ASSESSING QUALITY OF LIFE IN CHILDREN and ADOLESCENTS E.W. GERHARZ et al.

Current approaches to assessing the quality of life in children and adolescents

E.W. GERHARZ, C. EISER* and C.R.J. WOODHOUSE Department of Urology, Julius Maximilians University Medical School, Wrzburg, Germany, *Department of Psychology, University of Shefeld, and The Institute of Urology & Nephrology, Royal Free and University College Medical School, London, UK

SUMMARY Urologists are well aware of the importance of the quality of life (QoL) in determining the success of their treatments. The public are increasingly aware of this aspect of medicine. The advent of fetal ultrasonography has made knowledge of QoL in long-term survivors of congenital anomalies essential in counselling about pregnancy termination. It is becoming mandatory that clinical trials involving adults include an assessment of QoL. For children, measuring QoL has been restricted to lifethreatening conditions such as cancer or transplantation medicine. Measuring QoL is difcult in children and adolescents, and this is reected in the few suitable instruments available. The development of sexuality contributes a further difculty as many anxieties typical of those with genital anomalies may be common to all teenagers. Several instruments rely on the opinions of a proxy (parent or carer) but self-assessment by the child is preferable where possible. The need for child self-assessment is supported by nding little concordance between child and proxy assessments. While measuring QoL is challenging, we provide a substantial case for greater consideration of QoL in paediatric urology.

As a reection of this the concept of multidimensional health-related QoL is increasingly important, arguably becoming the most widely researched subject in the social and clinical sciences. However, what seems to be self-evident in the adult world is a rather recent development in children [2]. Although QoL is important for children with chronic diseases or congenital disorders, interest in children's QoL did not gain momentum until the 1980s [3]. The increasing attention given to measuring children's QoL can be attributed primarily to the changing epidemiology of childhood disease, from acute to chronic and from incurable to curable or palliative. Where it is possible to manage but not cure a disease, it is important to determine how far treatment and disease compromise the child's QoL. In this way, informed judgements can be made about whether or not treatment is appropriate, and critically, where there is a choice, which might be the best option for the child [4]. Measures of QoL are used in routine audit work or evaluating interventions, and certainly need to be considered when making end of life decisions, i.e. when it is appropriate to withhold treatment because the anticipated QoL is so poor. They may assist decisions about rationing of resources. The advent of fetal ultrasonography has facilitated the identication of a range of major but not life-threatening anomalies for which parents require counselling before deciding whether or not to terminate the pregnancy. At other times it may simply be useful to understand the child's perception of illness and its effect on daily life. These considerations apply to a broad range of issues in paediatric urology. Whereas the success of the numerous reconstructive procedures for urogenital abnormalities has been detailed meticulously in terms of mortality and physical morbidity, little research has explored the psychological impact and QoL associated with surgical treatment.

WHY QOL MUST BE MEASURED Children with chronic illness have a higher incidence of psychiatric, psychosocial, educational, and emotional problems than have healthy children. The effects are variable not only with the visibility or invisibility of the condition, but with a whole host of additional external and internal factors [5,6]. Explanations for these ndings have centred on generic variables which affect children regardless of the specic nature of their disease. These include school absence, which potentially compromises school achievement and social relationships; reduced opportunities to take part in sports and social activities; and changes in family relationships, especially where parents are responsible for home-based care. In addition to these generic factors, QoL is compromised depending on the characteristics of the specic condition. Thus, in diabetes, the need for regular blood tests, insulin injections and vigilance about diet are considered major obstacles to achieving a good QoL. In cystic brosis, diet is again important, as is daily physiotherapy. In the context of urological disorders there is a complex interplay between QoL and the control of evacuation. Children with urological conditions may be house-bound, as nurseries may not accept incontinent children. They may be unable to play games because of urinary leaks. Physically disabled children may not be able to take part in sports, except in similarly disabled groups. Hidden problems such as incontinence may have a more damaging effect than overt ones; incontinence may be a bar to swimming because of the fear of contamination. This forces a child to admit to a problem that might otherwise have remained private. Fecal incontinence is a particular difculty because of the smell. A child who needs special permission to leave class because of urinary urgency is singled out as different. The expression of QoL also depends on the child's age. In the passage to adult life, children must take control of their lives. In the chronically sick child, a very difcult moment

INTRODUCTION In these times of disappearing medical paternalism, in which increasingly selfcondent patients search the Internet before consulting their doctor, it seems almost redundant to state that decision-making in modern medicine must include consideration of the patient's perspective. At the same time, the traditional sanctity of life ideology governing medicine is under attack. The result is a growing acceptance that doctors must not only keep people alive, but also pay attention to the quality of life (QoL) of survivors [1]. There is a consensus that traditional endpoints such as the reduction in symptoms and improved survival are insufcient for evaluating medical outcomes. The inclusion of more holistic patientorientated variables is necessary. 150

2 0 0 3 B J U I N T E R N A T I O N A L | 9 1 , 1 5 0 1 5 4 | doi:10.1046/j.1464-4096.2003.04001.x

ASSESSING QUALITY OF LIFE IN CHILDREN AND ADOLESCENTS

comes when the child wishes to make decisions that previously have been made by the parents. As a part of this process the individual must be made aware of the nature of the disease. As sexuality becomes more important during adolescence, exstrophy patients and others with genital anomalies may experience particular challenges to achieving a completely fullled life [79]. Very badly deformed adolescents might wish that they had never been born. A pregnant mother might opt for termination of pregnancy if the fetus had spina bida, but the baby may be as loved as any other if she did elect to continue with the pregnancy. Only 8% of young women with spina bida would accept a termination if they were pregnant with a similarly affected fetus [10]. Scientic and anecdotal evidence shows that selective termination, based on a perception of longterm QoL, is reducing the incidence of babies with spina bida, exstrophy and prune-belly syndrome [11]. Even in cases of PUV there is a trend towards termination of pregnancy, even though there is no formal evidence that the QoL of patients with this condition is impaired. In contrast, there are many examples of handicapped adolescents leading heroic lives. A particular difculty arises if the adolescent makes a decision that the parents perceive to be wrong. In rare cases where treatment, especially surgery, is needed, a real conict can develop. Occasional cases have been referred for judicial review. However, more commonly the problem concerns a QoL issue, such as the need to become continent or develop sexuality. Here an objective QoL measure is of no help; if the adolescent does not perceive the need for treatment, there is no point in pursuing the matter. Conversely, it is wrong for parents to pressurize their offspring against treatment, especially in relation to sexual therapy. Adolescents with spina bida are sometimes over-protected by their families and such practice reduces the likelihood of independence. However, even with a cosseted upbringing, active programmes to teach independence in early adult life can be successful and can improve QoL [12,13]. Groups of adolescents have little tolerance for nonconformists. A year in the life of an adolescent is a long time and it is a common observation that lives can be devastated and triumphant in a short space of time, often as a result of changes in peer pressure and

behaviour. The obvious consequence is that comparison with objective criteria is almost impossible. In adulthood every effort must be made to establish independence from parents and close family. Education must aim at nancial, physical and emotional freedom, if at all possible. By the end of adolescence there must be a plan for adult life that takes into account the needs of the patient and of the family. There is one area in which adolescents and adults have the same desire, i.e. the freedom from lavatories, by having complete continence. Incontinence of urine, feces and atus are severe burdens, easily measurable, which limit the QoL in all ages. In adolescents with anorectal agenesis, poor continence correlates strongly with mental health problems and psychosocial disorders [14].

Obtaining data directly from the children may therefore be more desirable. However, there are circumstances where children are too young or ill to answer for themselves. All measures should therefore include provision for proxy ratings. In these situations, a critical question is how far ratings between child and proxy are similar. In practice, children and parents do not necessarily share the same perspective or experience of an event. Indeed, children may not want to share all their thoughts and feelings with their parents. Furthermore, parents have their own concerns which may colour their perception of the child's QoL. Ideally, all measures should include versions for self-completion by children as well as parallel forms for proxies. Questions should be re-phrased away from the search for concordance and towards understanding the circumstances in which parents and children agree (or disagree) about the child's QoL.

THE CHALLENGE OF MEASUREMENT Assessing QoL in children poses unique problems [4]; children do not share adult views about the cause and treatment of illness. They may interpret questions differently, and adopt a different time perspective about the course of a disease. In addition, their abilities to use rating scales, understand the language, and generally complete lengthy questionnaires of the type used in adult work, may be compromised by age and cognitive development. Understanding QoL in children will therefore not be achieved by scaling down ideas developed through work with adults. To match the features unique to childhood and adolescence a joint WHO/International Association for Child Psychology and Psychiatry Working Party [15] proposed in 1993 that new measures should be childcentred, use subjective self-reporting where possible, be age-related or at least developmentally appropriate, have a generic core and specic modules, and emphasise health-enhancing aspects of QoL. The measures should also be cross-culturally comparable and include the child's concept of health and illness (which may depend on developmental stage and cultural background). As children are often regarded as unreliable respondents, early attempts to rate children's QoL were based on mothers' reports. However, children and parents do not necessarily share similar views about the impact of illness. THE VALUE OF ADULT QOL MEASURES IN CHILDREN The issue of QoL has received considerable attention in adult work. Given this, it is tempting to take an adult measure, change the wording a little, and then use it for work with children. There are several problems with this approach. Adult measures tend to be too long for children; rating scales may be too complex, requiring sensitivity judgements beyond the child's abilities. In addition, it cannot be assumed that the domains of QoL considered appropriate for adults are necessarily the same for children. Central to children's QoL are issues about the ability to full age-related activities. For adolescents, any comprehensive assessment of QoL must include issues of body image, autonomy and aspirations for the future. These issues are rarely included in adult measures. For younger children, issues about attachment to the family, relationships with peers and siblings and developing cognitive competence are likely to be important.

ASSESSMENT OF QOL IN CHILDREN AND ADOLESCENTS: THE CURRENT STATUS An update of a comprehensive literature search [16] based on work between 1980 and 1998, using key electronic databases, illustrates that much needs to be done to improve QoL assessment in children. While 151

2003 BJU INTERNATIONAL

E . W. G E R H A R Z ET AL.

over 30 000 publications relevant to QoL in medicine were identied, only 12% were related to children, mostly involving theoretical and conceptual work. Over 90% of empirical studies used parents or clinic staff as reporters of the child's QoL. The conditions most frequently studied (oncology and transplantation medicine) tended to be those with signicant mortality rates and where treatment requires a high cost and levels of care. Other chronic conditions that received attention included asthma, epilepsy, diabetes and rheumatism. QoL was more often assessed for older children (1318 years) than for those aged 612 years. Many studies covered a wide range of ages with no stratication of the results. Interviews, questionnaires, clinical and social indicators have been used to assess QoL. A recent report commissioned by the NHS Health Technology Assessment Programme [4] was conducted to determine the scope and quality of measures currently available for work with sick children. A broad search strategy was devised to provide a comprehensive list of publications that have used direct or proxy measures for QoL for children and adolescents. Of 137 articles identied, 43 involved the development of a new measure, 79 reported further development and application of measures, and 15 assessed QoL on the basis of a battery approach, i.e. using several measures. Of the 43 measures retrieved, 19 were generic and 24 were disease-specic. Only three generic measures [1719] and two disease-specic measures [20,21] satised basic psychometric criteria. Many measures were identied for asthma (four), cancer (ve) and epilepsy (four). Sixteen measures allowed for completion by children and parent or caregiver. Measures were developed in the USA (18), the UK (eight), Canada (eight) and the Netherlands (two). Single measures were developed in Germany, Israel, Spain, Sweden, Norway, and Finland. The measures were described as QoL (30), health status (eight), functional status (two), perception of illness, life satisfaction, and quality of well-being (one each). Among the generic measures, one was targeted at children aged 05 years, six at children across a broad age range, two at children in middle childhood ( 611 years), four at adolescents, and four at children from 8 years to late adolescence; in addition, two measures were based on adult measures. 152

DISCUSSION While it is important to recognize the methodological difculties inherent in measuring QoL, over-emphasis on these problems must not be used as an excuse to focus on hard or clinical outcomes. In the nal analysis, the goals must be to improve quality as well as quantity of life. For those considering measuring QoL in their clinical or research work, decisions about how, what and where need to be made. Generic measures are assumed to be preferable when decisions need to be made about allocating resources from public health perspectives. In contrast, disease-specic measures have merit when assessing the impact of a change in treatment, or when evaluating outcomes in clinical trials. Among disease-specic measures, asthma, cancer and epilepsy have received most attention. For children with many other conditions it is only possible to rate QoL using a generic measure. Disease-specic measures are inappropriate where a child has more than one condition. Furthermore, the low incidence of some conditions (e.g. intersex, bladder exstrophy) may preclude the development of diseasespecic measures. Wallander et al. [22] argue that disease-specic measures are little more than indictors of how the individual perceives the illness. There is therefore a need to understand the relation between generic and disease-specic QoL. The development of a core generic instrument supplemented by disease-specic modules may be the solution. This allows for direct comparison between illness samples, and additional information to be obtained about specic disease. It is also imperative to accept that both child and proxy ratings have value. The question is to clarify how differences in the perception of QoL arise between the child and proxy. This also applies to ratings of QoL made by teachers and other proxies. Parents may be inuenced by other children they know, their expectations and hopes for the child, additional life stresses, and their own mental health. Based on ndings that children and parents differ in their understanding of illness and treatment, there is a widely endorsed view that children should rate their own QoL wherever possible. Despite this, many measures rely exclusively on parental

reporting. A few measures provide parallel forms for completion by both child and parent. These may be the measures of choice where children are well and able to rate their QoL. Measures are typically targeted at children across a broad age range, with very few measures available for those aged <8 years. There is much scope for creative measures that allow an insight into how treatments affect the QoL of the younger child. The question of norms must also be considered; in controlled studies there is always a difculty in selecting the control group. There seems to be little point in taking age-matched normal boys who would, in fact, be just the same as the local norms. On the other hand, what group will have had anything like the same medical experience as boys with epispadias or hypospadias? In recent publications on hypospadias, three control groups have been used, i.e. boys who have had appendectomies, hernia repairs and circumcisions [23,24]. These studies show that adolescents can have very real problems (although perhaps not to the extent that could be imagined), but give no clear picture of the degree to which QoL is affected. It is important to compare the adolescent with a peer group and not arbitrarily chosen norms. A child may well be continent as dened by a 3-h dry interval, because childhood revolves around disciplines based on school lessons, breaks and meals. The life of adolescents has very few dened landmarks, and continence means not leaking and not making visits to the lavatory at intervals different from the social circle, perhaps well considered as the number of drinks that can be taken before leakage or a visit to the lavatory. It has been suggested that studies of QoL after rehabilitation are so lacking in rigour, with only 13% involving controlled trials, that no conclusions can be drawn [25]. However, work with otherwise normal adolescents shows a preoccupation with sex and therefore the question of proper controls becomes even more important. Uncontrolled studies of adolescents with both hypospadias and epispadias show a high level of dissatisfaction with genitalia and sexuality. But even then, when absolutes are measured, e.g. the age of sexual debut and the number of regular sexual partnerships, the differences from local norms may be minimal. Translating a QoL instrument for use in different countries may appear a cheap

2003 BJU INTERNATIONAL

ASSESSING QUALITY OF LIFE IN CHILDREN AND ADOLESCENTS

and satisfactory option, but in fact requires extensive work to establish true comparability, given cultural differences in the meaning of illness, relationship between parents and children, and organization of healthcare services. Lack of attention to cross-cultural issues must be addressed before blindly adopting norms based on other cultural, social or illness groups. Thus, we conclude that there are methodological limitations in all the measures currently available. However, we feel strongly that measuring QoL should not be abandoned. The renement of existing measures and the future development of more sophisticated measures depends crucially on experience gained in using those that are currently available. This interface between constructing measures and clinical application requires multidisciplinary collaboration; the potential for mutual benet is enormous. An improved understanding of the psychosocial and QoL aspects of urogenital abnormalities and their treatment may provide a scientic basis for an interdisciplinary concept in the long-term care of such patients. In 1999, Schober [26] stated that no published study has adequately addressed the QoL of intersexuals. In a systematic overview of outcome studies in patients with bladder exstrophy, only 10 of 1260 articles (0.8%) focused on the mental or psychosocial aspects, and most of these had serious methodological deciencies [7]. Many studies state that outcomes were unsatisfactory without stating how satisfaction was measured. There is no doubt that much needs to be done to improve the quality of QoL measures, and hence the status of this work in clinical practice and research. However, recognition of QoL has done much to raise the prole of children's views about the treatment and organization of care. A recognition of the shortcomings of currently available measures must not be used as a reason to ignore QoL issues. At the least, attention to QoL has emphasized the need to consider the outcomes of paediatric medicine in terms of the whole child rather than a narrow range of clinical indicators.

made to use QoL measures in clinical research, to gain experience that will guide the development of a second generation of more sophisticated instruments. QoL researchers and clinicians need to cooperate to face this exciting challenge.

12

ACKNOWLEDGEMENTS Christine Eiser is supported by Cancer Research-UK. The review of QoL measures for children was commissioned by the NHS Health Technology Assessment (HTA) Programme. Elmar Gerharz was supported by the Incontinence Research Trust. The views and opinions expressed in this paper reect those of the authors and do not necessarily reect those of the HTA.

13

14

15

16 REFERENCES Browne JP. Health-related quality-of-life studies in urology. conceptual and methodological considerations. World J Urol 1999; 17: 1938 2 Gerharz EW. Quality of life research in children: Fashion or future? Dial Ped Urol 1997; 20: 12 3 Eiser C, Morse R. The measurement of quality of life in children: past and future perspectives. Dev Behav Ped 2001; 22: 24856 4 Eiser C, Morse R. Quality-of-life measures in chronic diseases of childhood. Health Technol Assess 2001; 5: 1147 5 Beck AL, Nethercut GE, Crittenden MR, Hewins J. Visibility of handicap, selfconcept and social maturity among young adult survivors of end stage renal disease. J Dev Behav Paed 1986; 7: 936 6 Pless IB, Nolan T. Revision, replication and neglect research on maladjustment in chronic illness. J Child Psych Psychiatr 1991; 32: 34765 7 Diseth TH, Emblem R, Schultz A. Mental health, psychosocial functioning, and quality of life in patients with bladder exstrophy and epispadias an overview. World J Urol 1999; 17: 23948 8 Berg R, Berg G, Svensson J. Penile malformation and mental health. Act Psychiatr Scand 1982; 66: 398416 9 Bracka AA. A long term view of hypospadias. Br J Plastic Surg 1989; 42: 2515 10 Boydell LR, Nevin NC. Is antenatal selection for spina bida possible? BMJ 1990; 301: 608 11 Chan A, Robertson EF, Haan EA. 1

17

18

19

20

21

22

23

CONCLUSIONS All currently available measures of QoL have limitations, but it is essential that attempts be

24

Prevalence of neural tube defects in South Australia, 196691: effectiveness and impact of prenatal diagnosis. BMJ 1993; 307: 703 Wolman C, Brasco DE. Factors inuencing self-esteem and self-consciousness in adolescents with spina bida. J Adolesc Health 1994; 15: 5438 Reiner S, Goldman T. A programme for the enhancement of autonomy in young adults with physical disabilities. Int J Rehab Res 1999; 22: 714 Diseth TH, Emblem R. Somatic function, mental health and psychosocial adjustment of adolescents with anorectal anomalies. J Ped Surg 1996; 31: 63843 World Health Organization. Measurement of quality of life in children. Division of Mental Health, Genf, 1993 Bullinger M, Ravens-Sieberer U. Health related quality of life assessment in children: a review of the literature. Revue Europeenne Psychologie Appliquee 1995; 45: 24554 Varni JW, Seid M, Rode CA. The PedsQL: measurement model for the pediatric quality of life inventory. Med Care 1999; 37: 12639 Feeny D, Furlong W, Barr RD, Torrance GW, Rosenbaum P, Weitzman S. A comprehensive multiattribute system for classifying the health status of survivors of childhood cancer. J Clin Oncol 1992; 10: 9238 Landgraf JM, Abetz LN. Functional status and well-being of children representing three cultural groups: Initial self-reports using the CHQ-CF87. Psychol Health 1997; 12: 83954 Varni JW, Katz ER, Quigging DJL, Friedman-Bender A. The Pediatric Cancer Quality of Life Inventory-32 (PQCL-32) I. Reliability and validity. Cancer, 1998; 82: 118496 Juniper EF, Guyatt GH, Feeny DH, Ferrie PJ, Grifth LE, Townsend M. Measuring quality of life in children with asthma. Qual Life Res, 1996; 5: 3546 Wallander JL, Schmitt M, Koot HM. Quality of life measurement in children and adolescents: issues, instruments, and applications. J Clin Psychol 2001; 57: 571 85 Mureau MAM, Slijper FME, Nijman RJM, van der Meulen JC, Verhulst FC, Koos Slob A. Psychosexual adjustment of children and adolescents after different types of hypospadias repair: a norm related study. J Urol 1995; 154: 19027 Aho MO, Tammela OKT, Somppi EMJ, Tammela TLJ. A long term comparative 153

2003 BJU INTERNATIONAL

E . W. G E R H A R Z ET AL.

follow up study of voiding, sexuality and satisfaction among men operated for hypospadias and phimosis during childhood. Eur Urol 2000; 37: 95101 25 Evans RL, Dingus CM, Haselkorn JK. Living with a disability: a synthesis and critique of the literature on quality of life, 1985

89. Psych Reports 1993; 72: 2738 26 Schober JM. Quality-of-life studies in patients with ambiguous genitalia. World J Urol 1999; 17: 24952 Correspondence: Dr E. Gerharz, Department of Urology, Julius Maximilians University

Medical School, Wrzburg, Germany. e-mail: elmar.gerharz@mail.uniwuerzburg.de Abbreviations: QoL, quality of life.

154

2003 BJU INTERNATIONAL