UBC Faculty Research and Publications

Health related quality of life in 3 and 4 year old children and their parents: preliminary findings about… Klassen, Anne F; Landgraf, Jeanne M; Lee, Shoo K; Barer, Morris; Raina, Parminder; Chan, Herbert W; Matthew, Derek; Brabyn, David Dec 22, 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12955_2003_Article_81.pdf [ 549.84kB ]
JSON: 52383-1.0223539.json
JSON-LD: 52383-1.0223539-ld.json
RDF/XML (Pretty): 52383-1.0223539-rdf.xml
RDF/JSON: 52383-1.0223539-rdf.json
Turtle: 52383-1.0223539-turtle.txt
N-Triples: 52383-1.0223539-rdf-ntriples.txt
Original Record: 52383-1.0223539-source.json
Full Text

Full Text

ralHealth and Quality of Life OutcomesssBioMed CentOpen AcceResearchHealth related quality of life in 3 and 4 year old children and their parents: preliminary findings about a new questionnaireAnne F Klassen*1, Jeanne M Landgraf2, Shoo K Lee3, Morris Barer4, Parminder Raina5, Herbert WP Chan1, Derek Matthew6 and David Brabyn7Address: 1Centre for Community Child Health Research, L408, 4480 Oak Street, Vancouver, BC, V6H 3V4, Canada, 2HealthAct 205 Newbury Street, Boston, MA USA, 3Centre for Healthcare Innovation and Improvement, Dept of Pediatrics, University of British Columbia, Vancouver, BC, Canada, 4Centre for Health Services and Policy Research, Department of Healthcare & Epidemiology, University of British Columbia, Vancouver, BC, Canada, 5Evidence-Based Practice Centre, Dept of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, L8N 3Z5, Canada, 6Greater Victoria Hospital Society, 35 Helmcken Road Victoria, BC, V8Z 6R5, Canada and 76470 Berkley Place, Burnaby, BC, V5E 4G5, CanadaEmail: Anne F Klassen* - afk@interchange.ubc.ca; Jeanne M Landgraf - jml@healthact.com; Shoo K Lee - shool@interchange.ubc.ca; Morris Barer - mbarer@chspr.ubc.ca; Parminder Raina - praina@mcmaster.ca; Herbert WP Chan - hwphan@cw.bc.ca; Derek Matthew - james.matthew@caphealth.org; David Brabyn - dbrabyn@shaw.ca* Corresponding author    AbstractBackground: Few measures of health related quality of life exist for use with preschool agedchildren. The objective of this study was to assess reliability and validity of a new multidimensionalgeneric measure of health-related quality of life developed for use with preschool children.Methods: Cross-sectional survey sent to parents as their child turned 3 1/2 years of age. Thesetting was the province of British Columbia, Canada. Patients included all babies admitted totertiary level neonatal intensive care units (NICU) at birth over a 16-month period, and aconsecutive sample of healthy babies. The main outcome measure was a new full-lengthquestionnaire consisting of 3 global items and 10 multi-item scales constructed to measure thephysical and emotional well-being of toddlers and their families.Results: The response rate was 67.9%. 91% (NICU) and 84% (healthy baby) of items correlatedwith their own domain above the recommended standard (0.40). 97% (NICU) and 87% (healthybaby) of items correlated more highly (≥ 2 S.E.) with their hypothesized scale than with otherscales. Cronbach's alpha coefficients varied between .80 and .96. Intra-class correlation coefficientswere above .70. Correlations between scales in the new measure and other instruments weremoderate to large, and were stronger than between non-related domains. Statistically significantdifferences in scale scores were observed between the NICU and healthy baby samples, as well asbetween those diagnosed with a health problem requiring medical attention in the past year versusthose with no health problems.Conclusions: Preliminary results indicate the new measure demonstrates acceptable reliabilityand construct validity in a sample of children requiring NICU care and a sample of healthy children.However, further development work is warranted.Published: 22 December 2003Health and Quality of Life Outcomes 2003, 1:81Received: 27 August 2003Accepted: 22 December 2003This article is available from: http://www.hqlo.com/content/1/1/81© 2003 Klassen et al; licensee BioMed Central Ltd. This is an Open Access article: verbatim copying and redistribution of this article are permitted in all media for any purpose, provided this notice is preserved along with the article's original URL.Page 1 of 11(page number not for citation purposes)Health and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/81BackgroundThere are now a number of validated health-related qual-ity of life (HRQL) instruments available for use withadults, and these are often routinely included in clinicaltrials. Such measures are based on the view that health ismultidimensional, that the concepts forming thesedimensions can be assessed only by subjective measures,and that quality of life should be evaluated by asking thepatient, or in some cases a proxy. Measurement of HRQLin children is based on these same principles, but is at anearlier stage of development [1].HRQL assessment in children is complicated by develop-mental issues and by the need to use proxies in certain cir-cumstances (e.g., preschool aged children). Somedevelopers have addressed these issues by creating sepa-rate questionnaires for specific age-groups and for parentand child report. The PedsQL generic measure of HRQL,for example, has 4 parent report measures (ages 2–4, 5–7,8–12 and 13–18 years old) and 3 child self-report meas-ures (ages 5–7, 8–12 and 13–18 years old) [2].Developmental issues are most relevant to the preschoolaged group, who undergo rapid growth and development[3]. Since preschool aged children are not able to com-plete a questionnaire for themselves, the use of a proxy isessential. A growing number of studies have looked at theproxy issue in school aged children. Eiser and Morse(2001) performed a systematic review and reported thatthat there was greater agreement for observable function-ing (e.g. physical HRQoL), and less for non-observablefunctioning (e.g. emotional or social HRQoL), and thatagreement was better between parents and chronicallysick children compared with parents and their healthychildren [4]. These authors suggest there remain strongarguments for obtaining information from both parentsand children whenever possible.A recent systematic review [1] and a number of otherreview articles [5-8] describe the range of generic healthrelated quality of life (HRQL) measures for childrendeveloped to date. At the time of the present study, genericquestionnaires were developed to measure HRQL forschool-aged children only. However, a full-length ques-tionnaire still under development – the Infant/ToddlerQuality of Life Questionnaire (ITQOL) – was made avail-able for purposes of further evaluation (9). The ITQOL isconceptually similar to the Child Health Questionnaire(there is some overlap of items and scales) [10]. Bothmeasures adopt the World Health Organization's defini-tion of health, which is "a state of complete physical,mental and social well-being and not merely the absenceof disease" [11]. The ITQOL was developed following a[12], which identified core child health concepts andresulted in the development of items and scales to meas-ure physical function, growth and development, bodilypain, temperament and moods, behavior and generalhealth perceptions. Like the CHQ, the ITQOL alsoincludes scales to measure parental impact (time andemotions).Since the inception of the current project, two new genericmeasures for pre-school aged have since become available[2,13]. In The Netherlands, Fekkes and colleagues [13]developed the TNO-AZL Preschool Quality Of Life(TAPQOL), a 43-item (12-domain) generic pre-schoolmeasure of health status, and used this instrument in astudy of preterm infants [14]. HRQL in this measure wasdefined as health status in 12 domains weighted by theimpact of health status problems on wellbeing. These 12domains measure aspects of physical, social, cognitiveand emotional function. Varni et al, in the USA [2], devel-oped the generic 23-item Pediatric Quality of Life Inven-tory (PedsQL), which can be used to measure 3 domainsof health (physical, mental and social) in children andadolescents aged 2 to 18.The aim of the current paper is to present preliminaryinformation about the psychometric properties of theITQOL questionnaire as applied in two samples of pre-school aged children: a population-based follow-up studyof children admitted at birth to level III neonatal intensivecare units (NICU) (i.e., regional neonatal-perinatal cent-ers that provide care for high risk pregnancies and inten-sive care for severely ill infants); and a comparison groupof healthy full-term births. The overall purpose of ourstudy was to link questionnaire survey data with adminis-trative health data for NICU children and their caregiversto examine relationships between health care utilization,initial NICU birth experience and long-term health out-comes for respondents. Research describes a range of neg-ative health outcomes associated with neonatal intensivecare [15-27]. Commonly reported adverse outcomesinclude cerebral palsy, mental retardation, deafness,blindness as well as more widespread problems such aslearning disabilities and behavioral problems. Results per-taining to HRQL outcomes in our sample of NICU gradu-ates are reported in a separate publication [28].MethodsNICU sampleOur sample included all surviving babies admitted formore than 24 hours to one of 3 level III NICUs in BritishColumbia (Canada) over a 16-month period (March1996 through June 1997 inclusive). These 3 units (atRoyal Columbian Hospital, Victoria General Hospital andPage 2 of 11(page number not for citation purposes)thorough review of the infant health literature and areview of developmental guidelines used by pediatriciansBritish Columbia Women's and Children's Hospital) pro-vided 100% of the tertiary care NICU beds in the provinceHealth and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/81at the time. Mothers' name and contact details wereobtained from each hospital. This population of babieswas then matched with provincial mortality records toidentify and exclude any babies that had died after dis-charge from the NICU. To ensure the data were independ-ent, only families with one child in the study sample wereincluded in this paper.Healthy baby sampleOur comparison sample of healthy term babies wasrecruited from the two hospitals with an affiliated hospi-tal-based primary care unit (BC Women's and Children'sHospital and the Royal Columbian Hospital). This sam-ple included all babies delivered over 11 months (March1996 through January 1997 inclusive) by any primary carephysician from these two units working within either ofthese two hospitals. Multiple births, babies with a siblingin the NICU sample, and babies subsequently admitted toa NICU for more than 24 hours were excluded. Contactdetails for the mother were obtained from the healthrecords department at one hospital and directly from theprimary care unit at the other.Data collectionA questionnaire booklet, that included a number of sepa-rate instruments, was sent to each mother as her childturned 3 1/2 years of age. A consent letter was included toobtain permission to link the questionnaire data withhospital birth records. The caregiver that had, to that pointin the child's life, spent the most amount of time with thechild was asked to complete the questionnaire. Non-respondents were sent a reminder letter and up to twomore copies of the questionnaire as necessary. Finally,phone calls were made as part of a final effort to reachfamilies. If the telephone number was not in service orreassigned, or a questionnaire was returned to us from thepost office as undeliverable, a comprehensive search strat-egy was implemented. The process involved searching theInternet and/or contacting the mothers' primary care phy-sician to obtain an address.Infant Toddler Quality of Life QuestionnaireThe questionnaire booklet included the developmentalfull-length version of the Infant Toddler Quality of LifeQuestionnaire (ITQOL) [9,29]. The prototype contains103-items that measure 8 infant and 5 parental concepts(see Table 1). This instrument was developed for infantsas young as 2 months and toddlers up to five years of ageusing developmental guidelines used by pediatricians andother published literature [12]. More than half the itemsin each scale must be answered in order to derive a score.Raw scores are calculated for each scale by computing thealgebraic mean of the items. Following published conven-tion [30], raw scores are then transformed to a scale fromTable 1: Infant Toddler Quality of Life Questionnaire – General ContentInfant concepts No. items General ContentPhysical Abilities 10 Amount of limitation in physical activities, such as eating, sleeping, grasping, and playing due to health problemsGrowth and Development 10 Satisfaction with development (physical growth, motor, language, cognitive), habits (eating, feeding, sleeping) and overall temperamentBodily Pain/Discomfort 3 Amount, frequency of bodily pain/discomfort and the extent to which pain/discomfort interferes with normal activitiesTemperament and Moods 18 Frequency of certain moods and temperaments, such as sleeping/eating difficulties, crankiness, fussiness, unresponsiveness, playfulness and alertnessGeneral Behavior Perceptions 13 Perceptions of current, past and future behaviorGetting Along with Others 15 Frequency of behavior problems, such as following directions, hitting, biting others, throwing tantrums, and easily distracted. Frequency of positive behaviors, such as ability to cooperate, appears to be sorry, and adjusts to new situationsGeneral Health Perceptions 12 Perceptions of current, past and future healthChange in health 1 Perceptions of changes in health over the past yearParent conceptsImpact-Emotional 7 Amount of worry experienced by parent due to child's eating/sleeping habits, physical and emotional well-being, learning abilities, temperament, behavior and ability to interact with others in an age-appropriate mannerImpact-Time 7 Amount of time limitations experienced by parent (time for his/her own needs) due to child's eating/sleeping habits, physical and emotional well-being, learning abilities, temperament, behavior and ability to interact with others in an age-appropriate mannerMental Health 5 Parent's general mental health, including depression, anxiety, behavioral-emotional control, and general positive affectGeneral Health 1 Rating of parent's overall healthFamily Cohesion 1 Rating of family's ability to get along with one anotherPage 3 of 11(page number not for citation purposes)Health and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/810 (worst health) to 100 (best health).Item-level analysisData completeness was measured by computing the per-centage of items completed for each scale and the instru-ment. Following published conventions [31-36], item-to-scale correlations (corrected for overlap) were consideredsatisfactory for items that correlated .40 or more with theirhypothesized scale. Item discriminant validity was con-sidered successful if the correlation between an item andits hypothesized scale was significantly higher (≥ 2 S.E.)than correlations between that item and all other scales.As advised with newly created scales [30], the percentageof correlations that were ≥1 S.E. higher for each item andits hypothesized scale were also examined.Scale-level analysisFor each scale, we determined the percentage of scoresthat could be computed. The distribution of scores wasexamined to determine potential floor and ceiling effect(i.e., people scoring at the absolute lowest and highestends of the continuum for each scale). Scale internal con-sistency was assessed in terms of Cronbach's α coefficient.Internal consistency was considered satisfactory if thecoefficient was at least .70 [37,38]. To evaluate the degreeto which each scale was "unique", correlations among allscales were examined and compared against the respectiveCronbach's α reliability coefficient observed for each indi-vidual scale. In general, the correlation between scalesshould be less than the alpha coefficient achieved for anindividual scale [37]. To examine test-retest reliability, arandom sample of 80 NICU respondents, who indicatedthey would be willing to participate in further research,was contacted by telephone. Those that agreed to partici-pate were sent a copy of the ITQOL in the mail. A secondcopy of the questionnaire was mailed out once it was con-firmed that the first copy had been completed. Test-retestreliability was assessed through intra-class correlationcoefficients. ICCs of at least .70 were considered satisfac-tory [37,38]Concurrent validityTo test concurrent validity, scale scores in the ITQOL werecorrelated with scores for similar and dissimilar scales inthree validated instruments: the Child Behavior Checklist/1.5-5 (CBCL/1.5-5)[39]; the SF-36 [40,41]; and the Fam-ily Assessment Device (FAD) [42]. Scales from each instru-ment that are intended to measure similar constructsshould have higher correlations (convergent validity)with each other than with scales that measure unrelatedconstructs (divergent validity). Correlations of <0.20 wereconsidered negligible; 0.20 to 0.34 weak; 0.35 to 0.50moderate; and >0.50 strong [43].Child Behavior Checklist (CBCL/1.5-5)Since no validated multidimensional generic measure ofHRQL was available for validation purposes, we used ameasure of behavior as 55% of items in the ITQOL meas-ure child behavior or temperament. The CBCL/1.5-5measures behavioral, emotional and social functioning inchildren 1 1/2 to 5 years of age. This 100-item instrumentmeasures both internalizing and externalizing syndromesand can be summed to produce a total problem score. Ahigher score reflects greater presence and severity ofsymptoms.Short Form 36The SF-36 [40,41] assesses the following 8 domains ofadult health: physical health; physical role limitations;emotional role limitations; mental health; social func-tion; energy; pain; and general health perception, and wasused to help validate the ITQOL parent-impact scales.Since the mental health domain and one item from gen-eral health perception are included in the ITQOL, theremaining 6 domains were used in the validation process.Scores on these domains can range from 0 (worst health)to 100 (best health).Family Assessment DeviceThe Family Assessment Device (FAD) [42] is a measure offamily functioning and was used to help validate the Fam-ily Cohesion item. Scores for this 12-item scale can rangefrom 0 to 36 with higher scores indicating greaterdysfunction.Discriminant validityThe ability of the ITQOL to discriminate between groupsof children with poorer expected outcomes was deter-mined by comparing ITQOL scale scores for the followingtwo dichotomous variables (using Mann-Whitney U-testfor statistical significance): (1) NICU vs. healthy babysample; and (2) children with one or more health prob-lems (from a list of 16 common childhood conditions) vs.children with no health problems. The NICU sample andthe group with one or more health problems wereexpected to have poorer reported health. Effect size statis-tics (i.e., mean difference divided by pooled s.d.) werecomputed to determine the magnitude of the difference inmean scores.ResultsQuestionnaires were sent to mothers of 1,907 NICUbabies and 718 healthy babies. Fifty percent of familieshad moved at least one time since the birth of their baby.Using our search strategy, we were able to locate 81% offamilies. The overall response rate (after 131 exclusions,e.g. deaths, language issues) was 54.9%, and the responsePage 4 of 11(page number not for citation purposes)rate for families we successfully located was 67.9%, withcompleted questionnaires received for 972 NICU familiesHealth and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/81and 393 healthy baby families. The response rate for theNICU sample did not vary from that of the healthy babysample. Five NICU respondents returned a signed consentform without a completed questionnaire and weredropped from the analysis.For both samples combined, the mean age of the respond-ents was 35 (s.d. 5.7; range 19 to 65). Most respondents,(98.1%), were the child's biological parent, most com-monly the child's mother (94.6%), and most (85.6%)were married or living in a common-law relationship. Nodifferences were found between the NICU and healthybaby group in terms of parental age, gender, marital statusor educational level. The proportion of boys in the samplewas 55.1%. The sample was composed of 926 (68%)three-year olds, 413 (30.3%) four-year olds, and 23(1.7%) five-year olds. The five-year old children havebeen excluded from the psychometric analysis since thisgroup is unlikely to be representative.Item-level analysisItem-level results are presented in Table 2. Sixty-seven per-cent of respondents in the NICU sample and 74% ofrespondents in the healthy baby sample answered all 103items. This was lower than the 83% (NICU sample) and94% (healthy baby sample) of respondents who com-pleted all items for the similar length CBCL/1.5-5. Ofthose missing at least one response on the ITQOL, threequarters of respondents in both samples missed answer-ing only 3 items or less. The rate of missing data withineach scale varied from 2.3% (Impact-emotional) to 8.9%(General Health Perception) for the NICU sample, andFor item-scale correlations, 91% (NICU sample) and 84%(healthy baby sample) of items correlated with their owndomain above the recommended standard (0.40). Withindomains, perfect results were obtained for 7 (NICU sam-ple) and 5 (healthy baby sample) scales. For item-discri-minant validity, 97% (NICU sample) and 87% (healthybaby sample) of items correlated more highly (≥ 2 S.E.)with their hypothesized scale than with other scales. Per-fect results (100%) were attained for 8 of the 10 scales inthe NICU sample, and 6 of the 10 scales in the healthybaby sample. Only 2 items in the NICU sample (in Get-ting Along) and 5 in the healthy baby sample (in Temper-ament and Moods, General Behavior and Getting Along)did not correlate ≥ 1 S.E. with its hypothesized scales.Scale-level analysisScale-level results are presented in Tables 3 and 4. The pro-portion of missing values for scored domains was small:2.9% (Physical Abilities) or less. There were no flooreffects, but ceiling effects (scores of 100%) were apparent.The largest ceiling effect (69.3% NICU; 85.8% healthybaby) was in the Physical Abilities scale. The range ofscores was particularly skewed for three scales (PhysicalAbilities, Growth/Development, Bodily Pain) where morethan 84% of respondents in both samples reported scoresof 75 or higher. Scores for scales that assess aspects ofemotional and behavioral function showed morevariability.For both samples, the Cronbach's alpha coefficients were.80 or higher. One scale (Physical Abilities) achieved acoefficient of .96. The correlations between the ITQOLTable 2: ITQOL item-level analysis for the NICU and healthy baby samplesNICU Healthy babyNo. items % missing Item internal consistencyItem discriminant validity% missing Item internal consistencyItem discriminant validityInfant scales -1 S.E. -2 S.E. -1 S.E. -2 S.E.Physical Abilities 10 4.1 100 100 100 2.8 100 100 100Growth and Development 10 2.4 100 100 100 3.1 100 100 100Bodily Pain/Discomfort 3 5.1 100 100 100 2.3 100 100 100Temperament and Moods 18 8.8 89 100 100 7.8 67 89 78General Behavior 13 5.3 92 100 100 6.2 92 92 92Getting Along with Others 15 8.1 60 87 87 7.2 60 87 53General Health Perceptions 12 8.9 100 100 100 5.2 83 100 100Parent scalesImpact-Emotional 7 2.3 100 100 86 0.8 86 100 86Impact-Time 7 2.7 100 100 100 1.6 100 100 100Mental Health 5 3.4 100 100 100 2.1 100 100 100Page 5 of 11(page number not for citation purposes)from 0.8% (Impact-emotional) to 7.8% (Temperamentand Moods) for the healthy baby sample.scales were on average moderate (see Table 5 and 6). AllHealth and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/81Table 3: ITQOL scale level analysis: % not scored and categorized percentile distribution of scores for the NICU sample% not scored Categorized percentile distribution: Scale 0–100Infant scales 0–24 25–49 50–74 75–99 100Physical Abilities 2.9 3.4 2.1 2.6 19.6 69.3Growth and Development 0.5 1.2 2.2 8.6 56.1 31.4Bodily Pain/Discomfort 0.8 1.2 2.1 11.6 36.1 48.2Temperament and Moods 0.9 0 1.9 24.1 71.6 1.5General Behavior 0.7 0.4 8.0 40.0 48.2 2.6Getting Along with Others 1.1 0 2.5 44.6 51.4 0.4General Health Perceptions 1.1 1.6 9.6 34.7 51.4 1.8Parent scalesImpact-Emotional 1.4 1.8 4.9 21.5 54.8 15.5Impact-Time 1.9 1.8 4.7 13.0 37.2 41.4Mental Health 1.9 2.1 6.9 30.5 56.2 2.4Table 4: ITQOL Scale Level Analysis: % not scored and categorized percentile distribution of scores for the healthy baby sample% not scored Categorized percentile distribution: Scale 0–1000–24 25–49 50–74 75–99 100Infant scalesPhysical Abilities 2.6 1.8 0.5 0.5 8.8 85.8Growth and Development 0.5 0.3 0.5 2.8 51.7 44.2Bodily Pain/Discomfort 0.8 0.3 1.3 8.8 41.1 47.8Temperament and Moods 1.6 0 0.5 17.8 78.3 1.8General Behavior 1.0 0.5 4.1 37.7 55.0 1.6Getting Along with Others 1.3 0 2.1 35.4 61.0 0.3General Health Perceptions 0.5 0.3 3.1 23.0 72.1 1.0Parent scalesImpact-Emotional 0.0 0.3 3.6 11.4 68.7 16.0Impact-Time 1.3 0.3 2.6 9.8 37.7 48.3Mental Health 1.0 0.3 7.8 28.4 59.7 2.8Table 5: Cronbach's α reliability coefficients and inter-scale correlations (Spearman) of the ITQOL scales for the NICU samplePA GD BP TM GB BE GH PI-E PI-T MHInfant scalesPhysical Abilities (PA) (.96)Growth Development (GD) .50 (.89)Bodily Pain/Discomfort (BP) .28 .34 (.88)Temperament and Moods (TM) .32 .53 .47 (.86)General Behavior (GB) .27 .44 .23 .55 (.88)Getting Along with Others (BE) .29 .48 .27 .67 .74 (.80)General Health Perceptions (GHP) .36 .43 .38 .42 .34 .39 (.86)Parent scalesImpact-Emotional (PI-E) .30 .50 .38 .62 .59 .61 .44 (.86)Impact-Time (PI-T) .35 .44 .39 .57 .50 .55 .37 .66 (.89)Page 6 of 11(page number not for citation purposes)Mental Health (MH) .18 .27 .23 .40 .32 .36 .36 .42 .43 (.84)Health and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/81correlations between scales were less than their reliabilitycoefficients, providing evidence of unique reliable varia-ble measured by each scale.For test-retest reliability, 2 copies of the ITQOL werereceived from 71% of the families who agreed to partici-pate. Mailings were separated on average by 13 (s.d. 5)days. Intra-class correlation coefficients exceeded the .70benchmark, and were as follows: Physical Abilities = .80;Growth/development = .84; Bodily Pain = .71;Temperament = .75; General Behavior = .94; GettingAlong = .87; General Health Perception = .89; MentalHealth = .83; Impact-emotional = .77; and Impact-time =.75.Concurrent validityCorrelations between related scales in the ITQOL andother standardized instruments were strong (see Tables 7and 8). Specifically, Getting Along, Temperament, andGeneral Behavior correlated more strongly with CBCLsyndrome and total problem scores and less strongly withdomains that measure aspects of physical health. Simi-larly, as anticipated, the parental impact scales (emotionaland time) correlated more strongly with SF-36 psychoso-cial scales than with SF-36 physical scales. The familycohesion item correlated strongly with the Family Func-tion Scale and weakly or moderately with all other scales.Table 6: Cronbach's α reliability coefficients and inter-scale correlations (Spearman) of the ITQOL scales for the healthy baby samplePA GD BP TM GB BE GH PI-E PI-T MHInfant scalesPhysical Abilities (PA) (.96)Growth and Development (GD) .38 (.82)Bodily Pain/Discomfort (BP) .12 .21 (.85)Temperament and Moods (TM) .25 .43 .30 (.82)General Behavior (GB) .26 .39 .07 .42 (.87)Getting Along with Others (BE) .25 .36 .09 .50 .70 (.80)General Health Perceptions (GH) .22 .27 .23 .29 .29 .36 (.80)Parent scalesImpact-Emotional (PI-E) .23 .43 .25 .48 .53 .49 .35 (.82)Impact-Time (PI-T) .27 .34 .27 .45 .43 .45 .32 .54 (.88)Mental Health (MH) .18 .22 .16 .30 .32 .37 .28 .39 .34 (.81)Single items are not included in these analyses.Table 7: Convergent and divergent validity for the NICU: Spearman's correlations between ITHQ domain scores and CBCL/1.5-5 scales, SF-36 domain scores and FADCBCL/1.5-5 SF-36 FADInfant scales Internal External TotalProblemPhysical RolePhysicalPain RoleMentalEnergy SocialfunctionPhysical abilities -.35 -.21 -.29 .17 .10 .10 .20 .12 .15 -.22Growth and Development -.51 -.39 -.51 .17 .14 .18 .24 .25 .26 -.29Bodily Pain/discomfort -.31 -.27 -.33 .14 .19 .23 .24 .17 .22 -.15Temperament and moods -.60 -.52 -.60 .24 .26 .24 .29 .30 .38 -.36General behavior -.47 -.63 -.59 .20 .15 .19 .18 .23 .26 -.35Getting along with others -.61 -.68 -.69 .27 .18 .23 .24 .24 .32 -.36General health perception -.39 -.29 -.37 .19 .14 .18 .23 .23 .25 -.28Parent scalesImpact-emotional -.57 -.59 -.63 .24 .23 .26 .32 .32 .35 -.33Impact-time -.46 -.50 -.51 .29 .26 .32 .35 .33 .38 -.34Mental health -.37 -.38 -.40 .25 .31 .35 .48 .57 .52 -.43General health – parent -.27 -.28 -.31 .38 .34 .45 .28 .48 .38 -.30Family cohesion -.24 -.26 -.27 .13 .18 .20 .29 .29 .33 -.58Page 7 of 11(page number not for citation purposes)CBCL/1.5-5 domains: internalizing syndromes; externalizing syndromes; total problems score; FAD: Family Assessment DeviceHealth and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/81Discriminant validityTable 9 and 10 presents findings for tests of discriminantvalidity. Parents of NICU children reported their childrenas having significantly poorer HRQL than children in thehealthy baby group for 5 of the child scales. Scores for theNICU sample were also lower for the 3 parent scales.In the NICU sample, those with children with at least onehealth problem that required treatment in the past yearhad poorer reported HRQL in all areas compared withthose without health problems. In the healthy baby sam-ple, significant differences were noted for 4 of the childscales.Table 8: Convergent and divergent validity for the healthy baby sample: Spearman's correlations between ITHQ domain scores and CBCL/1.5-5 scales, SF-36 domain scores and FADCBCL/1.5-5 SF-36 FADInfant scales Internal External TotalProblemPhysical RolePhysicalPain RoleMentalEnergy SocialfunctionPhysical abilities -.24 -.15 -.25 .13 .18 .16 .17 .15 .20 -.23Growth and Development -.40 -.38 -.48 .11 .18 .21 .13 .22 .20 -.27Bodily Pain/discomfort -.26 -.16 -.23 .10 .16 .20 .12 .11 .20 -.14Temperament and moods -.50 -.38 -.49 .17 .20 .21 .22 .31 .31 -.34General behavior -.39 -.57 -.54 .12 .14 .15 .20 .28 .16 -.33Getting along with others -.50 -.59 -.60 .20 .17 .19 .23 .28 .22 -.36General health perception -.33 -.24 -.34 .16 .16 .18 .16 .22 .22 -.20Parent scalesImpact-emotional -.50 -.53 -.59 .20 .17 .22 .30 .34 .29 -.24Impact-time -.40 -.39 -.45 .24 .25 .24 .25 .35 .33 -.28Mental health -.33 -.28 -.37 .22 .24 .26 .45 .56 .52 -.42General health – parent -.19 -.20 -.24 .40 .32 .43 .18 .38 .33 -.34Family cohesion -.21 -.26 -.28 .05 .05 .10 .19 .21 .16 -.53CBCL/1.5-5 domains: IN – internalizing syndromes; EX – externalizing syndromes; TOT – total problems score; FAD: Family Assessment DeviceTable 9: Discriminant Validity of the ITQOL: Comparison of mean (s.d.) ITQOL scales scores (s.d.) for the NICU and healthy baby samplesNICU Healthy Effect size p-valueInfant Scales (n = 952) (n = 387)Physical Abilities 92.7 (20) 97.2 (13) -.25 <.001Growth and Development 89.5 (16) 94.5 (10) -.35 <.001Bodily Pain/Discomfort 86.4 (18) 88.0 (15) -.09 .540Temperament and Moods 80.3 (12) 82.0 (10) -.15 .050General Behavior 73.4 (16) 75.6 (15) -.14 .022Getting Along with Others 73.8 (12) 76.4 (11) -.23 <.001General Health Perceptions73.1 (18) 80.9 (13) -.46 <.001Parent scalesImpact-emotional 80.5 (19) 85.1 (15) -.26 <.001Impact-time 86.3 (19) 90.0 (15) -.21 .003Mental Health 73.7 (17) 75.3 (15) -.09 .246General Health 77.3 (22) 80.5 (20) -.15 .044Family cohesion 76.7 (21) 78.8 (19) -.10 .255Scores range 0–100 – a higher score indicated more favorable quality of lifePage 8 of 11(page number not for citation purposes)These differences were all small in size (effect size .44 orsmaller).Health and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/81DiscussionIncreasingly, valid and reliable instruments are needed byresearchers and clinicians to facilitate the collection ofHRQL data in children. The preliminary results from thisstudy of children aged 3 and 4 years of age indicate thatthe ITQOL has acceptable reliability in a sample of chil-dren requiring neonatal intensive care and a sample ofhealthy peers born during the same time period. The vastmajority of items in the ITQOL were substantially linearlyrelated to their hypothesized scale, and correlations werestronger than with other scales. This finding suggestsacceptable item discriminant validity. Alpha coefficientsfor all but one scale (Physical Abilities .96) were between.80 and .90, indicating that each domain was internallyreliable. In addition, the ICCs were all satisfactory, indi-cating that parents were consistent in their ratings of theirchildren's health upon repeated assessments.The range of scores in three scales for both samples (Phys-ical Abilities; Growth and Development; Bodily Pain/Dis-comfort) was rather skewed. It is possible that the ceilingeffects may be due to the absence of younger children inour sample, or it could be because many of these children,after graduating from the NICU, are healthy. Question-naires were sent to parents of children as they turned 3 1/2 and were completed at different times (due to the lagtime for locating families that moved). Thus, our samplesincluded children ranging in age from 3 to 5 years. Thefive year olds were excluded since these data were unlikelyto be representative. Future validation research shouldlook at the full age-range from two months up to five-of infants and toddlers, it will be important to establishthat the same instrument can measure HRQL in a two-month old and a five-year old.In its present form, the main disadvantage of the ITQOLis its length. Evidence from the item-level analysis (certainitems did not satisfy scaling success criteria) suggests theremay be scope for reducing the questionnaire's length. In arecent systematic review of methods used to increaseresponse to postal surveys, the use of a short question-naire made response much more likely [44]. Since manyHRQL studies rely on postal surveys, the development ofa short-form, which is planned, may prove useful. Futurevalidation research will need to ensure a large enoughsample size across age groups to provide the opportunityto determine which items may be deleted and still retainthe psychometric properties deemed necessary.This study has certain limitations. First, we did not exploreconcurrent validity for all the instruments' domains.There was no suitable validated multidimensional meas-ure of HRQL for preschoolers at the time our study wassetup. We, therefore, chose to include a validated measureof behavior (CBCL/1.5-5), since the developmental ver-sion of the ITQOL is heavily weighted towards measuringbehavior. Had we included domain-specific measures forall domains in our study, the length of our questionnairebooklet would likely have been unacceptable to subjects.Using the CBCL/1.5-5, SF-36 and FAD, we found expectedcorrelations between similar and dissimilar constructs inthe various measures. Future research should explore con-Table 10: Discriminant validity of the ITQOL: mean ITQOL scale scores, for those with one or more health problems versus none for the NICU and healthy baby samplesNICU sample Healthy baby sampleInfant scales 1 (n = 395) 0 (n = 553) Effect size p-value 1 (n = 98) 0(n = 287) Effect size p-valuePhysical Abilities 89.6 (23) 94.9 (17) -.28 <.001 98.0 (8) 96.9 (15) .08 .121Growth and Development 85.1 (19) 92.5 (12) -.47 <.001 92.8 (10) 95.1 (9) -.26 .006Bodily Pain/Discomfort 81.6 (21) 89.7 (15) -.45 <.001 80.8 (18) 90.4 (13) -.63 <.001Temperament and Moods 78.1 (13) 81.7 (11) -.31 <.001 80.5 (11) 82.5 (9) -.24 .105General Behavior 70.5 (18) 75.4 (15) -.30 <.001 72.8 (16) 76.5 (15) -.29 .013Getting Along with Others 71.5 (12) 75.4 (11) -.33 <.001 76.0 (11) 76.7 (11) -.10 72General Health Perceptions 64.8 (19) 79.0 (15) -.80 <.001 73.1 (15) 83.8 (11) -.82 <.001Parent scalesImpact-emotional 75.4 (21) 84.0 (17) -.46 <.001 82.3 (17) 86.1 (14) -.28 .035Impact-time 81.8 (22) 89.5 (16) -.40 <.001 87.2 (18) 91.0 (14) -.28 .027Mental Health 70.2 (19) 76.2 (16) -.36 <.001 73.8 (17) 75.7 (15) -.13 .241General Health 72.1 (24) 80.9 (19) -.39 <.001 80.2 (19) 80.5 (20) -.02 .772Family cohesion 73.0 (24) 79.1 (20) -.28 <.001 76.9 (20) 79.3 (19) -.13 .261Scores range 0–100 – a higher score indicated more favorable quality of lifePage 9 of 11(page number not for citation purposes)years, as well as sub-populations (e.g., children with acuteand chronic disease). Given the rapidly changing naturecurrent and divergent validity for all the ITQOL domainsHealth and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/81and there are now validated instruments that would facil-itate this exercise.Second, although we made every effort to locate the entirecohort, we only found 81%, and only 67.9% of thesesubjects completed our study questionnaire. Thisresponse rate is within the range often obtained in a postalsurvey [45]. Many of the non-participants indicated (ver-bally or in writing) they were "too busy" to participate. Itis also likely that some questionnaires returned to usblank were from non-English speakers. Elsewhere wereport that where we had data and were able to look atresponse bias (NICU sample only), we found a few differ-ences between non-respondents and respondents chil-dren, which suggested that non-respondents had healthierbabies to begin with, and represents a potential source ofbias [28].Third, our group of healthy babies was not randomlyselected from all low-risk births in the province. However,they composed a consecutive sample of hospital deliveriesby all family physicians working within the primary careunits affiliated with 2 of the hospitals (the third hospitaldid not have such a unit).ConclusionThe results from this study indicate that the ITQOL hasgood reliability and construct validity in a sample of chil-dren who were healthy and another that had morbid con-ditions requiring neonatal intensive care. Limitationsinclude its length and possible ceiling effects. Futurevalidation work should include children of different agesand with different clinical problems.Author's contributionsAnne Klassen contributed to the study's conception anddesign; acquisition of data; analysis and interpretation ofdata; drafting of manuscript; revised the article criticallyfor important intellectual content; and gave final approvalof the version to be published.Jeanne M. Landgraf, contributed to analysis and interpre-tation of data; revised the article critically for importantintellectual content; and gave final approval of the versionto be publishedShoo Lee contributed to the study's conception anddesign, acquisition of data, analysis and interpretation ofdata; revised the article critically for important intellectualcontent and gave final approval of the version to bepublished.Morris Barer contributed to the analysis and interpreta-intellectual content; and gave final approval of the versionto be published.Parminder Raina contributed to the study's conceptionand design; the analysis and interpretation of data; revisedthe article critically for important intellectual content; andgave final approval of the version to be published.Herbert Chan contributed to the acquisition of data;revised the article critically for important intellectual con-tent; and gave final approval of the version to bepublished.Derek Matthew contributed to the acquisition of data;revised the article critically for important intellectual con-tent; and gave final approval of the version to bepublished.David Brabyn contributed to the acquisition of data;revised the article critically for important intellectual con-tent; and gave final approval of the version to bepublished.AcknowledgementsThe Hospital for Sick Children Foundation (Toronto) provided an operat-ing grant for this study. Anne Klassen was recipient of a Killam Postdoctoral Fellowship. From Canadian Institutes of Health Research, Anne Klassen holds a Senior Research Fellowship, and Parminder Raina holds a New Investigator Award. We would like to thank the families that participated in our study and the Canadian Neonatal Network.References1. Eiser C, Morse R: A review of measures of quality of life forchildren with chronic illness. Arch Dis Child 2001, 84:205-211.2. Varni JW, Seid M, Kurtin PS: PedsQL 4.0: reliability and validityof the Pediatric Quality of Life Inventory version 4.0 genericcore scales in healthy and patient populations. Med Care 2001,39:800-812.3. Eiser C, Mohay H, Morse R: The measurement of quality of lifein young children. Child Care Health Dev 2000, 26:401-414.4. Eiser C, Morse R: Can parents rate their child's health-relatedquality of life? Results of a systematic review. Qual Life Res2001, 10:347-357.5. Schmidt LJ, Garratt AM, Fitzpatrick R: Child/parent-assessed pop-ulation health outcome measures: a structured review. ChildCare Health Dev 2002, 28:227-237.6. Connolly MA, Johnson JA: Measuring Quality of life in PaediatricPatients. Pharmacoeconomics 1999, 16:605-625.7. Pal DK: Quality of life assessment in children: a review of con-ceptual and methodological issues in multidimensionalhealth status measures. J Epidemiol Community Health 1996,50:391-396.8. Spieth LE, Harris CV: Assessment of health-related quality oflife in children and adolescents: an integrative review. J PediatrPsych 1996, 21:175-193.9. Abetz L: The Infant/toddler Quality of Life Questionnaire: Conceptualframework, logic, content, and preliminary psychometric results Final Reportto Schering-Plough Laboratories and Health Technology Associates. NewEngland Medical Center; 1994. 10. Langraf JM, Abetz L, Ware JE: The Child Health Questionnaire (CHQ): AUser's Manual 2nd Printing Boston MA: HealthAct; 1999. 11. World Health Organization: Constitution of the World Health Organiza-Page 10 of 11(page number not for citation purposes)tion of data; revised the article critically for importanttion. WHO Basic Documents Geneva; 1948. Publish with BioMed Central   and  every scientist can read your work free of charge"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."Sir Paul Nurse, Cancer Research UKYour research papers will be:available free of charge to the entire biomedical communitypeer reviewed and published immediately upon acceptancecited in PubMed and archived on PubMed Central Health and Quality of Life Outcomes 2003, 1 http://www.hqlo.com/content/1/1/8112. Caplan F: The first twelve months of life: your baby's growth month bymonths Toronto, Canada: Bantam Books; 1975. 13. Fekkes M, Theunissen NCM, Brugman E, Veen S, Verrips EGH, Koop-man HM, Vogels T, Wit JM, Verloove-Vanhorick SP: Developmentand psychometric evaluation of the TAPQOL: a health-related quality of life instrument for 1–5-year-old children.Qual Life Res 2000, 9:961-972.14. Veen S, Feekes M, Koopman HM, Zwinderman KA, Brugman E, WitJM: Quality of life in preschool children born preterm. DevMed Child Neurol 2001, 43:460-465.15. Bhutta AT, Cleves MA, Casey PH, Cradock MM, Anand KJ: Cogni-tive and behavioral outcomes of school-aged children whowere born preterm: a meta-analysis. JAMA 2002, 288:728-737.16. Lorenz JM, Wooliever DE, Jetton JR, Paneth N: A quantitativereview of mortality and developmental disability inextremely premature newborns. Arch Pediatr Adolesc Med 1998,152:425-435.17. Escobar GJ, Littenberg B, Petitti DB: Outcome among survivingvery low birthweight infants: a meta-analysis. Archives of Dis-ease in Childhood 1991, 66:204-211.18. Aylward GP, Pfeiffer SI, Wright A, Verhulst SJ: Outcome studies oflow birth weight infants published in the last decade: Ametaanalysis. The Journal of Pediatrics 1989, 115:515-520.19. Saigal S: Perception of health status and quality of life ofextremely low-birth weight survivors. The consumer, theprovider, and the child. Clin Perinatol 2000, 27:403-419.20. Hack M, Flannery DJ, Schluchter M, Cartar L, Borawski E, Klein N:Outcomes in young adulthood for very-low-birth-weightinfants. NEJM 2002, 346:149-157.21. Victorian Infant Collaborative Study Group: Improved outcomeinto the 1990s for infants weighing 500–999 g at birth. TheVictorian Infant Collaborative Study Group. Arch Dis Child FetalNeonatal Ed 1997, 77:F91-94.22. Wolke D, Meyer R: Cognitive status, language attainment andprereading skills of 6-year-old very preterm children andtheir peers: the Bavarian Longitudinal Study. Dev Med ChildNeurol 1999, 41:94-109.23. Vohr BR, Wright LL, Dusick AM, Mele L, Verter J, Steichen JJ, SimonNP, Wilson DC, Broyles S, Bauer CR, Delaney-Black V, Yolton KA,Fleisher BE, Papile LA, Kaplan MD: Neurodevelopmental andfunctional outcomes of extremely low birth weight infants inthe National Institute of Child Health and Human Develop-ment Neonatal Research Network, 1993–4. Pediatrics 2000,105:1216-26.24. Walther FJ, den Ouden AL, Verloove-Vanhorick SP: Looking backin time: outcome of a national cohort of very preterm infantsborn in The Netherlands in 1983. Early Hum Dev 2000,59:175-91.25. Grunau RE, Whitfield MF, Davis C: Pattern of learning disabilitiesin children with extremely low birth weight and broadlyaverage intelligence. Arch Pediatr Adolesc Med 2002, 156:615-20.26. Siagal S, Stoskopf BL, Streiner DL, Burrow E: Physical growth andcurrent health status of infants who were of extremely lowbirth weight and controls as adolescence. Pediatrics 2001,108:407-415.27. Stjernqvist K, Svenningsen NW: Ten-year follow-up of childrenborn before 29 gestational weeks: health, cognitive develop-ment, behaviour and school achievement. Acta Paediatr 1999,88:557-62.28. Klassen AF, Lee SK, Raina P, Chan WPH, Matthew D, Brabyn D:Health status and health related quality of life in a popula-tion-based sample of NICU graduates. Pediatrics  in press.29. Klassen AF, Lee SK, Raina P, Chan H, Matthew D, Brabyn D: Relia-bility and validity of the Infant Toddler Quality of LifeQuestionnaire. Qual Life Res 2002, 11:684.30. Ware JE, Harris WJ, Gandek B, Rogers BW, Reese PR: MAP-R for Win-dows Version 1.1 Boston, MA: Health Assessment Lab; 1997. 31. Ware JE Jr, Gandek B for the IQOLA Project Group: Methods oftesting data quality, scaling assumptions, and reliability: TheIQOLA Project Approach. International Quality of Life Assessment. JClin Epidemiol 1998, 51:945-952.32. Hayashi T, Hays RD: A microcomputer program for analyzingmulti-trait-multimethod matrices. Beh Res Meth Instr Comp1987, 19:345-348.33. Campbell DT, Fiske DW: Convergent and discriminant valida-tion by the multitrait-multimethod matrix. Psychological Bulletin1959, 56:85-105.34. Howard KI, Forehand GC: A method for correcting item-totalcorrelations for the effect of relevant item inclusion. Educa-tional and Psychological Measurement 1962, 22:731-735.35. Helmstadter GC: Principles of psychological measurement New York:Appleton-Centry Crofts Inc; 1964. 36. Nunnally JC, Bernstein IH: Psychometric Theory 3rd edition. New York:McGraw-Hill; 1994. 37. Streiner DL, Norman GR: Health Measurement Scales. A Practical Guideto their Development and Use 2nd edition. Oxford, England: Oxford Uni-versity Press; 1995:64-65. 38. Fayers PM, Machin D: Quality of Life Assessment, Analysis andInterpretation Chichester, England: John Wiley & Sons, Ltd; 2000:149-50. 39. Achenbach TM, Rescorla LA: Manual for the ASEBA Preschool Forms andProfiles Burlington, VT: University of Vermont Department of Psychiatry;2000. 40. Ware JE, Snow KK, Kosinski M: SF-36 Health Survey: Manual and Inter-pretation Guide Lincoln, RI: QualityMetric Incorporated; 2000. 41. Ware JE, Kosinski M: SF-36 Physical & Mental Health Summary Scales: AManual for Users of Version 1 2nd edition. Lincoln, RI: QualityMetric;2001. 42. Cadman D, Rosenbaum P, Boyle M, Offord DR: Children withchronic illness: family and parent demographic characteris-tics and psychosocial adjustment. Pediatrics 1991, 87:884-9.43. Juniper EF, Guyatt GH, Jaeschke R: How to develop and validatea new health-related quality of life instrument. In B Quality oflife and pharmacoeconomics in clinical trials 2nd edition. Edited by:Spilker B. Philadelphia: Lippincott-Rave; 1996:49-56. 44. Edwards P, Roberts I, Clarke M, DiGuiseppi C, Pratap S, Wentz R,Kwan I: Increasing response rates to postal questionnaires:systematic review. BMJ 2002, 324:1183-1192.45. Asch DA, Jedrziewski MK, Christakis NA: Response rates to mailsurveys published in medical journals. J Clin Epidemiol 1997,50:1129-1136.yours — you keep the copyrightSubmit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.aspBioMedcentralPage 11 of 11(page number not for citation purposes)


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items