medical/pharmacy records
`
`Retrospective review of patient
`records or refills
`
`Objective
`Relatively simple to obtain
`
`Monitoring remaining dose
`counts or medication
`
`MDI canister weighing
`DPI doses left
`SVN doses or solution packages
`left
`
`Simple
`Objective
`Low cost
`
`Clinical judgment of provider
`
`Global judgment of health-care
`provider during clinic visits
`
`Quick
`Low cost
`
`Time required to obtain patient data
`Limited to detecting nonrefills
`No information on correct patient
`use or scheduling of drug with
`refills
`
`Possible patient deceit by wasting
`doses
`No information on actual dosing
`schedule
`Requires staff time
`
`Low validity and reliability14
`
`Patient self-report
`
`Periodic recall survey or interview
`Patient diary
`
`Fast for health-care provider
`Low cost
`Ease of use
`
`Vulnerable to patient error or
`deceit15
`
`*The methods are listed in order of relative accuracy, from greater to less. (Adapted from References 9 and 11.)
`MDI ⫽ metered-dose inhaler
`DPI ⫽ dry powder inhaler
`SVN ⫽ small-volume nebulizer
`
`General Studies of Adherence With Aerosol Therapy
`
`The general lack of adherence with prescribed aerosol
`therapy has been documented in a number of studies,
`
`Fig. 1. Percentage of adherence with prescribed metered-dose
`inhaler (MDI) medication among patients with chronic obstructive
`pulmonary disease, determined with 3 methods of monitoring: neb-
`ulizer chronolog (electronic MDI monitor), MDI canister weight
`change, and patient self-report. (Based on data from Reference 10.)
`
`as
`as well
`including patients with asthma,25–27
`COPD.15,28 –29 Rand and associates documented that
`COPD patients had poor adherence with prescribed
`3-times-daily MDI therapy, as measured with the neb-
`ulizer chronolog.15 Fewer than 20% of 70 patients used
`their MDIs an average of 2.5–3 times per day as in-
`structed, although almost 95% reported correct use as
`prescribed. Jo´nasson et al found a decline in adherence
`with twice-daily inhaled budesonide and placebo in
`mildly asthmatic children over a 27-month period of
`monitoring remaining doses with Turbuhaler DPIs.25 A
`disturbing finding from Mawhinney et al was that only
`1 subject out of 34 in a clinical trial of 2 nonbronchodi-
`lator anti-asthma drugs (cromolyn-like and corticoste-
`roid agents) was compliant with prescribed use, as mea-
`sured with a nebulizer chronolog for MDI.27 Such
`findings raise questions about the validity of clinical
`trials, when patient medication use is thought to be best.
`
`RESPIRATORY CARE • OCTOBER 2005 VOL 50 NO 10
`
`1349
`
`Liquidia's Exhibit 1033
`Page 4
`
`
`
`DETERMINANTS OF PATIENT ADHERENCE TO AN AEROSOL REGIMEN
`
`Table 3.
`
`Relation of Dosing Frequency to Compliance With a
`Prophylactic Inhaled Medication in Children Monitored
`With a Nebulizer Chronolog Monitor
`
`Prescribed Frequency
`(doses/day)
`
`Reported Compliance
`(% of days)
`
`Monitored Compliance
`(% of days)
`
`2
`3
`4
`
`96
`90
`69
`
`71
`34
`18
`
`(Adapted from Reference 16.)
`
`poor patient use with MDIs, compared to DPIs.38 In their
`study, 24% of patients used MDIs poorly; failure to cor-
`rectly perform essential steps with the Aerolizer, Turbu-
`haler, and Diskus was 17%, 23%, and 24%, respectively.
`Use of a large-volume spacer reduced poor MDI use from
`24% to only 3% of patients.
`
`Complexity of Inhalation Regimen
`
`The complexity of an inhalation regimen in managing
`airway disease can depend on the frequency with which an
`inhaled medication must be taken, the number of medica-
`tions to be taken, and whether different types of aerosol
`devices must be used (eg, a nebulizer for one drug and a
`DPI for another).
`
`Dosing Frequency
`
`Medication adherence has been linked to the frequency
`with which a drug must be taken, for both oral and in-
`haled-drug regimens. Eisen et al used electronically mon-
`itored pill containers to measure patient adherence with
`antihypertensive medication.39 Their study found that ad-
`herence improved from 59% with a 3-times daily regimen
`to 83.6% with a once-daily regimen. Similarly, Cramer et
`al found the mean (SD) adherence rate for oral antiepilep-
`tic drugs was 87% (11), 81% (17), 77% (12), and 39% (24)
`for daily, twice-a-day, 3-times-a-day, and every-6-hours
`dosing, respectively, using an electronic pill bottle dis-
`pensing system.40 Prescribed frequency of drug use simi-
`larly affects inhaled medications. Coutts et al performed a
`pilot trial of the nebulizer chronolog to study compliance
`with inhaled prophylactic medication (corticosteroids) in
`children.16 Table 3 gives the results of their study for
`twice-a-day, 3-times-a-day, and every-6-hours dosing fre-
`quencies, with patient self-report and nebulizer chronolog
`monitoring data. A “compliant day” was defined as one
`with the correct number of puffs at appropriate times. As
`reported for oral medications, compliance declined with
`increasing frequency of use. Mann et al assigned patients
`to 2 groups, with group A taking 4 inhalations of fluni-
`
`Fig. 2. Percentage of prescribed doses of inhaled 2 agonists and
`inhaled corticosteroids over 13 weeks among asthmatic children.
`The chronolog record is the raw percentage of prescribed doses
`taken. “Doses taken at correct times” represents the percentage
`of prescribed doses with the correct number of puffs taken within
`the correct time window. The error bars indicate the minimum and
`maximum percentages. The boxes indicate the lower and upper
`quartiles (25% and 75% of subjects). The thick black horizontal
`bars indicate the medians of values reported or measured. (Adapted
`from Reference 4, with permission.)
`
`Correct Aerosol Device Technique
`
`Lack of adherence to aerosol therapy can be due to lack
`of understanding correct aerosol device or drug use, and
`was termed “unintentional” nonadherence in Table 1. Far-
`ber et al found that 23% of parents (n ⫽ 131) misunder-
`stood the role of their asthmatic child’s inhaled anti-in-
`flammatory medication, believing that it was for treatment
`of symptoms after they occurred, not for prevention. This
`was associated with decreased adherence to its daily use.12
`A number of studies have documented problems pa-
`tients have using aerosol devices and common patient er-
`rors, particularly with MDIs.30 –34 While “press and breathe”
`seems simple when using an MDI, many patients lack the
`coordination for the split-second timing required between
`actuating the MDI and beginning a slow inhalation.31 Sub-
`optimal therapeutic response and poor control of airway
`disease can result from faulty technique.31,35
`Problems with patient use of aerosol devices can be
`worsened by inadequate knowledge of correct device use
`among health-care professionals. A study by Hanania et al
`of medical personnel’s knowledge of MDIs, MDIs with
`spacers, and a DPI had a mean ⫾ SD knowledge score of
`67 ⫾ 5% for respiratory therapists, 48 ⫾ 7% for house
`staff physicians, and 39 ⫾ 7% for registered nurses.36 A
`similar study of the same types of aerosol devices found
`that pharmacists lacked adequate knowledge to properly
`instruct patients in inhaler use.37 DPIs can remove the
`need for hand-breath coordination with MDIs (a common
`problem) because DPIs are breath-actuated. However, a
`recent study by Melani et al found similar percentages of
`
`1350
`
`RESPIRATORY CARE • OCTOBER 2005 VOL 50 NO 10
`
`Liquidia's Exhibit 1033
`Page 5
`
`
`
`DETERMINANTS OF PATIENT ADHERENCE TO AN AEROSOL REGIMEN
`
`solide twice a day, and group B taking 2 inhalations every
`6 hours.41 Correct use was 8 inhalations per day, for either
`group. Both groups had a run-in period with 4 inhalations
`twice-a-day. Compliance did not change for group A (twice-
`a-day dosing) from the run-in period. The percentage of
`days with less than 8 inhalations for group B increased
`from 20.2 ⫾ 40.3% during the run-in, to 57.1 ⫾ 49.6%
`with the change to every-6-hours dosing. The mean num-
`ber of daily inhalations in group B decreased from 7.9 ⫾
`2.5% to 6.8 ⫾ 3.1% (p ⬍ 0.01) between the 2 time peri-
`ods.
`
`Combination Formulations of Inhaled Drugs
`
`Combining 2 inhaled drugs into one formulation for
`inhalation could theoretically halve the number of times
`needed for drug administration, and thereby reduce the
`complexity for drug inhalation. Bosley et al reported a
`study in 1994 that compared separate inhalation of a cor-
`ticosteroid (budesonide) and a short-acting 
`2 agonist (ter-
`butaline) to a combination formulation of the two.42 All
`drugs were given using the Turbuhaler DPI and were to be
`taken twice daily. Adherence was monitored electronically
`with the Turbuhaler Inhalation Computer. When the 2 drugs
`were inhaled separately, compliance was similar for both
`the 
`2 agonist and the corticosteroid, at about 60 –70%.
`This was somewhat surprising, since compliance with in-
`haled corticosteroid therapy is often thought to be poor,
`and worse than with bronchodilators.43 In addition, com-
`pliance was no better in patients using the combined for-
`mulation. These results may have been due to use of a
`short-acting 
`2 agonist, which requires more frequent use
`per day than a long-acting agent.
`A study by Stoloff et al compared medication-refill per-
`sistence with (1) the corticosteroid fluticasone propionate
`and the long-acting 
`2 agonist salmeterol in combination
`in a single inhaler; (2) fluticasone propionate and salme-
`terol inhaled separately from 2 inhalers; (3) fluticasone
`propionate and montelukast taken together (inhaled, oral);
`and (4) fluticasone propionate and montelukast each taken
`singly as monotherapy.44 The cohort that used fluticasone
`plus salmeterol from a single inhaler had significantly bet-
`ter adherence (4.06 refills per 12-month period) than the
`other cohorts that used fluticasone (2.35 refills per 12-
`month period in the group that inhaled fluticasone and
`salmeterol from separate inhalers; 1.83 refills per 12-month
`period in the group that used fluticasone plus montelukast;
`and 2.27 refills per 12-month period in the group that used
`fluticasone alone). The combination formulation (flutica-
`sone plus salmeterol in one inhaler) had refill persistence
`similar to that of the oral leukotriene modifier montelukast
`taken alone (4.51 refills per 12-month period), although
`montelukast monotherapy had the highest refill persistence.
`
`Fig. 3. Percentage of patient compliance with oral theophylline
`versus inhaled corticosteroids and cromolyn sodium, based on
`pharmacy claims data with a group of asthmatic subjects. The
`error bars represent the standard deviations. (Based on data from
`Reference 45.)
`
`The difference in results between the study by Bosley et
`al,42 with a short-acting bronchodilator, and that of Stoloff
`et al,44 with a long-acting bronchodilator, may well be due
`to the frequency of administration. In addition, the sim-
`plest form of drug therapy in the Stoloff et al study was
`oral montelukast taken as monotherapy, which had the
`highest adherence.44 The recommended dosage for mon-
`telukast is once daily, taken as a pill.
`
`Route of Administration: Oral Versus Aerosol
`
`Taking a dose of medication as a pill is reasonably
`simple and quick, assuming normal swallowing ability and
`consciousness. In terms of time needed for a dose, the
`MDI and DPI are closest among the various aerosol de-
`vices to pill-taking, although the multiple steps needed for
`correct use of either (MDI: shaking, exhaling, actuating,
`slow inhalation, and breath-hold; DPI: multi-step prepara-
`tion, breath-hold) certainly requires a minute or more. In
`terms of simplicity, I would argue that a pill taken orally
`is far simpler than MDI or DPI use.
`Kelloway et al used medical records together with phar-
`macy claims data to measure the compliance of patients
`prescribed oral theophylline and inhaled medications.45 All
`subjects used oral theophylline; 97% used inhaled corti-
`costeroids and 8.4% used inhaled cromolyn sodium, with
`5% taking both inhaled cromolyn and inhaled steroids.
`Both theophylline and inhaled corticosteroid dosing regi-
`mens ranged between 2 and 3 times daily. Cromolyn is
`usually prescribed on a 4-times-daily basis. As shown in
`Figure 3, the highest compliance was with oral theophyl-
`line (79 ⫾ 34%), with inhaled corticosteroid and cromolyn
`at 54 ⫾ 43% and 44 ⫾ 34%, respectively. Since only a
`few patients had 2 inhaled formulations, the data from
`Kelloway et al suggest better adherence with oral drugs
`than with inhaled drugs. Compliance for oral theophylline
`was similar when patients were stratified into age groups
`of 12–17 years versus 18 – 65 years. Inhaled corticosteroid
`compliance was 30% in the younger group and 57% in the
`
`RESPIRATORY CARE • OCTOBER 2005 VOL 50 NO 10
`
`1351
`
`Liquidia's Exhibit 1033
`Page 6
`
`
`
`DETERMINANTS OF PATIENT ADHERENCE TO AN AEROSOL REGIMEN
`
`Table 4.
`
`Results on Adherence With Oral Versus Inhaled Medications From a Number of Studies
`
`First Author, Year
`
`Condition
`
`Age Range
`(y)
`
`Measurement
`
`Adherence (%)
`
`Oral Drug
`
`Inhaled Drug
`
`Kelloway 199445
`
`Asthma
`
`12–65
`
`Sherman 200146
`Maspero 200147
`Bukstein 200348
`Jones 200349
`
`Asthma
`Asthma
`Asthma
`Asthma
`
`Pediatric
`6–11
`6–11
`6–55
`
`Medical records
`Pharmacy data
`Prescription refills
`Patient interview
`Patient self-report
`Pharmacy claims
`
`Theophylline 79 ⫾ 34
`
`Montelukast 59 (95% CI 48–65)
`Montelukast 82
`Montelukast 78
`Leukotriene modifier 67.7
`
`Corticosteroid 54 ⫾ 43
`Cromolyn 44 ⫾ 34
`Fluticasone 44 (95% CI 35–50)
`Beclomethasone 45
`Cromolyn 42
`Inhaled corticosteroid 33.8
`Long-acting 
`2 agonist 40.0
`
`CI ⫽ confidence interval
`
`older group, indicating an age difference in that particular
`study.
`Table 4 summarizes results from the study by Kelloway
`et al,45 together with other studies46 – 49 that compared pa-
`tient adherence with oral versus inhaled drug therapy.
`Unfortunately, the studies listed in Table 4 all include
`inhaled therapy that must be taken multiple times daily.
`With the exception of the study by Kelloway et al, in
`which theophylline was prescribed, all of the other studies
`examined use of leukotriene modifiers, and most of these
`were the once-daily montelukast taken orally. Thus, there
`is some confounding of results between route of adminis-
`tration (oral vs inhaled) and frequency of dosing, with
`higher frequency of dosing for the inhaled drugs.
`
`Type of Inhaled Medication:
`Inhaled Corticosteroids Versus 2 Agonists
`
`There is a perception among clinicians that patient ad-
`herence with prescribed inhaled corticosteroids is worse
`than with inhaled 
`2 agonists. This has been attributed to
`the absence of immediate relief or perceptible effect from
`inhaled corticosteroids, compared to short-acting 
`2 ago-
`nists.43 A 2000 literature review by Cochrane et al of
`compliance with inhaled corticosteroids noted that studies
`have shown that patients took the recommended dose on
`20 –73% of days.50 Bosley et al compared a combination
`corticosteroid and 
`2 agonist inhaled formulation with sep-
`arate delivery and found no difference in compliance when
`the 2 drugs were taken separately.42 At least 2 other stud-
`ies have measured differences in adherence with inhaled
`corticosteroids and 
`2 agonists. Milgrom et al measured
`adherence of children with asthma to regimens of both
`inhaled corticosteroids and 
`2 agonists, using the MDI
`chronolog (also termed the nebulizer chronolog monitor).4
`They found that doses taken within the correct time win-
`dow, as prescribed (the “time-corrected compliance”), were
`48% for 
`2 agonists and 32% for inhaled corticosteroids.
`Median days without medication were 20.4% for 
`2 ago-
`
`Fig. 4. Median percentage of days with no, minimal, or complete
`inhaled-medication use for corticosteroids and 2 agonists among
`24 asthmatic children. (Based on data from Reference 51.)
`
`nists and 24.4% for inhaled steroids. They noted that 25%
`of patients did not take inhaled corticosteroids on more
`than 60% of the days studied.
`Bender et al also found better adherence with 
`2 ago-
`nists than with inhaled corticosteroids.51 The results of
`their study are shown in Figure 4. The studies by Milgrom
`et al4 and Bender et al51 both support the view that inhaled
`corticosteroid adherence appears to be worse than adher-
`ence with inhaled 
`2 agonists. It should be noted that in
`both studies the 
`2 agonists were probably short-acting, as
`opposed to long-acting, although the specific drugs were
`not identified. Since inhaled corticosteroids and short-act-
`ing 
`2 agonists were both prescribed multiple times daily,
`it would not seem that the poorer results with inhaled
`corticosteroids were due to frequency of dosing. It is not
`clear if similar results would be found if adherence with
`inhaled corticosteroids were compared to long-acting 
`2
`agonists.
`
`Patient Awareness of Monitoring and Effect of
`Feedback On Monitoring
`
`Studies of patient adherence with inhaled medications
`have found that informing patients that they are being
`
`1352
`
`RESPIRATORY CARE • OCTOBER 2005 VOL 50 NO 10
`
`Liquidia's Exhibit 1033
`Page 7
`
`
`
`DETERMINANTS OF PATIENT ADHERENCE TO AN AEROSOL REGIMEN
`
`jects who were not informed of the function of the nebu-
`lizer chronolog actuated their inhalers ⬎ 100 times within
`a 3-hour interval on at least one occasion.29 Only 1 of 135
`subjects who had full knowledge of the nebulizer chro-
`nolog’s recording ability did so. Dumping episodes usu-
`ally occurred shortly before a clinic follow-up visit.
`
`Patient Beliefs, Sociocultural,
`and Psychological Factors
`
`In addition to the explicit factors such as understanding
`device use and instructions, complexity of inhalation reg-
`imen, and giving patients feedback on adherence, adher-
`ence can be influenced by a number of personal factors.
`These include health beliefs, such as need for medication,
`severity of disease, and risks of adverse effects, and so-
`ciocultural and psychological factors. Table 5 summarizes
`results from studies that examined the associati