Complementary and Alternative Medicine Use and Adherence

to Asthma Medications Among Latino and Non-Latino White

FamiliesElizabeth L. McQuaid, PhD, ABPP; David A. Fedele, PhD; Sue K. Adams, PhD;Daphne Koinis-Mitchell, PhD; Jessica Mitchell, MS; Sheryl J. Kopel, MSc; Ronald Seifer, PhD;Barbara Jandasek, PhD; Gregory K. Fritz, MD; Glorisa Canino, PhD

From the Bradley/Hasbro Children’s Research Center, Alpert Medical School of Brown University, Providence, RI (Drs McQuaid, Fedele,Koinis-Mitchell, Seifer, Jandasek, and Fritz; and Ms Mitchell and Ms Kopel); Department of Human Development and Family Studies,University of Rhode Island, Kingston, RI (Dr Adams); and Behavioral Sciences Research Institute, University of Puerto RicoMedical SciencesCampus, San Juan, PR (Dr Canino)The authors declare that they have no conflict of interest.Address correspondence to Elizabeth L. McQuaid, PhD, ABPP, Bradley/Hasbro Children’s Research Center, 1 Hoppin St, Providence,RI 02903 (e-mail: Elizabeth_McQuaid@brown.edu).Received for publication December 26, 2012; accepted September 22, 2013.

ABSTRACT

AC

OBJECTIVE: The current study sought to evaluate patterns ofcomplementary and alternative medicine (CAM) use in a sam-ple of Latino and non-Latinowhite (NLW) children with asthmato determine whether parental beliefs about conventional med-ications and barriers to obtaining these medications were relatedto CAM use and to assess whether CAM use was associatedwith decreased adherence to controller medications.METHODS: Participants included 574 families of children withasthma from NLW, Puerto Rican (PR), and Dominican back-grounds fromRhode Island (RI) and from Island PR. All parentscompleted a brief checklist of barriers to medication use and anassessment of CAM approaches. A subsample of 259 familieshad controller medication use monitored objectively for approx-imately 1 month by MDILog (fluticasone propionate), Track-Cap (montelukast), or dosage counter (fluticasone/salmeterolcombination).

CADEMIC PEDIATRICSopyright ª 2014 by Academic Pediatric Association 192

RESULTS: Prevalence of CAM use was high among Latinofamilies. Perceived barriers to obtaining medication wererelated to increased CAM use in PR families from RI. Elevatedmedication concerns were positively associated with CAM useamong NLW and Island PR families. CAM use was positivelyrelated to objective adherence within NLW families, and unre-lated in other groups.CONCLUSIONS: CAM use is common among Latino familieswith asthma. Among some families, CAM use may be initiatedas a way to cope with barriers to obtaining medication or whenparents have concerns about conventional medications. Fam-ilies who report CAM use do not appear to be substitutingCAM for conventional asthma medication.KEYWORDS: adherence; asthma; complementary and alterna-tive medicine

ACADEMIC PEDIATRICS 2014;14:192–199

WHAT’S NEW

This study evaluates associations between complemen-tary and alternative medicine (CAM) use and parentalbeliefs about conventional medications and barriers toobtaining these medications among Latino and non-Latino white (NLW) families of children with asthmaand how this use relates to medication adherence.CAM use was positively related to adherence inNLWs, but its use was unrelated in Latinos.

PEDIATRIC ASTHMA IS associated with impaired qualityof life and increased health care utilization.1 Minority anddisadvantaged populations are disproportionately af-fected.2 Latinos of the Caribbean have particularly highprevalence, morbidity, and health care utilization.3 Thesources of these disparities are multifaceted and complex;they include genetics, individual and family beliefs, healthcare system features, and broad environmental variables

such as housing and pollution.4 Attitudes toward con-ventional medication use and barriers to obtaining pre-scribed medications may be modifiable sources of asthmadisparities.5

National guidelines for asthma management includedaily controllermedication use to prevent symptoms and ex-acerbations for thosewith chronic asthma.6 Research amongethnic minorities consistently demonstrates concernsabout controller medication use7,8 and lower medicationadherence.9 We previously found that controller medicationadherence was low among island Puerto Rican (PR) andLatino families in the United States relative to non-Latinowhite (NLW) families.10 Data indicate greater acceptanceof complementary and alternative medicine (CAM) useamong ethnic minority families with asthma.11 Few studies,however, have examined how preference for CAM mayrelate to adherence to prescribed controller medications.12

CAM is defined as “a group of diverse medical andhealth care systems, practices, and products that are not

Volume 14, Number 2March–April 2014

ACADEMIC PEDIATRICS COMPLEMENTARY AND ALTERNATIVE MEDICINE USE 193

generally considered part of conventional medicine.”13

Prevalence of CAM use among families of children withasthma ranges from 33% to 89% across studies.14 Ratesof CAM use for ethnic minority families with asthma rangebetween 48% and 89%, depending on ethnicity.15,16

Reasons for CAM use in ethnic minority families arelikely multifaceted. In earlier work examining CAM useamong PR families, patterns of acculturation were relatedto medication adherence, with higher adherence amongfamilies whose parents reported a bicultural orientation.16

CAM use may be related to families’ broader ethnoculturalbelief systems regarding health and illness.17

POTENTIAL RELATION OF CAM USE TO

BELIEFS AND BARRIERS

Parental beliefs and concerns about conventional asthmamedication may be associated with CAM use. Studiesincluding Latino8 and mixed-ethnicity samples7,18

indicate that parents often have concerns surrounding theuse of conventional asthma medication, includingimmediate adverse effects, medication dependence, andlong-term effects. These concerns are related to a lower re-ported asthma medication adherence7 and a higher likeli-hood of uncontrolled asthma.19 Studies relating parentalmedication beliefs and child asthma medication adherence,however, have used predominantly NLW samples and havenot assessed CAM use.18

Ethnic minority families may be more likely to face bar-riers to obtaining prescription asthmamedications.20 Theseinclude financial burden, sociocultural factors (eg, limitedEnglish proficiency), and practical constraints (eg, trans-portation difficulties).21 For instance, CAM use has beenfound to be higher among uninsured families and thosewho perceived financial barriers to asthma care.22 In areport from the sample presented in this article, barriersto medication use were high among Island PR familiesrelative to Latino and NLW families.20 These barriersalso may play a role in greater CAM use and lower adher-ence to conventional asthma medications.

CAM AND ADHERENCE TO ASTHMA

MEDICATION

Latino parents of children with asthma may substituteCAM for asthma medications to prevent symptoms.8

CAM may also be used as a first-line treatment for epi-sodes.15 Research regarding CAM use and adherence tocontroller medications has produced mixed findings, how-ever. Positive caregiver beliefs surrounding CAM werelinked to risk for self-reported nonadherence to asthmamedication in a largely minority sample.23 Alternatively,parent report of child medication adherence was not asso-ciated with initiation of CAM within a predominantlyNLW sample.12 To date, no studies have examined whetherCAM use is associated with asthma medication adherenceusing objective electronic monitoring methods andwhether that association differs by ethnic group.

THE CURRENT STUDY

In a prior report including this sample, we found thatadherence to controller medications differed by ethnicgroup, with the lowest adherence among families of chil-dren from Island PR and the highest among NLW fam-ilies.10 In the current study, we evaluated CAM use in asample of families of children with asthma from NLW,PR, and Dominican Republic (DR) backgrounds fromRhode Island (RI) and Island PR, to evaluate patterns ofoverall CAM use by ethnic group. We also sought to deter-mine whether parental beliefs about conventional asthmamedications and barriers to obtaining medications wererelated to CAM use, and whether language preferenceand years living outside the United States (proxies foracculturation) were associated with CAM use among ourmainland Latino subgroups (PR and DR). Last, we evalu-ated whether CAM use was associated with objectiveadherence to asthma controller medications.We hypothesized that: 1) CAM use would be higher in

families of PR and DR descent, relative to NLW families,2) negative beliefs surrounding conventional asthma medi-cation and barriers to asthma medication use would berelated to greater CAM use for all families, 3) lower levelsof acculturation would be associated with reports of CAMuse for mainland Latino families, and 4) increased CAMuse would be related to lower levels of adherenceto controller medications, but only among Latinofamilies.12,23

METHODS

Participants were a subset from a larger study(n ¼ 805)24 evaluating mechanisms underlying pediatricasthma disparities between Latino and NLW groups. Previ-ous published work from this study has addressed contrib-uting factors to variations in asthma outcomes acrossLatino and NLW children, such as asthma severity,25

asthma symptom perception,26 and objective rates ofcontroller medication use.10

Data collection occurred in PR and RI. Participants(aged 7 to 16 years) were recruited from hospital-basedand community primary care clinics, community events,and asthma classes. Asthma diagnosis was confirmedthrough medical history, physical examination, andspirometry; those with complicating respiratory conditionswere excluded. PR families with a child with asthma wereeligible at the PR site. Ethnicity in RI was determined byparent report, and caregivers of PR, DR, or NLWethnicitywere eligible. For this report, we include only children withpersistent asthma (as a result of the greater amount ofmedication need) and those who completed an evaluationof CAM use.

PROCEDURE

Data collection occurred in 4 study visits over the courseof approximately 4 months. Visits were generally conduct-ed in the home by research assistants; the clinic visit wasconducted in a hospital setting by a study physician ornurse practitioner. Informed consent and demographic

194 MCQUAID ET AL ACADEMIC PEDIATRICS

information were obtained at enrollment. Data included inthis study were collected at the second visit, typically 1 to 2weeks after enrollment. During this visit, the family was in-terviewed regarding asthma management strategies using astandardized interview27 and completed surveys, and thefamily was oriented to adherence monitoring procedures.Adherence devices were collected approximately 5 weekslater. Families were compensated for participation. Theprotocol was approved by institutional review boards atboth sites.

MEASURES

ASTHMA SEVERITY

Asthma severity was evaluated by a study clinician dur-ing the clinic visit on the basis of the results of the physicalexamination, patient history, report of symptom frequency,prealbuterol forced expiratory volume in 1 second (FEV1)value, and current controller medication dose. The 4 levelsof asthma severity were classified as mild intermittent,mild persistent, moderate persistent, and severe persistent,consistent with guidelines in place at the time of thestudy.28

MEDICATION BELIEFS

Parents completed a 10-item self-report measure29 toassess parental beliefs about asthma medication, resultingin index scores of perceived medication necessity (eg, themedicines are necessary for the child’s health) and medica-tion concerns (eg, dependence, adverse effects).

BARRIERS TO FILLING PRESCRIPTIONS

Potential barriers to filling prescriptions were evaluatedusing an 8-item list devised for the larger study.20,23 Itemsrepresented health care system and contextual barriers suchas cost, transportation problems, language barriers, andpharmacy wait time.

ACCULTURATION

Language preferencewas coded as preferred language ofresearch interview for RI PR and DR families; this categor-ical variable and the number of years the caregiver hadlived outside of the mainland United States were used asproxies for acculturation.

CAM CHECKLIST

CAM use was assessed with the CAM Checklist, devel-oped for the larger study to assess CAM use among Latinoand NLWs,23 administered in the context of an establishedasthma management interview.27 Items were originallyderived through literature review,17 ongoing researchwith the target population,30 and a series of focus groupswith families from different ethnic subgroups. A list ofasthma treatment strategies reported by families was inde-pendently reviewed by experts in asthma management andcultural issues in RI and PR. Participants indicated whetherthey had utilized various types of CAM to treat their chil-dren’s asthma.

MEDICATION ADHERENCE MONITORING

A subset of participants who were prescribed asthmacontroller medications that could be monitored objectivelyparticipated in the medication adherence monitoring proto-col. Objective adherence monitoring in the larger study isdescribed in detail elsewhere.10 In brief, adherence wasmonitored for controller medications in 3 distinct ways.Adherence to oral montelukast was monitored using theTrackCapMEMS monitor (AARDEX), which is a replace-ment pill bottle with an electronic cap that registers the dateand time of each opening. Adherence to inhaled fluticasonewas monitored using the MDIlog II (Westmed), a smallelectronic device attached to an inhaler sleeve that fitsover the inhaler that records the date and time of inhaleractuation with inhalation. Adherence to fluticasone/salme-terol combination medication (diskus) was monitored byusing a standard protocol to record dosage counter readingsat the beginning and end of the monitoring period.Adherence was monitored for a minimum of 28 days.

Data were inspected and cleaned according to standardprocedures.31 Seventy-five cases were eliminated fromthe adherence analysis for a range of reasons (see McQuaidet al10 for specific details regarding device loss and datacleaning), resulting in 277 usable cases. Of these, 259 fam-ilies had also completed the CAM Checklist. Data used inadherence analyses were available from 77 island PR chil-dren, 51 RI PR children, 56 RI DR children, and 75 RINLW children.Adherence estimates were calculated as total doses per

day divided by prescribed doses per day. For participantsreceiving more than 1 monitored medication, the adherenceestimate represented the average of the 2 adherence mea-surements. Given that participants could be prescribedmore than 1 monitored medication and that medicationadherence is known to differ by medication type,32,33 acategorical variable was constructed to represent eachmedication combination encountered in this study, namelymontelukast alone (61.4%), fluticasone alone (7.7%),fluticasone/salmeterol combination (8.9%), fluticasoneand montelukast (8.5%), and fluticasone/salmeterol com-bination with montelukast (13.5%).

OVERVIEW OF ANALYSES

Chi-square analyses were conducted to determinewhether the prevalence of CAM use and CAM typesdiffered by site/ethnicity; t tests, chi square analyses, andlogistic regressions were conducted to examine whetheracculturation variables, caregiver-perceived barriers to pre-scription medication, and beliefs about asthma medicationwere predictive of CAM use (yes/no). The total number ofbarriers to prescription medication and the medicationnecessity and concerns subscales were entered simulta-neously into each regression model.Adherence estimates, which represent proportional data,

were adjusted using a probit transformation before analysis.Regression analyses controlling for covariates, specificallythe child’s asthma severity level, medication monitoringtype (eg, MDIlog, TrackCap, or dosage counter), and child

Table 1. Demographics of Children and Caregivers

Characteristic

Total Sample

(n ¼ 574)

No Adherence Data

Subsample (n ¼ 315)

Adherence Data

Subsample (n ¼ 259) P Value *

Child ethnicity/site, n (%) <.001Island PR 267 (46.5%) 190 (60.3%) 77 (29.7%)RI PR 92 (16%) 41 (13.0%) 51 (19.7%)RI DR 107 (18.6%) 51 (16.2%) 56 (21.6%)RI NLW 108 (18.8) 33 (10.5%) 75 (29.0%)

Child age, mean (SD) 10.54 (2.48) 10.73 (2.51) 10.32 (2.43) .053Asthma severity, n (%) <.001

Mild persistent 209 (36.4%) 132 (41.9%) 77 (29.7%)Moderate persistent 231 (40.2%) 143 (45.4%) 88 (34.0%)Severe persistent 134 (23.3%) 40 (12.7%) 94 (36.3%)

Child gender, female, n (%) 242 (42.2%) 133 (42.2%) 109 (42.1%) .974Caregiver identity, n (%) .051

Biological mother 545 (94.9%) 299 (94.9%) 246 (95.0%)Biological father 12 (2.1%) 3 (1.0%) 9 (3.5%)Other female relative 16 (2.7%) 12 (3.8%) 4 (1.6%)Other male relative 1 (.2%) 1 (.2%) 0 (0%)

Below poverty threshold, n (%) 303 (52.8%) 59.4% 116 (44.8%) <.001Combined caregiver years of education, mean (SD) 12.72 (2.69) 12.59 (2.67) 12.88 (2.71) .118Spanish language preference, n (%) .617

RI PR 52 (56.5) 20 (6.3%) 32 (12.4%)RI DR 92 (86.0) 45 (14.3%) 47 (18.1%)

Years caregiver lived outside United States, mean(SD)RI PR 13.58 (11.74) 12.31 (12.41) 14.63 (11.18) .352RI DR 21.51 (10.25) 21.78 (9.72) 21.26 (10.80) .795

PR ¼ Puerto Rico; RI ¼ Rhode Island; DR ¼ Dominican Republic; NLW, non-Latino white; SD ¼ standard deviation.

*Comparisons between adherence subsamples were made by c2 or t tests.

ACADEMIC PEDIATRICS COMPLEMENTARY AND ALTERNATIVE MEDICINE USE 195

age were conducted to determine whether reportedCAM use was associated with objective medicationadherence.31,32

Total Sample n = 805

Persistent Asthma Only n = 609

CAM Data n = 574

Medication Adherence Subsetn = 259

Figure 1. Participant flow.

RESULTS

PARTICIPANTS

From an initial sample of 805 families, the current sam-ple included children with persistent asthma who com-pleted the CAM checklist. This resulted in a sample of574 families across sites, including 267 island PR children,92 RI PR children, 107 RI DR children, and 108 NLW chil-dren in RI (Table 1). A subsample of 259 children also hadobjective adherence data available (Fig. 1).

PREVALENCE OF OVERALL CAM USE BY SITE/ETHNICITY

Reported CAM use differed by site and ethnicity [c2(3),n ¼ 574) ¼ 97.01, P < .001]. Island PR families reportedthe highest rates of CAM use (73%) compared to RI PR(51%), RI DR (62%), and NLW families (19%). Subse-quent analyses were conducted within groups.

PREVALENCE OF CAM TYPE ACROSS SITE/ETHNICITY

Reported types of CAM also differed by site andethnicity; specifically, mind–body interventions [c2(3),n ¼ 574, 81.70, P < .001], biologically based therapies[c2(3), n¼ 574, 91.63, P< .001], and body-based methods[c2(3), n ¼ 574, 101.30, P < .001] all differed by our site/ethnicity groups (Fig. 2). Island PR families reported thehighest rates of CAM use across the 3 types, followed by

RI DR, RI PR, and NLW families. Data regarding use byCAM item and type are presented in Table 2.

ASSOCIATION OF CAM USE TO ACCULTURATION

VARIABLES, PERCEIVED BARRIERS, AND MEDICATION

BELIEFS

ACCULTURATION VARIABLES

The number of years caregivers reported living outsidethe United States was unrelated to CAM use across groups.

Figure 2. Percentages of CAM type by site/ethnicity. CAM indicates complementary and alternative medicine; PR Puerto Rico; DR,

Dominican Republic; and NLW, non-Latino white. *P < .001 by c2 square analysis comparing CAM type by site/ethnicity.

196 MCQUAID ET AL ACADEMIC PEDIATRICS

Language preference was only related to CAM use in RIDR families [c2(1), n ¼ 125, 5.34, P ¼ .021]. CAM usewas higher among Spanish-speaking (57.6%) thanEnglish-speaking (5.6%) RI DR families.

Table 2. Percentage of Endorsement of CAM Items

CAM Type Island PR RI PR RI DR NLW c2

Biological basedWhale oil 3 0 3.7 0 6.76Cod liver oil 24.3 8.7 24.3 0 40.37*Honey or royal jelly 49.4 21.7 39.3 4.6 76.52*Onion or garlic 15.7 6.5 32.7 0.9 49.34*Almond oil 6.7 0 2.8 0 15.02*Castor oil 7.5 2.2 7.5 0 11.54*Oregano 2.2 1.1 4.7 0 6.28Lemon 30.3 15.2 27.1 3.7 35.60*Aloe vera juice 34.5 6.5 16.8 0 72.42*Siete jarabes† 28.8 13 8.4 0 55.07*Agua maravilla‡ 16.9 0 0.9 0 53.00*Chamomile tea 8.2 5.4 8.4 3.7 3.12Eucalyptus tea 8.2 2.2 0.9 0.9 16.12*Jarabe de maguey§ 12.4 0 1.9 0 34.61*Turtle blood or meat 0.4 0 1.9 0 4.90Cat or snake meat 0.4 0 1.9 0 4.90Lizard in boiled milk 1.1 0 0.9 0 2.19Many oils mixed together 8.6 4.3 19.6 0 29.53*

Mind–bodyPraying 51.7 21.7 30.8 8.3 74.51*Azabachek 1.1 0 1.9 0 3.19Prayer candles 3.4 0 3.7 0 7.16Azogue{ 1.1 0 0 0 3.47

Body basedVicks VapoRub 62.9 44.6 53.3 14.8 72.71*Massage 46.4 13 21.5 3.7 88.32*

CAM ¼ complementary and alternative medicine; PR ¼ Puerto

Rico; RI ¼ Rhode Island; DR ¼ Dominican Republic; NLW ¼ non-

Latino white.

*Statistically significant (p < .05).

†Honey syrup that may contain a mixture of almond oil, castor, oil,

wild cherry, and licorice.

‡Witch hazel compound that may contain aloe vera juice, honey,

garlic, and onion.

§Agave nectar.

kStone used to ward off evil spirits.

{Mercury used to ward off evil spirits.

PERCEIVED BARRIERS TO PRESCRIPTION MEDICATION

A higher number of perceived barriers to prescriptionmedication use was associated with CAM use among RIPR families (b ¼ 0.76, standard error [SE] 0.39,P ¼ .050; Table 3) but not among other groups.

BELIEFS ABOUT MEDICATION

Elevated medication concerns were associated withCAM use among Island PR (b ¼ 0.50, SE 0.22,P ¼ .026) and RI NLW (b ¼ 0.85, SE 0.39, P ¼ .030;Table 3) families, but not among RI PR and RI DR families.Medication necessity was not related to CAM use withinany group.

RELATION OF CAM TO ADHERENCE

CAM use was related to increased adherence tocontroller medications within NLW families (b ¼ 0.23,P ¼ .047; Table 4). For children and families from Latinobackgrounds, CAM use was not related to controller medi-cation adherence.

DISCUSSION

Pediatric asthma continues to represent a source of healthdisparities between Latinos and NLWs.1,3 Given thatadherence to controller medications remains challengingfor many families,31,34 and for minority families inparticular,10,32,33 cognitive and behavioral variables suchas attitudes toward medication use, barriers to obtainingmedications, and preference for CAM use representpotentially modifiable intervention targets. In our sample,nearly three-quarters of Island PR and over half of RI Latinofamilies reported some CAM use, whereas only one-fifth ofNLW families reported using CAM. Results are consistentwith previous research demonstrating a high prevalence ofCAM use in Latino families.17,30

Self-report of CAM use was associated with languagepreference among our DR group, suggesting that lower-acculturated families were more likely to use CAM.CAMusewas modestly associated with beliefs and barriersto medication use, and only for certain groups. Among our

Table 3. Association of CAM Use to Perceived Barriers and Beliefs About Medication

Characteristic

Island PR RI PR RI DR RI NLW

b SE Wald c2 OR b SE Wald c2 OR b SE Wald c2 OR b SE Wald c2 OR

Medication barriers 0.00 0.13 0.00 1.00 0.76 0.39 3.86* 2.14 0.31 0.31 0.98 1.36 0.34 0.45 0.59 1.41Medication necessity �0.24 0.19 1.62 .20 .17 0.32 .27 1.18 .39 .34 1.28 1.47 .04 0.41 .01 1.04Medication concerns .50 0.22 4.93* 1.64 .16 .31 .27 1.17 .02 0.29 0.01 1.03 .85 .39 4.71* 2.34

CAM ¼ complementary and alternative medicine; PR ¼ Puerto Rico; RI ¼ Rhode Island; DR ¼ Dominican Republic; NLW ¼ non-Latino

white.

*Statistically significant. Tests are logistic regressions. (p < .05).

ACADEMIC PEDIATRICS COMPLEMENTARY AND ALTERNATIVE MEDICINE USE 197

RI PR group, barriers to obtaining medications weremodestly associated with CAM use. It is possible that thesefamilies may turn to CAM when they encounter barriers toobtaining asthma medication. Because of financial strain,barriers to obtaining medications may also serve as barriersto obtaining CAM (eg, rubs available at the pharmacy,teas). Elevated medication concern was positively associ-ated with CAM use only in the Island PR and NLW groups.Parental concerns could be a factor in the high prevalencerate of CAM use within the Island PR group, and similarconcerns may play a role among NLW families. Specificfactors that may predict CAM use not measured in thisreport bear further study, including ease of access to cultur-ally sanctioned CAMmethods, such as through a bodega orherbalist in one’s neighborhood. Amore nuanced examina-tion of acculturation and immigration factors (eg, culturalidentity), health care system factors (eg, consistent asthmacare provider), and CAM use in relation to conventionalmedication use is warranted.

We previously noted the lowest rates of adherence tocontroller medications among Island PR families and thehighest rates among RI NLW families.10 Our results sug-gest that factors other than CAM use may account forthis effect. CAM use was unrelated to medication adher-ence among our Latino families in both RI and PR. Thesefindings are consistent with previous research,12 indicatingno relation between child medication adherence and CAMuse in a predominantly NLW sample.

CAM use was positively related to medication adher-ence among NLW families in our sample. Some familieswho are concerned about their child’s asthma may try mul-tiple strategies (both conventional and CAM) to managesymptoms. For Latino families in RI and PR, this relation-ship is less clear. CAM use may be related to medicationuse at multiple levels and may change over time as a result

Table 4. CAM Use and Objective Adherence

Step Variable

Island PR R

b t R2 b

1 Child asthma severity �0.03 �0.24 0.03 �0.04Child age �0.16 �1.31 �0.15Medication monitoring type 0.01 0.05 0.11

2 CAM 0.08 0.63 0.03 0.20

CAM ¼ complementary and alternative medicine; PR ¼ Puerto Rico;

white.

*P < .05.

†P < .10.

of family circumstances. For example, families may hesi-tate to fill a prescription once received, or they may useit intermittently according to their financial circumstances.If a medication has run out and they do not have the re-sources for a refill, praying or using oils may seem a viablealternative if useful in the past. Overall, these results reflectthe notion that the relationship between CAM use and useof conventional medications is complex, and CAM usemay not indicate poor medication adherence. Research isneeded to clarify dynamic multilevel factors that may pre-dict CAM use and use of conventional asthma medicationsamong specific ethnic groups.A number of study limitations bear mention. Our check-

list was developed to assess CAM use specific to Latinos,and it may havemissed certain strategies more typically im-plemented by NLW families (eg, chiropractic care).35 Useof objectivemeasures ofmedication adherence is a strength;however, we were only able to assess adherence to medica-tions that had been prescribed and filled; hence, it did notcapture under prescription of controller medications or non-adherence through failure to fill prescriptions. Qualitativework may be necessary to fully understand how familiesmake decisions about CAM use and how this affects pat-terns of filling and using conventional medications.Although study personnel were fully bilingual and

trained in cultural sensitivity surrounding asthma practices,it is possible that RI Latino families underreported CAMuse as a result of perceptions that these strategies mightbe less acceptable in the dominant US culture. Addition-ally, administration of measures regarding asthma manage-ment before adherence data collection may have primedfamilies to be more adherent; however, it is not clear thatthis would lead to group differences. Group sample sizeswere modest, which may have limited our power to find ef-fects. Last, our sample was recruited on the basis of likely

I PR RI DR RI NLW

t R2 b t R2 b t R2

�0.24 0.03 0.17 1.29 0.15 0.04 0.35 0.09�0.99 �0.37 �2.74* �0.30 �2.61*0.65 0.00 0.00 1.00 0.00 �0.031.42 0.07 �0.23 �1.77† 0.20 0.23 2.03* 0.14

RI ¼ Rhode Island; DR ¼ Dominican Republic; NLW ¼ non-Latino

198 MCQUAID ET AL ACADEMIC PEDIATRICS

asthma diagnosis and through convenience sampling, andhence it may not adequately represent families with chil-dren underdiagnosed or undertreated for asthma.

There is an increasing emphasis on culturally centeredcare in the treatment of children with asthma.36 Incorpo-rating the cultural beliefs and practices of families intoasthma-related care can improve asthma outcomescompared to standard treatment approaches.37 Followingthe Awareness–Assessment–Negotiation model38 may aidclinicians in integrating these findings into patient care.According to this model, clinicians should become awareof commonly held beliefs and practices among culturesfor which they provide care, assess these beliefs and prac-tices, and find ways to integrate beliefs and practices intopatient care. Clinicians should be encouraged to discussand assess CAM use and how it may be related to concernsregarding use of conventional asthma medications andfamily specific barriers to medication use.17 Culturallycentered care may enhance the patient–provider relation-ship and ultimately improve asthma management amongfamilies of children at most risk for asthma morbidity.

ACKNOWLEDGMENTS

Supported in part by the National Heart, Lung, and Blood Institute (U01

HL 072438, G. Fritz and G. Canino, PIs) the and Eunice Kennedy Shriver

Institute of Children’s Health and Human Development (K24 HD 058794,

E. L. McQuaid, PI). Funded by the National Institutes of Health.

REFERENCES

1. Akinbami LJ, Moorman JE, Liu X. Asthma prevalence, health care

use, and mortality: United States, 2005–2009. Natl Health Stat Rep.

2011;1–14.

2. Hunninghake GM, Weiss ST, Celed�on JC. State of the art: asthma in

Hispanics. Am J Respir Crit Care Med. 2006;173:143–163.

3. Lara M, Akinbami L, Flores G, Morgenstern H. Heterogeneity of

childhood asthma among Hispanic children: Puerto Rican children

bear a disproportionate burden. Pediatrics. 2006;117:43–53.

4. Canino G, McQuaid EL, Rand CS. Addressing asthma health dispar-

ities: a multilevel challenge. J Allergy Clin Immunol. 2009;123:

1209–1217.

5. Canino G, Koinis-Mitchell D, Ortega, AN, et al. Asthma disparities in

the prevalence, morbidity, and treatment of Latino children. Soc Sci

Med. 63:2926–37.

6. National Asthma Education and Prevention Program. Expert Panel

Report 3: Guidelines for the Diagnosis and Management of Asthma.

Bethesda, MD: US Dept of Health and Human Services, National

Institutes of Health; 2007.

7. Conn KM, Halterman JS, Fisher SG, et al. Parental beliefs about

medications and medication adherence among urban children with

asthma. Ambul Pediatr. 2005;5:306–310.

8. Bearison DJ, Minian N, Granowetter L. Medical management of

asthma and folk medicine in a Hispanic community. J Pediatr

Psychol. 2002;27:385–392.

9. Apter AJ, Reisine ST, Affleck G, et al. Adherence with twice-daily

dosing of inhaled steroids. Socioeconomic and health-belief differ-

ences. Am J Respir Crit Care Med. 1998;157(6 pt 1):1810–1817.

10. McQuaid EL, Everhart RS, Seifer R, et al. Medication adherence

among Latino and non-Latino white children with asthma. Pediatrics.

2012;129:e1404–e1410.

11. Adams S, Koinis Mitchell D. Perspectives on complementary and

alternative therapies in asthma. Expert Rev Clin Immunol. 2008;4:

703–711.

12. Philp JC, Maselli J, Pachter LM, Cabana MD. Complementary and

alternative medicine use and adherence with pediatric asthma treat-

ment. Pediatrics. 2012;129:e1148.

13. National Center for Complementary and Alternative Medicine

(NCCAM). What is CAM? March 8, 2007. Available at: http://

nccam.nih.gov/health/whatiscam/. Accessed November 1, 2012.

14. Slader CA, Reddel HK, Jenkins CR, et al. Complementary and alter-

native medicine use in asthma: who is using what? Respirology. 2006;

11:373–387.

15. Braganza S, Ozuah PO, Sharif I. The use of complementary therapies

in inner-city asthmatic children. J Asthma. 2003;40:823–827.

16. Pachter LM, Weller SC. Acculturation and compliance with medical

therapy. J Dev Behav Pediatr. 1993;14:163–168.

17. Pachter LM, Cloutier MM, Bernstein BA. Ethnomedical (folk) rem-

edies for childhood asthma in a mainland Puerto Rican community.

Arch Pediatr Adolesc Med. 1995;149:982–988.

18. Conn KM, Halterman JS, Lynch K, Cabana MD. The impact of par-

ents’ medication beliefs on asthma management. Pediatrics. 2007;

120:e521–e526.

19. Koster ES, Wijga AH, Koppelman GH, et al. Uncontrolled asthma at

age 8: the importance of parental perception toward medication.

Pediatr Allergy Immunol. 2011;22:461–468.

20. McQuaid EL, Vasquez J, Canino G, et al. Beliefs and barriers to medi-

cation use in parents of Latino childrenwith asthma.Pediatr Pulmonol.

2009;44:892–898.

21. Flores G, Abreu M, Olivar MA, Kastner B. Access barriers to health

care for Latino children. Arch Pediatr Adolesc Med. 1998;152:

1119–1125.

22. Shen J, Oraka E. Complementary and alternative medicine (CAM)

use among children with current asthma. Prev Med. 2012;54:27–31.

23. Adams SK, Murdock KK,McQuaid EL. Complementary and alterna-

tive medication (CAM) use and asthma outcomes in children: an

urban perspective. J Asthma. 2007;44:775–782.

24. Canino G, McQuaid EL, Alvarez M, et al. Issues and methods in

disparities research: the Rhode Island–Puerto Rico asthma center.

Pediatr Pulmonol. 2009;44:899–908.

25. Esteban CA, Klein RB, McQuaid EL, et al. Conundrums in childhood

asthma severity, control, and health care use: Puerto Rico versus

Rhode Island. J Allergy Clin Immunol. 2009;124:238–244.e5.

26. Fritz GK, McQuaid EL, Kopel SJ, et al. Ethnic differences in percep-

tion of lung function: a factor in pediatric asthma disparities? Am J

Respir Crit Care Med. 2010;182:12–18.

27. McQuaid EL, Walders N, Kopel SJ, et al. Pediatric asthma manage-

ment in the family context: the family asthma management system

scale. J Pediatr Psychol. 2005;30:492–502.

28. Global Initiative for Asthma. 2006 Revision: GINA Report, Global

Strategy for Asthma Management and Prevention. Vancouver, Wash:

Global Initiative for Asthma. Available at: http://www.ginasthma.

org/documents/5/documents_variants/31; 2006. Accessed October

20, 2011.

29. Horne R, Weinman J. Patients’ beliefs about prescribed medicines

and their role in adherence to treatment in chronic physical illness.

J Psychosom Res. 1999;47:555–567.

30. Koinis-Mitchell D,McQuaid EL, FriedmanD, et al. Latino caregivers’

beliefs about asthma: causes, symptoms, and practices. J Asthma.

2008;45:205–210.

31. McQuaid EL, Kopel SJ, Klein RB, Fritz GK.Medication adherence in

pediatric asthma: reasoning, responsibility, and behavior. J Pediatr

Psychol. 2003;28:323–333.

32. Celano MP, Linzer JF, Demi A, et al. Treatment adherence among

low-income, African American children with persistent asthma.

J Asthma. 2010;47:317–322.

33. McNally KA, Rohan J, Schluchter M, et al. Adherence to combined

montelukast and fluticasone treatment in economically disadvantaged

African American youth with asthma. J Asthma. 2009;46:921–927.

34. Lasmar L, Camargos P, Champs NS, et al. Adherence rate to inhaled

corticosteroids and their impact on asthma control. Allergy. 2009;64:

784–789.

ACADEMIC PEDIATRICS COMPLEMENTARY AND ALTERNATIVE MEDICINE USE 199

35. Barnes PM, Powell-Griner E, McFann K, Nahin RL. Complementary

and Alternative Medicine Use Among Adults: United States, 2002. Be-

thesda,Md:USDept ofHealth andHumanServices, Centers forDisease

Control. Available at: http://nccam.nih.gov/sites/nccam.nih.gov/files/

news/camstats/2002/report.pdf; 2004. Accessed OCtober 15, 2012.

36. Partridge MR. In what way may race, ethnicity or culture influence

asthma outcomes? Thorax. 2000;55:175–176.

37. Bailey EJ, Cates CJ, Kruske SG, et al. Culture-specific programs for

children and adults fromminority groups who have asthma.Cochrane

Database Syst Rev. 2009;CD006580.

38. Pachter L. Working With Patients’ Health Beliefs and Behaviors:

The Awareness–Assessment–Negotiation Model in Clinical Care.

Columbus, Ohio: Ross Products Division, Abbott Laboratories;

2000.