The primary prevention of asthma in children study: Design of a multifaceted prevention program

The primary prevention of asthma in childrenstudy: Design of a multifaceted preventionprogram

Childhood asthma, a chronic disease of theairways, is predominantly newly diagnosed ininfants and small children. Asthma is the maincause of school absence (1, 2), involves high costsand is known to reduce the quality of life ofchildren and their parents. During the lastdecades there has been a substantial increase inthe prevalence of asthma worldwide (3, 4), whichemphasizes the need to prevent its development.It is generally thought that genetic as well asenvironmental factors are involved in the devel-opment of asthma (5, 6). Because it is not yet

possible to influence the genetic factors, preven-tion programs should focus on influencing theenvironmental factors. There are reasons tobelieve that the first signs of sensitization occuralready in the prenatal stage by the interactionbetween the maternal environment and the foetus(7). Therefore, it is important to start interven-tion in the prenatal stage. Several primaryprevention studies are being performed, and arestill in progress. In most primary preventionstudies only one (8–10) intervention was testedwhereas in others two single interventions with or

Kuiper S, Maas T, van Schayck CP, Muris JWM, Schonberger HJAM,Dompeling E, Gijsbers B, van Weel C, Andre Knottnerus J on behalf ofthe PREVASC group. The primary prevention of asthma in childrenstudy: Design of a multifaceted prevention program.PediatrAllergyImmunol2005:16:321–331.�2005BlackwellMunksgaard

The PREVASC study addresses the primary prevention of asthma ininfants and small children. The objective of this study is to investigatewhether a multifaceted prenatally started intervention strategy in high-risk infants leads to a decrease in the occurrence of (severe) asthma andwhether a refinement of the prevention strategy leads to an increase inthe adherence to the prevention program. The primary preventionprogram includes house dust mite impermeable bed coverings, educa-tion on breast feeding, hypoallergenic feeding, timing of introduction ofsolid food and smoking cessation. A total of 888 infants were prenatallyincluded. By the time of inclusion the mothers were 3–7 months preg-nant. About 27 infants were excluded from the study and 18 droppedout. Of the remaining 843 infants 535 had a first-degree familial pre-disposition of asthma (high-risk group), whereas a reference group of308 (162 boys) infants was not predisposed for asthma in the first-degree(low-risk group). To evaluate the (cost-)effectiveness of the preventiveintervention, 222 (118 boys) infants of the high-risk group allocated tothe intervention group and 221 (112 boys) allocated to a control groupare followed up. The low-risk infants served as controls to evaluate thepredictive value of high risk (first-degree familial predisposition ofasthma). The infants are followed from the prenatal stage until theyreach the age of 6 yr. The remaining 92 high-risk infants were includedin an optimized randomized-clinical adherence trial (RCAT). Of these92 infants, 45 (20 boys) were allocated to an intervention group and 47(24 boys) to a control group. Until now all infants have been followedfor at least 1 yr.

Sandra Kuiper1, Tanja Maas1,Constant P. van Schayck1,Jean W. M. Muris1,Huub J. A. M. Schçnberger1,Edward Dompeling2,Barbara Gijsbers1, Chris van Weel3

and J. Andr� Knottnerus1 on behalf ofthe PREVASC group1Department of General Practice, Care and PublicHealth Research Institute, University of Maastricht,Maastricht, 2Department of Pediatrics, Care andPublic Health Research Institute, University HospitalMaastricht, Maastricht, 3Department of GeneralPractice, University Medical Centre Nijmegen,Nijmegen, The Netherlands

Key words: asthma; allergen avoidance; infants;house dust mite; multifaceted primary prevention;randomized-controlled trial

Sandra Kuiper, Department of General Practice, Careand Public Health Research Institute, University ofMaastricht, PO Box 616, 6200 MD Maastricht,The NetherlandsTel.: +31 43 388 21 84Fax: +31 43 361 93 44E-mail: [email protected]

Accepted 21 February 2005

Pediatr Allergy Immunol 2005: 16: 321–331 Copyright � 2005 Blackwell Munksgaard

PEDIATRIC ALLERGY AND

IMMUNOLOGYDOI: 10.1111/j.1399-3038.2005.00278.x

321

without a factorial design were tested (11–13),which may lead to underestimation of the effectof prevention. To our knowledge there are threemultifaceted primary prevention studies (14–16),in which several interventions are combined intoone program and compared with one othermodality, like usual care.We report on the objectives and the design of

the prevention of asthma in children (PRE-VASC) a research program focussed on primaryprevention of asthma in children. The objectivesof this program are to estimate the predictivevalue of a positive family history of asthmaassessed in the prenatal period by the generalpractitioner (GP) for the development of child-hood asthma and, to assess the effectiveness of aprenatally started multifaceted primary preven-tion program focused on reducing the exposureto indoor and food allergens and environmentaltobacco smoke (ETS) in children at high risk ofdeveloping asthma. Furthermore, it will beassessed whether the duration of the preventionprogram has any effect on the development ofasthma as measured at 6 yr of age and whetheroptimizing the intervention leads to increasedadherence. Finally, the health-economic benefitsof high-risk identification and primary preven-tion of asthma will be assessed.

Research planGeneral design

Our study comprises three parts, one observa-tional cohort and two intervention studies.To assess the clinical effectiveness of the

PREVASC prevention program, a randomized-clinical prevention trial (RCPT) was started in agroup of prenatally selected high-risk infants.Additionally a cohort of prenatally selected low-risk infants was composed of a natural historystudy, to assess the clinical course of asthma andallergies in high- and low-risk children and toassess the predictive value of �high-risk�. To beable to identify whether (the parents of) theinfants were compliant to the prevention pro-gram, the adherence to the intervention wasevaluated. In addition, to get a better insight intothe determinants of adherence and to improvethe preventive program, a �focus� group approachwas applied in a subgroup of trial participants. Inaddition, an �optimized randomized-clinicaladherence trial� (RCAT) was evaluated in anewly recruited group of prenatally selectedinfants with a family history of asthma. Cost-effectiveness of the prevention program will bestudied on the basis of the RCPT, taking

adherence aspects and validity of screening intoaccount.In the participants section a general descrip-

tion will be given of how the participants wereincluded in the study. In the section concerningthe intervention studies and the natural historystudy the outline of both intervention studies andthe natural history study will be described.

Participants

The infants were recruited prenatally when theirmothers were in their first two trimesters ofpregnancy. The community-based recruitmentstarted in January 1997 and continued untilFebruary 2002. For the RCPT this was per-formed by the primary caregivers (GP andmidwives) and by advertisements. For the nat-ural history study recruitment was exclusivelyperformed by the GP, whereas for the optimizedRCAT recruitment was performed by midwivesand by advertisements. These complementaryways of recruitment were necessary to include888 participants.�High-risk� was defined as �at least one first-

degree family member (i.e. the pregnant woman,the biological father of the unborn child or asibling) suffering from GP registered asthma�,whereas �low-risk� was defined as �absence of GPregistered asthma in first-degree family mem-bers�. Asthma was defined according to theInternational Classification of Primary Care(ICPC) (17) as recurrent episodes of reversibleacute bronchial obstruction with wheeze and/ordry cough. This was checked by the GP andscreening questionnaires.For the RCPT, the natural history study and

the RCAT, a family was excluded from the studywhen there was intrauterine or neonatal death,major language problems, serious birth defects(illness or malformation) and moving abroad.

Intervention studies

In the following specific elements of the variousstudy parts are described. Thereafter aspects thatrelate to all the three study parts are presentedtogether.

Randomized-clinical prevention trialThe goal of this study was to assess the effect-iveness of (the duration of) a prenatally startedmultifaceted primary prevention program inchildren at high risk for development of asthma.The study was designed to follow the high-riskchildren until they reach the age of 6. By thattime lung function measurement can reliably be

Kuiper et al.

322

performed which makes an objective diagnosis ofasthma possible.

RecruitmentIn the first part of the study (the period from theprenatal stage until the age of 2) 476 pregnantwomen were included (Fig. 1) and after rand-omization equally divided over two study groups:an intervention group and a control group.About 443 of the descendants could be followedup, 222 (118 boys) of the intervention group and221 (112 boys) of the control group.

Intervention first partThe intervention started prenatally in order toreach a low house dust mite (HDM) and petallergen exposure level at the time the child wasborn. In addition, the intervention focussed onavoidance of prenatal and postnatal passivesmoking and on avoidance of food allergenspostnatal.

House dust mite reduction intervention star-ted before the month 7 of pregnancy. Thisintervention consisted of advice on ventilationand cleaning as well as application of HDMimpermeable covers on the parents and chil-dren’s bed. To prevent exposure to pet allergens,the advice was given to keep pets outdoors fromthe month 6 of pregnancy. Smoking interven-tion consisted of advice on smoking cessation ofthe mother as early as possible in pregnancy andno smoking of the father as well as the motherin the presence of the baby postnatal. Dietaryintervention contained advice on exclusivelybreast feeding (or hypoallergen formula feeding)for at least 6 months and postponing introduc-tion of solid food until 6 months after birth.The high-risk control group received usual care,according to the Guidelines of the DutchCollege of GPs that include criteria-based diag-nosis and an initial treatment with short actingbronchodilators and when no adequate asthma

RCPT-Cn= 221

RCPT-IVn= 222

RCPT-LIV

RCPT-Cn=220

Exclusionn = 14

No follow-upn = 1No follow-up

n = 3

No follow-upn = 3

RCPT-SIV

Inclusionn = 888

RCPT-Cn = 221

RCPT-IVn = 222

RCPT-LIVn = 111

RCPT-Cn = 220

2 2 yr

6 6 yr

1 1 yr

No follow-upn = 6

Exclusionn = 10

Exclusionn = 0

Exclusionn = 0

Exclusionn = 3

RCPT-SIVn = 108

No follow-upn = 3

No follow-upn = 0

No follow-upn = 6

RCPT-IVn = 228

RCPT-IVn = 242

RCAT-IVn = 48

RCAT-IVn = 45

RCPTn = 476

RCATn = 95

NH-Cn = 317

NH-Cn = 308

RCPT-Cn = 224

RCPT-Cn = 234

RCAT-Cn = 47

RCAT-Cn = 47

Randomisation

Prenatalstage

Fig. 1. Schematic representation of the PREVASC studies. RCPT-IV, randomized-clinical prevention trial interventiongroup; RCPT-C, randomized-clinical prevention trial control group; RCAT-IV, randomized-clinical adherence trial inter-vention group; RCAT-C, randomized-clinical adherence trial control group; NH-C, natural history control group; RCPT-SIV, RCPT short intervention; RCPT-LIV, RCPT long intervention.

Primary prevention of asthma in children

323

control is achieved with a step-up approachwith inhaled corticosteroids in increasing dosage(18).

Intervention second partIn the second part of the study the children arefollowed from 2 to 6 yr of age. For this part ofthe study – which is still going on – the childrenin the original intervention group were random-ized again into two groups (Fig. 1). One groupwas no longer exposed to advice and HDMimpermeable mattress covers (short interventionperiod). The other group still received an inter-vention program focused on diminishing allergenand irritant exposure and to keep on using HDMimpermeable mattress covers (extended inter-vention period). The children that participated inthe control group in the first part continue toparticipate in the control group in the secondpart of the study as well and receive usual care.

Randomized-clinical adherence trialEighty-one randomly selected families who par-ticipated in the intervention group of the RCPTreceived an invitation letter for a focus-groupinterview to investigate which determinantsinfluenced the actual adherence behaviour. Forthese interviews the stepwise guidelines of VanAssema et al. (19) for conducting a focus-groupinterview were used.About 43 households (59 participants) were

scheduled to meet at times and locations thatmight be convenient for the participants. Sixgroups were mixed as to gender and one groupwas formed containing only fathers, becausefathers may have different opinions and thoughtsabout breast feeding and postponement of solidfoods, which might be missed in the presence ofwomen.The reasons for the 39 non-respondent house-

holds were: interested, but date not convenient(n ¼ 14), no time (n ¼ 7), location of interviewnot convenient (n ¼ 3), on holiday (n ¼ 2),unknown (n ¼ 13). No significant differenceswere found between the participation and non-participation group with respect to level ofeducation, feeding behaviour and family historyof asthma.On the basis of the results of the determinant-

analysis, a refined prevention educationalprogram was developed.

Recruitment refined intervention programIn this phase, the RCAT study, 92 familiesexpecting a child at high risk of developing asthmawere selected. About 45 families received inter-vention and 47 served as controls (Fig. 1).

Intervention measuresEducational materials and the educational mes-sage to be provided orally by research nurseswere refined. The effect of the renewed educationon the degree of adherence with the advisedmeasures was evaluated in the next phase of thisprogram.

Natural history studyThe goal of this study was to assess the clinicalcourse of asthma and allergies in high- and low-risk children and to assess the predictive value of�high-risk�.

RecruitmentFor this study 317 pregnant women, withoutasthma or asthma in the family, were included.About 308 (162 boys) of the descendants couldbe followed up (Fig. 1).

Clinical follow upThe clinical course of symptoms of asthma andallergies over the first 2 yr of life in this cohortwas compared with the RCPT control group inorder to evaluate the predictive value of thefamily history. The low-risk infants receivedusual care.

Randomization, withdrawal procedures and blinding

For those women who were recruited by the GP,randomization was done in zipcode clusters onpractice level and not on patient level in order tominimize the probability of exchange of infor-mation between participants.Families who were recruited by the GP were

screened by the GP. Families who were recruitedfor the RCPT study by midwives or advertise-ments completed questionnaires concerning thepresence of asthma in their family. Whenever itturned out from the questionnaire that there wasno asthma in a �high-risk� family, the GP wascontacted. When the GP confirmed the question-naire outcome, the family was withdrawn fromthe study. Families who were recruited by mid-wives or advertisements for the RCAT studywere screened by the GP.For the RCPT and the RCAT, blinding for

group assignment was performed at family level.

Economical evaluation

To assess the efficiency of �high-risk� identifica-tion and the cost-effectiveness of the PREVASCintervention, it will be evaluated whether costsavings [associated with a decrease in (severe)asthma cases and an increase in the quality of

Kuiper et al.

324

life] outweigh the costs of the prevention pro-gram. This will be achieved in two steps, byassessing (i) the cost-effectiveness of the PRE-VASC intervention among the high-risk groupand (ii) the cost-effectiveness of the completePREVASC RCPT program including the high-risk identification at the population level.

Ad (1)It will be assessed what the PREVASC prevent-ive intervention with its allied costs yields interms of health outcome and savings concerningdirect and indirect health-related costs in thehigh-risk group. For this purpose, the clinicalcourse, health outcomes and costs of primaryprevention of asthma (PREVASC interventiongroup) will be compared with the clinical course,health outcomes and costs of usual care (PRE-VASC control group). To be able to assess thecost-effectiveness, information concerning thedirect and indirect medical costs, the costs of theprevention program, and the health status of thechildren will be collected, using information fromquestionnaires that will be completed by theparents. By use of these questionnaires the use ofmedical care and products can be measured. Thisuse will be valued against national recommendedprices (20) to be able to determine the costs ofcare per individual. The health status of thechildren will be measured using the InternationalStudy of Asthma and Allergies in Childhood(ISAAC) questionnaire as well as the morbidityregistration of the GP. When the children havereached the age of 6 yr asthma will objectively bediagnosed, which makes it possible to expressasthma as a dichotomous variable. In this waythe costs per asthma case prevented can bedetermined.

Ad (2)By connecting the results Ad (1) (the cost-effectiveness of the primary prevention programamong the high-risk group) to the predictivevalue of the �high-risk� identification, the costs ofprenatal high-risk identification, the prevalenceof a positive family history, and the observeddisease course in the absence of intervention inboth the high-risk control group and the low-riskgroup, the expected cost-effectiveness of thecomplete PREVASC program at population le-vel will be assessed.

Data collection procedure

Symptoms/diagnosisFirst-degree familial predisposition of asthmawas determined by the GP and by use of a

questionnaire directly at the inclusion. For theRCPT and the �natural history� study this was re-evaluated at the age of 2 yr. Information oncomplaints, diagnosis, laboratory tests andmedication concerning respiratory morbidity areprospectively registered by the GP at every con-sultation. Additionally the parents were asked tocomplete weekly reports, concerning questionsabout asthma-related complaints, over-the-counter (OTC) medication and asthma-relatedcosts, during the first 2 yr of the infants� life andto complete the internationally accepted ISAACquestionnaire (21) yearly until the age of 6(Tables 1–3).

Mite collection and humidityDust samples were collected (Table 1) to meas-ure the compliance with the intervention advicein the intervention group (RCPT and RCAT)and to correlate the intensity of exposure toinhalant allergens with respiratory morbidity(RCPT, RCAT and natural history study) andto evaluate the validity of using the low-riskinfants as control group. The collection wasperformed in a standardized manner. Sampleswere taken, using a vacuum cleaner (BoschBSA1100 1300 W; Bosch BV, Brussels, Belgium)with a special cassette (ALK Abello, Nieu-wegein, The Netherlands) containing a What-man GF/F 70 mm filter; Whatman, Breda, TheNetherlands, from the living room floor(2 · 1 m2), the parents� mattress (1 m2) and themattress of the baby (whole mattress). Thesamples were kept on 4�C until they were ana-lysed for mite, cat and dog allergens by use ofenzyme-linked immunosorbent assay (ELISA)(22–24). Humidity was measured by use of ahygrometer in the parents� bedroom and, whenin use, the infants� bedroom.

Type of feeding and ETS exposureThe parents completed questionnaires concern-ing the type of feeding, environmental allergenexposure and ETS exposure.

IgEBlood samples were collected in the RCPT andnatural history group to measure the concentra-tion of total immunoglobulin E (IgE). IgE wasanalysed using the method of Aalberse et al. (25).

Carbon monoxideTo study the compliance with the smoking advicein the intervention group, to detect spontaneoussmoking cessation by the mother during preg-nancy in the control groups and to correlate theintensity of exposure to smoke with respiratory


325

Table 1. Timing schedule of the measurements of environmental exposures and outcome variables in the RCPT

Time-point Environmental exposures Outcome measures

3–5 months of pregnancy Dust (Der p 1, Can f 1, Fel d 1)HumidityCO

7–8 months of pregnancy (intervention group only) HumidityCO

Weekly from birth until 2 yr of age Type of feedingDay care attendance

Asthma-related complaintsHealth-care consumptionAbsent from work

Around 4 wk after birth (intervention group only) HumidityCO

3–5 days after birth Total IgE6 months after birth Type of feeding Amount of wheezing episodes in the last 6 months

Nightly wheezing in the last 6 monthsNocturnal coughing in the last 6 monthsDoctors diagnosed asthmaConstitutional eczemaFood allergyNon-viral rhinitisOther upper airway infectionsOther lower airway infectionsDyspnoea

7–9 months after birth (exactly 1 yr after first dust sampling) Dust (Der p 1, Can f 1, Fel d 1)HumidityCO

1 yr after birth HumidityCOETSExposure to house dust mite andanimal allergens

Amount of wheezing episodes in the last 6 monthsNightly wheezing in the last 6 monthsNocturnal coughing in the last 6 monthsSpecific IgE (Der p 1, Fel d 1, Can f 1)Doctors diagnosed asthmaTotal IgEConstitutional eczemaFood allergyNon-viral rhinitisOther upper airway infectionsOther lower airway infectionsDyspnoea

2 yr after birth HumidityCO

Amount of wheezing episodes in the last yearNightly wheezing in the last yearNocturnal coughing in the last yearSpecific IgE (Der p 1, Fel d 1, Can f 1)Doctors diagnosed asthmaTotal IgEConstitutional eczema in the last yearFood allergy in the last yearNon-viral rhinitis in the last yearOther upper airway infectionsOther lower airway infectionsDyspnoea

3 yr after birth ETSAllergen exposure

Amount of wheezing episodes in the last yearNightly wheezing in the last yearNocturnal coughing in the last yearDoctors diagnosed asthmaDoctors diagnosed cows milk allergyOther food-related allergies diagnosed by a doctorDoctors diagnosed inhalation allergiesNon-viral rhinitis in the last yearOther upper airway infections in the last yearOther lower airway infections in the last yearDyspnoea in the last year

Kuiper et al.

326

morbidity, carbon monoxide was measured inexpired air.

Tidal airway resistance and lung functionTo assess the tidal airway resistance by means ofan interrupter technique, microRint measure-ments are performed when the children are 4 and6 yr of age, while lung function [forced expira-tory volume in 1 s (FEV1), forced vital capacity(FVC)], reversibility on b2-mimetics (change inFEV1 after 400 lg salbutamol through a spacer)and bronchial responsiveness to increasing dosesof histamine (PC20) will be assessed when thechildren reach the age of 6 yr.

Statistical aspects

Power calculationThe power calculation of the RCPT study wasbased on a study of Saarinen and Kajosaari (26),in which it was shown that breast feeding as pre-ventive measure, in a comparable group of infantsas in the RCPT, resulted in a 30% reduction in thedevelopment of asthma. This percentage waschosen in the RCPT as the minimum clinicallyrelevant difference. A minimum amount of 194

infants in the intervention as well as in the controlgroup was required to reduce the asthma inci-dence with at least 30% (taking a type I error of0.05 and a power of 0.80 into account).The power calculation of the RCAT was based

on the number of parents required to measure thecompliance of the parents with the optimizedprevention program. Based on literature studies(27) an effect-size of 0.46 was expected. With aminimum of 38 participants in each group apower of 0.80 could be reached (taking a type Ierror of 0.05 and a drop out of at the most 18%into account).The power calculation of the natural history

study was based on a sample size that would belarge enough to show a statistically significantdifference, between the high- and low-riskgroups, in the development of asthma at 6 yr ofage (28). An odds ratio of at least 2 was assumedfor the association of a first-degree family historyof asthma with the risk to develop asthma indescendants. Taking a cumulative asthma inci-dence of 10%, a maximum drop out of 10% anda type I error of 0.05 into account a power of0.80 could be reached with an addition of 308low-risk infants.

Table 1. Continued


4 yr after birth Dust (Der p 1, Can f 1, Fel d 1)HumidityCOETS

Tidal airway resistance (microRint)Amount of wheezing episodes in the last yearNightly wheezing in the last yearNocturnal coughing in the last yearSpecific IgE (Der p 1, Fel d 1, Can f 1)Doctors diagnosed asthmaDoctors diagnosed cows milk allergyOther food-related allergies diagnosed by a doctorDoctors diagnosed inhalation allergiesTotal IgENon-viral rhinitis in the last yearOther upper airway infections in the last yearOther lower airway infections in the last yearDyspnoea in the last year


Amount of wheezing episodes in the last yearNightly wheezing in the last yearNocturnal coughing in the last yearDoctors diagnosed asthmaDoctors diagnosed cows milk allergyOther food-related allergies diagnosed by a doctorDoctors diagnosed inhalation allergiesNon-viral rhinitis in the last yearOther upper airway infections in the last yearOther lower airway infections in the last yearDyspnoea in the last year


AsthmaSpecific IgE (Der p 1, Fel d 1, Can f 1)

CO, carbon monoxide; RCPT, randomized-clinical prevention trial; IgE, immunoglobulin E; ETS, environmental tobacco smoke.


327

Data analysisIn the RCPT compliance will be evaluated interms of reduced exposure to HDM, cat and dogallergens as well as reduced smoking andincreased breast feeding.Differences in proportions between groups in

bivariate analyses will be tested with the chi-square test. Differences in mean values forcontinuous normally distributed variables areanalysed using the Student’s t-test. Multiplelogistic and linear regression techniques are usedfor multivariate analyses with a fixed follow-uptime, while Cox proportion hazard analyseswill be performed when the follow-up time isvariable.Concerning the cost-effectiveness one-way sen-

sitivity analyses regarding important assump-tions will be performed. In addition, statisticaluncertainty regarding the cost-effectiveness esti-mates will be assessed using non-parametricbootstrap re-sample methods. By means of theanalysis of effect modification it will be evaluated

whether the effect differs between subgroups. Formost analyses a p-value smaller than 0.05 isconsidered as significant.Statistical analyses are performed using the

Statistical Package for the Social Sciences (SPSS,Chicago, IL, USA) version 10.0.

Ethics approval and informed consent

Ethical approval for this study was obtainedfrom the ethics committees of the participatinginstitutes. All participants gave written, informedconsent.

Preliminary results of the RCPT

Preliminary results indicate that infants in theintervention group were significantly morebreast-fed (p ¼ 0.001) and/or received signifi-cantly more hypoallergenic formula feeding thaninfants in the control group. The first intake ofsolid food was significantly more often

Table 2. Timing schedule of the measurements of environmental exposures and outcome variables in the natural history study


Weekly from birth until 2 yr of age Day-care attendance Asthma-related complaintsHealth-care consumptionAbsent from work

3–5 days after birth Total IgE6 months after birth Type of feeding Amount of wheezing episodes in the last 6 months

Nightly wheezing in the last 6 monthsNocturnal coughing in the last 6 monthsDoctors diagnosed asthmaConstitutional eczema in the last 6 monthsFood allergy in the last 6 monthsNon-viral rhinitis in the last 6 monthsOther upper airway infectionsOther lower airway infectionsDyspnoea

7–9 months after birth Dust (Der p 1, Can f 1, Fel d 1)1 yr after birth ETS

Exposure to house dust mite and animal allergensAmount of wheezing episodes in the last 6 monthsNightly wheezing in the last 6 monthsNocturnal coughing in the last 6 monthsDoctors diagnosed asthmaConstitutional eczema in the last 6 monthsFood allergy in the last 6 monthsNon-viral rhinitis in the last 6 monthsOther upper airway infectionsOther lower airway infectionsDyspnoea

2 yr after birth Amount of wheezing episodes in the last yearNightly wheezing in the last yearNocturnal coughing in the last yearDoctors diagnosed asthmaConstitutional eczema in the last yearFood allergy in the last yearNon-viral rhinitis in the last yearOther upper airway infectionsOther lower airway infectionsDyspnoea

ETS, environmental tobacco smoke.

Kuiper et al.

328

postponed to the age of 6 months in the inter-vention group compared with the control group(p ¼ 0.001). No difference was found in theintervention compared with the control groupconcerning the exposure to tobacco smoke.Measurement of environmental exposure toHDM, cat and dog allergens was performedduring pregnancy (baseline) and exactly 1 yrlater. One year after baseline measurements theconcentration of HDM, cat and dog allergens,measured at the living room floor (except for dogallergens), the mattress of the parents and themattress of the baby, were statistically significantdecreased in the intervention group, but not inthe control group (29).

Discussion

The PREVASC study is a longitudinal pros-pective birth cohort study with its focus on thedevelopment of a primary prevention programfor asthma in children. The main goal of thePREVASC study is to assess the effectiveness of(the duration of) an intervention programfocussed on reducing the exposure to indoorand food allergens and ETS in children athigh risk of developing asthma. This is inves-tigated in a RCPT. To be able to define towhat extent first-degree familial predispositionfor asthma is predictive for the development ofasthma, a natural history study was started

with children with a high- or low-risk ofdeveloping asthma.Whether the participants are compliant to the

prevention program and whether optimizing theintervention leads to an increase in adherence isinvestigated in a RCAT. The health-economicbenefits of �high-risk� identification and primaryprevention of asthma will be assessed on the basisof the RCPT, taking compliance aspects andvalidity of screening into account.There are few prospective birth cohort studies

in which the effect of primary prevention ofasthma is studied (8–16). Most of these studiesfocus on one or two preventive measures (8–13).To our knowledge there are three multifacetedintervention studies, the study of Halken et al.(14), the CAAP (15) study and SPACE (16)study. The PREVASC study differs from thesestudies in several aspects. In contrast to thesestudies a low-risk group was included in thePREVASC study, which enables an estimation ofthe high-risk prediction and in additionfacilitates the evaluation of the external validityof the primary prevention study. In the PRE-VASC study, �high-risk� was defined as presenceof asthma in first-degree family members. Thisdefinition was chosen because asthma in first-degree family members is expected to be the mostpredictive risk factor for the development ofasthma in descendants (30) although this is oftenbased on retrospective and cross-sectional data.

Table 3. Timing schedule of the measurements of environmental exposures and outcome variables in the RCAT

Time-point Environmental exposures Primary and secondary outcome measures

Month 3 of pregnancy Dust (derp1)HumidityCO

Month 7 of pregnancy HumidityCO

2–3 wk after birth HumidityCO

Compliance with dietary advices

6 months after birth Dust (derp1)HumidityCO

Compliance with dietary advicesAmount of wheezing episodes in the last 6 monthsNightly wheezing in the last 6 monthsNocturnal coughing in the last 6 monthsDoctors diagnosed asthmaConstitutional eczema in the last 6 monthsNon-viral rhinitis in the last 6 monthsOther upper airway infectionsOther lower airway infections

1 yr after birth HumidityCO

Amount of wheezing episodes in the last 6 monthsNightly wheezing in the last 6 monthsNocturnal coughing in the last 6 monthsDoctors diagnosed asthmaConstitutional eczema in the last 6 monthsNon-viral rhinitis in the last 6 monthsOther upper airway infectionsOther lower airway infections

RCAT, randomized-clinical adherence trial.


329

We expect the strongest effect of the interventionprogram in children with a first-degree hereditarypredisposition of asthma. Concerning the inter-vention program, in the PREVASC study thedietary advices for the baby were exclusivelybreast feeding for at least 6 months as well asdelay of solid food until 6 months of age. Themother was not advised to follow a diet. As thegoal of the prevention program was that itshould be easy to implement into practice,prevention was based on easily applicable inter-vention measures. Recruitment was performedpredominantly by the GP. The Dutch health caresystem is organized in such a way that the GPkeeps complete medical records all registeredfamilies which makes the GP the most appropri-ate health care worker to assess whether anunborn child is at high risk of developing futureasthma.There are however, some limitations to our

study. Although there is a well-defined cleaningregimen some exposure could not be avoided.However, by keeping the cleaning regimen closeto daily practice, the HDM reduction that will bereached is realistic and can easily be obtained. Ingeneral, we expect that people will better adhereto the intervention program when the interven-tion is as close to daily practice as possible. Instudying a multifaceted intervention strategy it isimpossible to directly investigate the effectivenessof all single intervention measures in all possiblecombinations (like would be done in a factorialdesign), as this should result in a sample size thatis far too large for in one trial. Moreover,synergistic effects can be missed in comparingsingle interventions only, while the most pro-nounced clinical effect possible should be expec-ted from a combined intervention. Finally, theGP and the parents could not be blinded to theintervention, which might result into classifica-tion bias and reporting bias with respect to med-ical records and questionnaires in relation tohealth outcome. Therefore, it is important tofollow the cohort until the children have reachedthe age of 6 yr, at which age lung functionmeasurement can be performed.

AcknowledgmentsThe PREVASC study was financially supported by theDutch Asthma Foundation, the Prevention Fund (ZON-MW), the Royal Academy of Science (KNAW), Astra-Zeneca. The study was a joint project of the Departments ofGeneral Practice of the University of Maastricht and theRadboud University Nijmegen. Authors would like tothank the general practitioners and practice assistants alliedto these departments and the participating midwives forrecruiting eligible families. Authors are grateful to all fam-ilies for their participation in the study. Also thank the

research assistants: Kittie Coolen, Ilse van Beerendonck,Kitty van der Meer, Frieda Waasdorp, Bernadette Zinzen,Anita Legtenberg, Kirsten Ritter, Maaike Jilderda, ManonLanghoor; the data manager: Jacqueline Pisters; the statis-tician: Jildou Sijbrandij; the nurses who visited the partici-pating families and sampled data: Cindy Brink, TwannyJeijsman, Marjorie Rovers, Elly de Vree, Ineke Schaap,Marjan Terhaar, Gertie van Driel, Mariken Dorsemagen,Mieke Witte, the PREVASC secretaries: Annemarie Span-inks and Angelique Lefebure and the secretary of Care andPublic Health Research Institute (CAPHRI): AnnemarieKorsten; and finally thank all other people who were sci-entifically involved in the PREVASC study: Hans Severens,Geertjan Wesseling, Marielle Goossens, Guido van denBoom, Ilse Mesters, Lisette van den Bemt, Annemiek Nij-holt, Karin Manders, Alfons Schrooten, Ruud Schmeitz,Han Hendriks, Rein Jobsis, Bart Thoonen, Willem vanGerwen, Wil van den Bosch and Job Metsemakers.

References

1. Austin JB, Selvaraj S, Russell G. Childhood asthmain the highlands of Scotland – morbidity and schoolabsence. Scott Med J 2004: 49: 18–21.

2. Vermeire PA, Rabe KF, Soraiano JB, Maier WC.Asthma control and differences in managementpractices across seven European countries. Respir Med2002: 96: 142–9.

3. Peat JK, van den Berg RH, Green WF, Mellis CM,Leeder SR, Woolcock AJ. Changing prevalence ofasthma in Australian children. BMJ 1994: 308: 1591–6.

4. Rona RJ, Chin S, Burney PG. Trends in the preval-ence of asthma in Scottish and English primary schoolchildren 1982–92. Thorax 1995: 50: 992–3.

5. Heinzmann A, Deichmann KA. Genes for atopy andasthma. Curr Opin Allergy Clin Immunol 2001: 1: 387–92.

6. Von Mutius E. The environmental predictors of allergicdisease. J Allergy Clin Immunol 2000: 105: 9–19.

7. Warner JA, Jones CA, Jones AC, Warner JO. Pre-natal origins of allergic disease. J Allergy Clin Immunol2000: 105: S493–6.

8. Zeiger RS, Heller S, Mellon MH. Effect of com-bined maternal and infant food-allergen avoidance ondevelopment of atopy in early infancy: a randomizedstudy. J Allergy Clin Immunol 1989: 84: 72–89.

9. Custovic A, Simpson BM, Murray CS, Lowe L,Woodcock A. The national asthma campaign Man-chester asthma and allergy study. J Allergy ClinImmunol 2000: 105: 252–8.

10. Brunekreef B, Smit J, de Jongste J, et al. The pre-vention and incidence of asthma and mite allergy (PI-AMA) birth cohort study: design and first results.Pediatr Allergy Immunol 2002: 13 (Suppl. 15): 55–60.

11. Arshad SH, Matthews S, Gant C, Hide DW. Effect ofallergen avoidance on allergic disorders in infancy.Lancet 1992: 339: 1439–97.

12. Marini A, Agosti M, Motta G, Mosca F. Effects of adietary and environmental prevention program on theincidence of allergic symptoms in high atopic risk in-fants: three years� follow-up. Acta Paediatr Suppl 1996:414: 1–21.

13. Mihrshahi S, Peat JK, Webb K, et al. The ChildhoodAsthma Prevention Study (CAPS): design and researchprotocol of a randomised trial for the primary preven-tion of asthma. Control Clin Trials 2001: 22: 333–54.

Kuiper et al.

330

14. Halken S, Host A, Hansen LG, Osterballe O. Effectof an allergy prevention program on incidence of atopicsymptoms in infancy. A prospective study of 159 �high-risk� infants. Allergy 1992: 47: 545–53.

15. Chan-Yeung M, Manfreda J, Kimich-ward H,Ferguson A, Watson W, Becker A. A randomizedcontrolled study on the effectiveness of a multifacetedintervention program in the primary prevention ofasthma in high-risk infants. Arch Pediatr Adolesc Med2000: 154: 657–63.

16. Halmerbauer G, Gartner C, Schierl M, et al. Studyon the Prevention of Allergy in Children in Europe(SPACE): allergic sensitization in children at 1 year ofage in a controlled trial of allergen avoidance frombirth. Pediatr Allergy Immunol 2002: 13 (Suppl. 15):47–54.

17. The WONCA Classification Committee. ICPC-2.International Classification of Primary Care. Oxford,UK: Oxford University Press, 1998.

18. Dirksen WJ, Geijer RMM, de Haan M, de Koning G,Flikweert S, Kolnaar BGM. Astma bij kinderen.NHG-Standaard. Huisarts Wet 1998: 41: 130–43.

19. Van Assema P, Mesters I, Kok G. Het focusgroep-interview: een stappenplan (The focus group interview:stepwise guidelines). Tijdschr Soc Gezondheidsz 1992:70: 431–7.

20. Oostenbrink JB, Koopmanschap MA, Rutten FFH.Handleiding voor kostenonderzoek. Methoden enrichtlijnprijzen voor economische evaluaties in degezondheidszorg. Amstelveen: College voor zorgverze-keringen, 2000.

21. Asher MI, Keil U, Anderson HR, et al. InternationalStudy of Asthma and Allergies in Childhood (ISAAC):rationale and methods. Eur Respir J 1995: 8: 483–91.

22. Van der Heide S, van Aalderen WM, Kauman HF,Dubois AE, de Monchy JG. Clinical effects of aircleaners in homes of asthmatic children sensitized to pet

allergens. J Allergy Clin Immunol 1999: 104 (2 Pt 1):447–51.

23. Luczynska CM, Arruda LK, Platts-Mills TA,Miller JD, Lopez M, Chapman MD. A two-sitemonoclonal antibody ELISA for the quantification ofthe major Dermatophagoides spp. allergens, Der p I andDer f I. J Immunol Methods 1989: 118: 227–35.

24. Ingram JM, Sporik R, Rose G, Honsinger R, Chap-

man MD, Platts-Mills TA. Quantitative assessmentof exposure to dog (Can f 1) and cat (Fel d 1) allergens:relation to sensitization and asthma among childrenliving in Los Alamos, New Mexico. J Allergy ClinImmunol 1995: 96: 449–56.

25. Aalberse RC, Koshte V, Clemens JG. Immunoglob-ulin E antibodies that crossreact with vegetable foods,pollen en hymenoptera venom. J Allergy Clin Immunol1981: 68: 356–64.

26. Saarinen UM, Kajosaari M. Breastfeeding as pro-phylaxis against atopic disease: prospective follow-upstudy until 17 years old. Lancet 1995: 346: 1065–9.

27. Kok G, van den Borne B, Mullen PD. Effectiveness ofhealth education and health promotion: meta-analysesof effect studies and determinants of effectiveness.Patient Education Counseling 1997: 30: 19–27.

28. Martinez FD, Wright AL, Taussig LM, Holberg CJ,Halonen M, Morgan WJ. Asthma and wheezingin the first six years of life. New Engl J Med 1995: 332:133–8.

29. Schonberger HJAM, Maas T, Dompeling E, Knot-

tnerus JA, van Weel C, van Schayck CP. Complianceof asthmatic families with a primary prevention pro-gram of asthma and effectiveness of measures to reduceinhalant allergens – a randomised trial. Clin Exp Allergy2004: 34: 1024–31.

30. Burke W, Fesinmeyer M, Reed K, Hampson L,Carlsten Ch. Family history as a predictor of asthmarisk. Am J Prev Med 2003: 24: 160–9.


331

The primary prevention of asthma in children study: Design of a multifaceted prevention program

Documents

Transcript of The primary prevention of asthma in children study: Design of a multifaceted prevention program