J Clinical practice in medical genetics Thegenetic testing ... · JMedGenet 1994;31:785-797...

J Med Genet 1994;31:785-797

Clinical practice in medical genetics

The genetic testing of children

Report of a Working Party of the Clinical Genetics Society (UK).Chairman: Dr Angus Clarke

Members of theWorking Party

Angus Clarke (chair)Senior Lecturer, Instituteof Medical Genetics,Cardiff CF4 4XN, UKDavid Fielding (BritishPaediatric Associationrepresentative)Consultant Paediatrician,Countess of ChesterHospital, Chester, UKLauren Kerzin-StorrarSenior GeneticAssociate, Department ofMedical Genetics, StMary's Hospital,Manchester, UKHelen Middleton-PricePrincipal Scientist,Institute of Child Health,London, UKJonathan MontgomeryLecturer in Law, TheUniversity,Southampton, UKHeather PayneSenior Clinical MedicalOfficer, Department ofCommunity ChildHealth, Cardiff, UKEmily SimonoffSenior Clinical Scientist,Institute of Psychiatry,London, UKAudrey TylerHonorary ResearchOfficer, Institute ofMedical Genetics,Cardiff, UK

Correspondence toDr Clarke.

Conclusions and recommendations(1) The predictive genetic testing of children isclearly appropriate where onset ofthe conditionregularly occurs in childhood or there are usefulmedical interventions that can be offered (forexample, diet, medication, surveillance forcomplications).

(2) In contrast, the working party believes thatpredictive testing for an adult onset disordershould generally not be undertaken if the childis healthy and there are no medical in-terventions established as useful that can beoffered in the event of a positive test result. Wewould generally advise against such testing,unless there are clear cut and unusual ar-guments in favour. This does not entail ourrecommending that families should avoid dis-cussing the issues with younger children, butrather that formal genetic testing should gen-erally wait until the "children" request suchtests for themselves, as autonomous adults.This respect for autonomy and confidentialitywould entail the deferral of testing until theperson is either adult, or is able to appreciatenot only the genetic facts of the matter but alsothe emotional and social consequences of thevarious possible test results.

In circumstances when this type of testingis being contemplated, there should be fulldiscussions both within the family and betweenparents and genetic health professionals (clin-ical geneticists or non-mecical genetic coun-sellors); the more serious the disorder, thestronger the arguments in favour of testingwould need to be.

(3) For some disorders, there is insufficientevidence to know whether a diagnosis in child-hood is helpful in the medical management ofthe possibly (not yet) affected child. Researchin these areas will be worthwhile and important.When such research is planned, however, itwill be important to incorporate a social andpsychological evaluation of the genetic testing,as well as a technical and more strictly medicalevaluation, because the results of the psy-chosocial evaluation may be critical in futureclinical judgements if the medical benefits re-main uncertain or are shown to be minor.Furthermore, the psychosocial study of testingfor these conditions, where the existence ofpossible medical benefits justifies the study ofthe testing, may throw light upon the likelypsychosocial effects of testing for other dis-

orders; hence such studies may be of moregeneral applicability.

(4) The situation with regard to testing childrenfor their carrier status for recessive disordersand balanced, familial chromosomal re-arrangements is more complex. In general, theworking party would make a presumptionagainst testing children to determine their car-rier status, where this would be of purely re-productive significance to the child in thefuture.(A) Circumstances may arise, however, inwhich the genetic testing of children could behelpful in the provision of accurate informationto other family members. Even in families withapparently balanced chromosomal trans-locations, however, we think that this occursonly occasionally. It is important that childrenin such families are not tested "as a routine",but that each situation is considered on itsmerits so that children are tested only whenthe results will contribute to the counselling ofother family members. Otherwise, if the resultswould only be of future reproductive concernto the child, then it is wiser to defer the testinguntil the child is able to understand the issuesand requests testing in person.(B) Where such (carrier) testing is, or has been,taking place, it would be useful to instituteprospective and retrospective psychosocialevaluations of the impact of the testing on thechildren and their families, so that future policycan be guided by evidence rather than con-jecture and anecdote.(C) If the testing is not to be performed inchildhood, then a certain obligation rests uponthe health care system and the family togetherto ensure that testing is offered when the childis older. While testing in childhood may allowparents and physicians to feel that they havedone their duty, this may still leave both partieswith an obligation to ensure that the testedchild is offered counselling (and possibly anupdated genetic test too) when he or she"comes of age".(5) There are additional factors to be con-sidered with a healthy but "at risk" child re-ferred for adoption, in so far as the results ofthe testing might influence decisions made onbehalf of the child. However, it should not beassumed that genetic (predictive or carrier)testing will be required before a suitable place-ment can be achieved. In each case, we would

785

on June 13, 2020 by guest. Protected by copyright.

http://jmg.bm

j.com/

J Med G

enet: first published as 10.1136/jmg.31.10.785 on 1 O

ctober 1994. Dow

nloaded from

http://jmg.bmj.com/

Clinical Genetics Society Working Party

advise discussion between the medical adviserto the adoption agency and a clinical geneticist.The important factors other than the possiblelaboratory test results need to be identified forfuture attention in advance of any test beingperformed.(6) Because some of these recommendations(1-5) are likely to diverge from the practiceand beliefs of many within the medical pro-fession, and even within clinical genetics, it isimportant that further discussion and debatetake place. We believe that the medical pro-fession should work towards a consensus onthese issues before such tests become morewidely available through commercial labor-atories, which may pay little respect to the goalofcoupling laboratory testing with the provisionof counselling and support as a package ofgenetic services. The ability of the workingparty to arrive at such a consensus, despite ourholding different views initially, suggests thatthis may be a realistic goal. It will also beimportant to extend this work to achieve abroader consensus across the professions ofnursing, social work, and the law, as well asmedicine.

BackgroundOpportunities for genetic testingTechnical advances in recent years have greatlyenhanced our ability to identify persons carry-ing faulty genes. It has been possible for someyears to identify those carrying adult polycystickidney disease by ultrasound scanning beforethe onset of symptoms; the healthy carriers ofsome haemoglobin disorders can be identifiedby routine, automated haematological analysis.However, it is now possible to identify thoseapparently healthy persons likely to developone of a large number of inherited disordersby molecular genetic technologies. It is alsopossible to identify those who carry faulty genesor chromosomes that will cause them no healthproblem, but which may result in problems fortheir future children. Much of this new testinghas become possible because the techniques ofmolecular genetics can be applied to a personof any age, and do not depend upon the at riskperson manifesting early signs of the disorder.However, the principles involved are no differ-ent for molecular genetic testing than for moretraditional diagnostic techniques.

Ethical concerns raised by genetictesting of childrenOne early concern addressed the ethics of test-ing children at risk of developing Huntington'sdisease (HD) in later (adult) life.' A consensuswas reached by those involved in the predictivetesting of those at risk of HD, who decidedthat children should not be tested.23 The know-ledge of HD gene status can be burdensomefor an adult, even when the result is low risk4and only a minority of at risk adults in factchoose to undergo such testing (about 10-15%).5-7 The right of the child to decide inadult life whether or not to be tested would be

removed if their genetic status was determinedat the request ofparents or others; such requestshave been made on many occasions.89 To whatextent is HD unique, and to what extent hasthe discussion about HD served to awaken usto similar issues raised by the testing of childrenfor other disorders?'0

Similar ethical concerns arise in the case ofother adult onset neurodegenerative disorderssuch as Alzheimer's disease and prion de-mentia." We already know that discriminationagainst persons on the basis of their geneticconstitution has been practised by insurancecompanies in Britain and North America'2 andhas even been practised by employers againsthealthy workers who happen to carry a recessivedisorder, such as sickle cell trait. ' If childrenare to have predictive tests performed, theymay lose the opportunity to obtain life in-surance before clarifying their genetic status.Harper'4'5 has suggested a moratorium on theuse of genetic testing by insurance companiesto avoid generating new problems without anopportunity for reflection and for the de-velopment of a socially accepted consensus.Two areas were regarded as potentially prob-

lematical. (1) Predictive testing of apparentlyhealthy children (that is, presymptomatic test-ing) for late onset disorders, usually of auto-somal dominant inheritance, in which clinicalmanifestations are unlikely until well into adultlife (and in which early treatment or sur-veillance for complications would not be help-ful). (2) Testing healthy children to determinetheir carrier status for inherited disorders thatwould have no implications for their ownhealth, but might affect the health of theirfuture children (usually autosomal recessive orsex linked inheritance, or balanced chro-mosomal rearrangements).

In both these areas, children's future auto-nomy, their ability to decide for themselves atsome stage whether or not to be tested, couldbe undermined. In addition, the confidentialityto which any adult being tested would be en-titled, would have been breached when theresults were disclosed to the child's parents.The potential harms caused by childhood

genetic testing might include damage to thechild's self-esteem, distortion of the family'sperceptions of the child, loss of future adultautonomy and confidentiality, discriminationagainst the child in education, employment, orinsurance, and adverse effects on the child'scapacity to form future relationships. Such test-ing breaches the policy ofproviding counsellingbefore or in parallel with the testing process,when viewed from the perspective of the childbeing tested; this could damage the professionalattempts to provide genetic services of coun-selling and testing together as a package.Although concerns about predictive and car-

rier testing in childhood have been voiced, theremay be advantages to such testing, includingthe opportunity for the child to adjust to cir-cumstances, the fostering of openness withinthe family, the resolution of parental un-certainty, and (for carrier status tests) ensuringthat testing has been offered to the whole family.These points will require consideration.

786


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


Clearly, considerations of the possible harmdone by genetic testing do not apply to alarge number of disorders, where it is to theadvantage of the child that the condition bediagnosed at the earliest possible stage. Thisapplies wherever therapy, or surveillance forpossible complications of a disorder, can use-

fully be started in childhood. Also, where an

affected person is likely to manifest the disorderduring childhood, these ethical difficulties ofpresymptomatic genetic testing in childhooddo not arise.This working party was constituted to ad-

dress these concerns.

Working party aimsThe working party was asked to: examine cur-

rent attitudes and practices; focus attention onany difficulties raised by the genetic testing ofchildren; and make appropriate re-

commendations about future practice. Theseobjectives led us to examine two questions. Towhat extent does such testing take place now?Under what circumstances is such testing jus-tified, and when might it be better deferreduntil the child is older?Even before embarking upon this project,

we acknowledged that there were likely to bedifferences of opinion and practice amongmembers of the medical profession. We foundsuch differences, which were reflected withinthe working party. On the assumption that itis desirable, is it possible for health pro-fessionals to work towards a consensus on theseissues? Some apparent differences of approachare likely to have arisen for historical reasonsrather than from any major philosophicalschism: how can such differences be resolved,so as to avoid arbitrary inconsistencies of prac-tice?

ResultsCurrent practice and attitudes inBritain: a questionnaire surveyWe wrote to molecular genetic and cytogeneticlaboratories engaged in clinical diagnostic workrequesting information about the number andnature of the tests carried out on children. Wealso asked whether requests were declined onethical grounds.We carried out a questionnaire survey of

approximately 3000 health professionals inBritain (as described in appendix 1), seekinginformation about their practice and attitudesconcerning the genetic testing of children. De-tailed results are given in appendix 1.

This survey has confirmed that there is wide-spread testing of children for genetic disorders,but most of it falls outside our remit: it eitherinvolves testing children for disorders that usu-ally become manifest during childhood (or inwhich there is a significant risk of this hap-pening), or it is otherwise clearly appropriateas part of good medical practice. By this, wemean that these tests are used in managementdecisions concerning therapy, or the sur-

veillance for complications ofgenetic disorders.Among geneticists and co-workers, the pre-

dominant view is that no requests for testing butthose initiated by a child's parents or medicaladvisers should be entertained. This view isshared by paediatricians, except that theywould be willing for adoption agencies to ini-tiate such testing. Respondents consider gen-etic tests in childhood to be justified if a usefulintervention can be made as a result (diet,treatment, surveillance for complications), orif medical management may be otherwise im-proved. There is also general agreement thatpredictive testing is best avoided among un-treatable late onset disorders such as HD, prionprotein and Alzheimer dementias, and otherneurodegenerative disorders.

In contrast to the consensus on those topics,there was a lack of consensus concerning thewisdom of testing for other late onset disordersfor which no useful interventions would beavailable; 16% of respondents considered thatpredictive testing in childhood was generallyundesirable. There was even some dis-agreement about whether or not such tests werejustified even where some clinical interventionis possible. Predictive testing, it was thoughtby some, might also be helpful in clarifying thenatural history of some conditions.There was a wide range of views on the

question of testing children for gene carrierstatus where this would have no health re-percussions for the child, as with the carrierstate for recessive disorders and chromosomalrearrangements. It is certainly common prac-tice among paediatricians to test the healthysibs of affected subjects to see if they may becarriers of, for example, cystic fibrosis or abalanced chromosomal translocation.There were also differences of attitude con-

cerning the right of parents to arrange genetictesting for their children. Geneticists and co-workers were less likely to view parental wishesalone as sufficient to justify testing than werepaediatricians and others. Only a minority ofrespondents had a firm age limit below whichthey did not test children's genetic status, andthese age limits varied widely.

Finally, it was clear from the questionnairereplies that geneticists and co-workers are morewary than paediatricians or haematologists ofusing phrases that could be regarded as pa-ternalistic, eugenicist, or directive. Thus, gen-eticists and co-workers did not support thenotion of genetic testing in childhood to in-crease "responsible" attitudes to reproduction,whereas other groups did support such sen-timents. These differences may arise fromdifferent understandings of the meaning orvalue implications of the word "responsible",or may indicate that the genetics communityhas become sensitised to such language, andno longer regard it as acceptable terminology.Alternatively, those involved in clinical geneticsmay have decided that "responsible" re-productive behaviour should be encouraged,but that genetic testing in childhood does not,or in practice is unlikely to, result in such"responsible" adult reproductive behaviour.We cannot distinguish these possible ex-planations from the replies we received.

787


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


Current practice and attitudes inBritain: a prospective studyBecause the retrospective gathering of data isdifficult, and is likely to be unreliable, we ini-tiated a prospective survey of genetic testing inchildren (under 16 years) for 12 months from1.1.92. Seventeen molecular genetic labor-atories supplied data, although it was concededthat these were incomplete even when con-siderable pains had been taken to gather in-formation on all cases. We also circulatedclinical geneticists, CGS members, and con-sultant paediatricians. These data thereforeonly provide an indication of the pattern ofchildhood testing in Britain, not an accuratemeasure of current activity, and the numberslisted are likely to be an underestimate. Noattempt was made to enquire into the testscarried out by biochemistry or haematologylaboratories, and the generation ofcarrier statusresults by newborn screening programmes (forexample, for certain haemoglobinopathies) wasnot studied.We have listed in table 5 in appendix 2

the numbers of tests reported to us as eitherperformed or deferred in 1992, grouped underthe major categories of test. We were informedabout 165 tests performed and 37 tests de-ferred. The most frequent test performed wasthat for cystic fibrosis carrier status. Forty-twotests (one quarter) were predictive, the restbeing carrier status tests. There was no evid-ence that the tests on children were usuallyperformed in later childhood; the age dis-tribution of children tested appeared to berandom, and was not concentrated in the earlyteenage years. Two exceptions to the apparentlyrandom distribution of ages were the tests forfragileX normal transmitting males (the young-est reported was at age 4; before 4 years, the testwould also identify children likely to manifestlearning problems, and would therefore bediagnostic) and for familial adenomatouspolyposis coli (where nine of the childrenwere tested in the age range 9-11 years,presumably as an aid to management, and theremaining 13 children were tested in theage range 0-6 years).

Attitudes of family and patient supportgroupsA letter and a brief set of questions were sentto 108 separate family support groups affiliatedto GIG, the Genetic Interest Group. We hopedto arouse interest in these issues, to elicit ac-counts of the experiences of families in whichtests had been performed in childhood, andto discover the attitudes of those affected bygenetic disorders and of members of their fam-ilies. We received 78 replies; some were fromwhole groups or local branches of a group,and others were from individual persons withingroups. The purpose of this enquiry was notto ballot the members of these specific diseaseoriented groups, but to obtain informationabout the range of opinions held within thesegroups by those whose lives have been touchedby the various genetic conditions.

It is possible to group the replies into very

broad categories. Forty-one replies favoured apolicy of not performing predictive or carriergenetic tests in childhood.Another group of 14 replies indicated a will-

ingness to carry out carrier tests in childhoodbut not predictive tests, although some re-servations were made by a few respondents,such as the child having the right to decide atan appropriate age.The third group of 19 replies indicated a

willingness for predictive tests to be performedin childhood. Interestingly, this group wasabout evenly divided as to whether or not testsfor carrier status should also be performed onchildren.Four replies were unclassifiable.Specific comments from respondents are

listed in appendix 3.

Legal considerationsRecent developments in child law, most im-portantly the Gillick decision in 198516 and theChildren Act 1989, have considerably alteredthe legal status of children. Whereas it waspreviously possible to regard children as littlemore than the property of their parents, thisis no longer the case. Instead, parents haveresponsibility for the care of their children,rather than rights over them, and are expectedto act in their best interests. Parental decisionsabout testing for genetic disorders should there-fore be made according to whether the childwill benefit, not in order to relieve the anxietiesof the parents. If necessary, in cases of dispute,a court can be called upon to determinewhether a particular child should be tested,although this would be rare in practice. Thecourts have suggested that children should begiven the chance to take decisions on medicalcare for themselves if it is possible to wait untilthey are able to do so without risking theirhealth. 7The law ensures that decisions about testing

are taken jointly by parents and professionals.'8In effect, parents have a veto because testingcannot be carried out without their consent,although a general consent to diagnostic testingwould be sufficient without a full explanationof the conditions being tested for. The consentof any one person with parental responsibilitywill suffice, unless a court order specificallydeals with the matter. Fathers who have neverbeen married to the mother will not normallyhave parental responsibility. However, in Eng-land and Wales, they may obtain it by applyingfor a court order giving it to them, or makinga "parental responsibility agreement" with themother and having it registered with the court.Where older children are concerned, they

will be able to consent on their own behalf ifthey have sufficient understanding of the issueto make a choice"6 (Age of Legal Capacity(Scotland) Act 1991, s 2 (4)). By statute, it ispresumed that children of 16 have this capacity(Family Law Reform Act 1969, s 8; Age ofLegal Capacity (Scotland) Act, s 1), but if achild below that age can understand the testbeing proposed and its significance then theytoo will be able to consent. Even if a child is

788


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


mature enough to be able to give a consent,recent English cases have established that par-ents can still authorise treatment, even againstthe child's wishes. 19-21 However, even if it is legalto do so, it may well be considered unethical totest for genetic status against the wishes of achild with a good understanding of the issues.Where children do not have the intellectualcapacity to agree to treatment, parental consentmust be obtained. Such consent may be givenin respect of children until they reach the ageof 18 years.Although children cannot be tested without

consent, either their own or that of their par-ents, health professionals cannot be forced tooffer care that they consider inappropriate.22Thus in cases where it would be contrary tothe child's interests to be tested, it is legitimateto refuse to do so even if the parents requestit. Nor are professionals obliged to reveal in-formation on the genetic status of a child whenthey believe it would harm the well being ofthe child. In principle, parents are entitled tosee their children's records under the Access toHealth Records Act 1990 (applying to manualrecords created after 1 November 1991) andthe Data Protection Act 1984 (for com-puterised records). However, health pro-fessionals are entitled to refuse access wherethey believe that revealing the informationwould cause serious harm to the physical ormental health of the child or others.The general principles of malpractice law

also require health professionals to give dueconsideration to the welfare ofthe child patient.They will be negligent if they fail to practise ina manner accepted as proper by a responsiblebody ofprofessional opinion in their specialty.23Thus, legal standards reflect professional ones.As long as practitioners conform to responsiblemedical practices, they will be safe from legalaction. This means that as professional ethicsdevelop, legal requirements will also change toreflect the greater understanding. Given thegeneral consensus that it is unwise to test chil-dren for HD, to do so might well be regardedas negligent except in very particular cir-cumstances. In other areas, however, both test-ing and refusing to test would probably beacceptable providing the child's position wasconsidered. However, an over rigid policy mightbe found negligent because it was not addressedto the particular child's needs and would there-fore be unacceptable to professional opinion.Withholding testing could only leave a doctoropen to litigation if it appeared to be negligenton the above test.

It should be emphasised at this point thatmost potential disagreements among familymembers, or between family members andhealth professionals, about the desirability ofgenetic testing for individual children can beavoided by sensitive counselling. This dis-cussion of the legal issues should not be takento imply that recourse to the law is likely to berequired on a regular basis.

Issues relating to adoptionThe issues relating to genetic disease and gen-etic testing in the context of adoption are com-

plex, and are discussed more extensivelyelsewhere.2425 However, we raise the questionas to whether there are particular con-siderations that might justify the genetic testingof a child being considered for adoption, re-stricting our attention (as elsewhere in thisreport) to tests of (unaffected) carrier status,and to predictive tests for adult onset disorders.

Prospective adoptive parents may have akeen interest in the genetic status of a childthat they are considering for adoption. Likemost people (if given the choice) adopters willusually want healthy children, and the adoptionagency is under an obligation to gather in-formation about the child's circumstances, in-cluding the health of the child and of membersof the family. All relevant available informationis then given to prospective adopters so thatthey can make as informed a decision as pos-sible when deciding whether to accept a child.Good practice indicates that all available in-formation should be passed on.

Adoptive parents are mostly from the 10%of involuntarily infertile couples, but some willhave made a conscious decision to limit theirfamily because of a heritable disorder. Theirpersonal experiences are likely to affect theirviews on genetic conditions in a prospectiveadoptive child.

TESTS OF CARRIER STATUSIt is our view that there will usually be noparticular justification for testing the child anyearlier than would be the case with a child stillin the birth family. Adoptive parents have thesame legitimate interests in the health of theirchildren and grandchildren as do biologicalparents. However, there are specific difficultiesthat may arise relating to carrier testing in thecontext of adoption.

First, the adoptive family may be informedabout a family history of possible relevance tothe adopted child's future reproductive plans,but may fail to remember this or to pass onthe information to the child at an appropriateage.

Secondly, the adoptive parents may focus onthe possible carrier state of their child andaccord it more emotional significance than iswarranted.

In these situations, the lack of familiaritywith the condition in question may complicatethe task of discussing the issues with the grow-ing child, leading to over- or under-emphasison the possibility of "genetic risk". Difficultiesof communication may also arise when a gen-etic disease in the birth family comes to lightafter the adopted child has been settled in thenew family. Similar problems of com-munication may arise in reverse if the adoptedchild develops a genetic disease after settlingin the adoptive family.Openness between adoptive parents and

child, and the maintenance of links betweenthe adoption agency and both the adoptive andthe birth families, may help to minimise theseproblems. However, the questions of principlerelating to carrier testing in childhood are nodifferent in the context of adoption from the

789


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


issues encountered in other families, even ifthe pathways of communication are more vul-nerable to rupture. They do not amount to areason for ensuring that the tests are carriedout preplacement.

PREDICTIVE TESTINGFor adult onset conditions where predictivegenetic testing is available, and where the ad-optive child is at risk, the question arises as towhether or not the fact of the child's adoption(prospective or established) is a justification forcarrying out a predictive test which might nototherwise be performed until the adopted childbecame an adult and chose to undergo testing.Arguments for testing include those in table

1, and the specific point that appropriate carersmay more easily be found for the child (ifpreplacement).Arguments against testing also include the

list in table 1, and some specific points relatedto adoption: that the diagnosis will label thechild and affect the (already difficult) processof identity development, that it is irrelevant tothe needs of the child for acceptance as he/sheis, and that is may cause more problems byexcluding paternity.The arguments will have to be weighed in

each case, but their force will not differ greatlyfrom the standard case of a child in the originalbirth family, unless it proves difficult to findsuitable prospective adoptive parents for a childat risk of a late onset genetic disorder becauseof the uncertainity surrounding the child's pos-sible genetic status. Then, either the decisionto put the child forward for adoption, or thedecision about genetic testing, will need to bereconsidered. In practice, this situation mayarise only infrequently, but will call for a carefulconsideration ofthe child's overall best interestswhen it does so.

In general, it would seem best, whereverpossible, to find adopters who can accept thechild as a whole, and subsequently participatein any testing that is appropriate for the childas a confirmed member of their family.

Finally, it should be noted that the status of

Table 1 Possible advantages and disadvantages ofpredictive testing in childhood

Advantages(1) Relieves anxiety about possible early signs of the

disorder.(2) Family uncertainty about the future is reduced.(3) More accurate genetic counselling becomes possible.(4) The child's attitude towards reproduction in adulthood

will be more responsible.(5) Children who might benefit from genetic counselling in

the future will be identified.(6) Practical planning for education and career, housing,

and family finances becomes possible.(7) Parental expectations of the child's behaviour become

altered.Disadvantages(1) Removes the child's right to decide whether or not to

be tested in adulthood.(2) Parental expectations of the child's future reproductive

behaviour become altered.(3) Damages the child's sense of self-esteem.(4) Generates unwarranted anxiety about possible early

signs, before any genuine manifestation of the disorder.(5) Leads to future difficulties in obtaining life insurance.(6) *Rarely leads to clarification of the genetic status of

other family members.

* This is not so much a disadvantage as a lack of advantage,but was included as a disadvantage for the purposes of thequestionnaire.

a child at risk of developing a genetic disorder(even in adult life) has some parallels with thestatus of a child with special needs. Theseinclude financial implications for the local au-thority, as a child who is at "substantial risk ofdeveloping a serious disorder, known at thetime of adoption" may require an adoptionallowance, payable until the age of 18.

SUMMARYInherited disorders have a wide range of sig-nificance in adoption because ofthe multiplicityof viewpoints that exist between the membersof the adoption triangle (adopted person, birthfamily, adoptive family) and the AdoptionAgency. When inherited disorders arise in thecontext of adoption placement decisions, theymay be highly significant and have an importantbearing on the outcome of an adoption.The management of genetic disorders in ad-

option should begin with determining the at-titudes and expectations of adopters withregard to present or future disability. Specificissues such as predictive and carrier testing arethe same as for any similarly situated personat genetic risk, with the additional con-sideration of the effect of the decision (and, ifa test is performed, the possible test results)on the child's placement. The legal situationregarding consent to testing in the adoptionprocess has not been clarified judicially.

DiscussionThe two central questions to be answered, orat least considered, are: What is good clinicalpractice with regard to predictive or carriertesting in childhood? and Who has the re-sponsibility or the right to arrange predictivetesting or carrier testing in childhood: the fam-ilies, the individual persons (the children asfuture adults), adoption agencies, or doctors?

Unfortunately, there is a dearth of firm evid-ence on which to base policies. Various blendsof ethics, anecdote, and prejudice are likely toshape people's attitudes, including our own.While there is real concern about the abrogationof a child's future autonomy as an adult, andabout the loss of confidentiality entailed inchildhood testing, we do not know whetherharm actually results from such testing (butsee below). There are concerns that the parents'attitudes to the child may change, and thatunfavourable parental expectations about achild's future mental, physical, and emotionaldevelopment could be self-fulfilling. These fac-tors could result in low self-esteem in the child,but there is little evidence that genetic testsproduce such damage in practice. This may bebecause such testing does not cause harm, orbecause such testing in the past has been limitedin scope, and the resulting harm has not cometo light because evidence of it has not beensought. We have had to arrive at our responsesto these questions in the absence of adequateempirical data, relying upon our various ex-periences and upon our concern to maintainthe generally accepted principles of medicalethics.

790


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


Predictive testing (where there is nodirect health benefit to the child). "Whatis good practice with regard to predictivetesting in childhood?"The working party has arrived at a consensusview that there should be a general presumptionagainst such testing, although we recognise thatcircumstances may arise when testing could beappropriate. We now set out the reasoning thathas led us to this common understanding.There is the precedent of the decision made

by those performing predictive testing forHD, aconsensus reached on the basis of the supposedlikely effects of such testing and not from theexperience of harmful effects (the consensuswas reached before such testing of children wasever undertaken). In families with HD, personsmay be singled out for no rational reason asbeing destined to develop the disorder. Suchfamily "preselection" of children as being likelyto develop HD is recognised as having po-tentially profound adverse effects on the emo-tional development of the unfortunatelylabelled child.26 These effects could be evenmore severe if medical evidence supported, andperhaps directed, the collusive family dynamicthat has selected one child as the "victim".

Adults undergoing predictive testing for HDhave extensive counselling, in which they arerequired to confront the effects that the test islikely to have on their whole life. Insurance(of life and health), home ownership, career,marriage, and reproductive plans may all bedeeply affected by the results of the testing.Only some 10 to 15% of at risk adults chooseto accept such testing.56 It may be difficult forparents to appreciate the reluctance to undergotesting that is felt by many adults at risk ofHD, and which their at risk child may alsocome to experience. A sensitive exploration ofthe issues will often help parents to appreciatethe limitations of their perspective, and howtheir feelings may differ from those of theirchild in the future; this may lead them toreconsider their request for testing. However,where the family has already "preselected" thechild, as being affected or unaffected, theymay continue to press for testing. In thesecircumstances, it could be particularly in-appropriate to perform the test.There is little evidence as to ill effects re-

sulting from predictive testing in childhood forHuntington's disease or for other disorders,although the fundamental principles involvedare much the same for any late onset condition,especially for other neurodegenerative dis-orders. Only limited evidence is available con-cerning the Swedish newborn screeningprogramme.for alpha-l-antitrypsin deficiency,which was discontinued because of severe ad-verse psychosocial consequences.27 However, itis clear that predictive testing in childhood forlate onset disorders, even in the context of ahigh risk family situation, can raise as manyproblems and as much anxiety as is generatedby continuing anxiety about the child's geneticstatus, and the knowledge that a child willdevelop such a disorder may, at least in somefamily contexts, cause worse problems thancontinued uncertainty. At present, we have no

means of identifying those families that wouldbe helped by having uncertainty resolved bygenetic tests and distinguishing them fromother families in which the results of testingwould be harmful. Indeed, there is not evenany agreed means of deciding whether or nota family might have been helped or harmedby such an intervention, and what time scaleshould be considered in coming to a judgementon this.Given the dearth of evidence, it is the ethical

consequences ofchildhood testing (loss ofadultautonomy and confidentiality, and the pos-sibility ofcausingharm to the developing child),together with the limited empirical evidencethat is available, that has led us to advocatethis cautious policy, erring towards a pre-sumption of non-maleficence ("primum nonnocere - the first goal is to cause no harm").

Furthermore, there are now good legalreasons for professionals to make decisions inthis area primarily on the basis of the long termbest interests ofthe child; failure to do so couldlead to professionals later being sued for actingagainst the interests of the child, even if thiswas at the request of the parents.

Despite our judgement that it is wise to avoidpredictive tests in childhood, we recognise thatexperience and information may accumulateover time, and the apparent disadvantages ofpredictive testing for HD in childhood maynot apply in practice to all other late onsetdisorders, or even to Huntington's disease inevery situation and for all time. The situationwill certainly change when effective in-terventions become available for these con-ditions.

Ifthe child stands to gain some health benefitfrom predictive testing, then it is obviouslygood practice to make the testing available.However, predictive testing carried out to sat-isfy the curiosity of a medical practitioner (ex-cept in the context of certain types of research,see below), or to relieve parental anxiety, wouldnot seem to us in general to be appropriate.There may be circumstances where such testingis justified, but we expect these circumstancesto be unusual. In our judgement, the possibleemotional harm to the child, the abrogationof their autonomy and the breaching of theirconfidentiality, will generally outweigh the pos-sible benefits that we can see.An additional caution about the use of pre-

dictive tests in childhood must be noted: genetesting is not a substitute for good clinicalassessment. This issue has been raised pre-viously in the context of HD, where the oc-currence of juvenile onset cases is infrequentbut well recognised. If an at risk child developsclinical features compatible with the early stagesof juvenile HD, then some alternative ex-planation for the suspicious clinical featuresmay still be more likely to be correct, becausethe early signs of HD can be gradual and non-specific. In such circumstances, performing apredictive genetic test (not "presymptomatic"in the usual sense of the word) may be con-sidered, but it is unlikely to be helpful.'28 Show-ing that a child has a mutation in the HD genedoes not indicate that the clinical features are

791


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


caused by HD, and may discourage the furtherdiagnostic efforts needed to arrive at the correctexplanation. If the child's symptoms are causedby another disorder, then an unfavourable pre-dictive test has been carried out on a minorwithout their consent.

Similar scenarios can arise with myotonicdystrophy and other mendelian disorders. Iden-tifying children as being at increased risk ofdeveloping one of the common multifactorialdiseases (cardiovascular disease, hypertension,diabetes melitus, cancers, ...) has the samedrawbacks as predictive testing for late onsetmonogenic diseases unless there is some dem-onstrably effective therapy, or unless the testingis carried out as part of a therapeutic trial inwhich the allocation of treatments dependsupon identifying the subjects with increasedrisk at the outset.

Carrier testingThe arguments here are less clear, and noconsensus exists among health professionals.The working party has again arrived at a con-sensus view that carrier tests in childhoodshould generally be deferred. The argumentsfor and against testing can be (and indeedalready often are) discussed openly in geneticcounselling sessions with the parents of chil-dren who might be carriers ofa genetic disorder.In most cases, this will rapidly lead to a con-sensus that testing can be deferred withoutany penalty, to allow the child eventually toparticipate in the decision as an autonomousperson. We present the contrary arguments andthe reasons for our view in the next six sections.

It may be better for a child to know about theircarrier status in childhood: this may then beaccepted as a simple matter offact, without theemotional problems that could arise from a laterdisclosureThis argument would only be valid if the onealternative to testing in childhood were to besilence on the issue within the family. We wouldlike to emphasise that a decision not to testa child for carrier status is not intended todiscourage a family from talking about thepossible carrier status of the child. In practice,talk about such matters may actually be easierwithout firm knowledge; it is not difficult toexplain to a child that the decision as to whenand whether to test is being left for them todecide once they are older. This explicit grant-ing of control to the child may actually behelpful in enhancing their self-esteem, and inallowing them to come to terms with possiblyunfavourable results once adult.We recognise that 50% of "possible" carrier

children would be shown not to be carriers bytesting in childhood, and that life might besimplified for them by childhood testing. How-ever, it is our judgement that the possible be-nefits experienced by these children must bebalanced against the possible harm done byidentifying carriers in childhood. Such harmmight be especially likely in a family wheresome children are shown to be carriers, and

others are shown not to be. There may also befamilies in which being identified in childhoodas not a carrier could be emotionally dis-advantageous.

Testing their child may decrease the level ofanxiety of the parents simply by giving them thisknowledgeBut it may well increase their anxiety if thechild is shown to be a carrier, and the testingmay need to be repeated in any case, as tech-niques and methods improve.

There is no direct evidence to indicate that carriertesting in childhood is harmfulNor is there good evidence of benefit, andsome evidence has accumulated suggesting thatknowledge of carrier status can have adverseeffects, which might be magnified if the carrier(or the non-carrier) is identified involuntarilywhen young, and is labelled in this way bythe whole family unit. We have arrived at ourjudgement on the basis of the currently avail-able information, and recognised that furtherresearch is needed to examine these issues.Problems may arise with carrier testing be-

cause of stigmatisation or the fear of stig-matisation, even in an area where the carrierfrequency is high.29 Adult carriers ofTay-Sachsdisease have been shown to view their futurehealth with less optimism than others30; it isnot clear how this would affect those carriersidentified in childhood. An earlier study hasshown that 19% of Tay-Sachs carriers wereworried about their carrier status at follow upsome years later,3' and 10% of sickle cell andthalassaemia carriers in Rochester, New Yorkare unable to state that being a carrier will notdamage their health.32 We do not know whetherchildren identified as carriers will have the samedegree of misunderstanding, or less, or more.The wider issues of social stigma have alsocaused concern even in the recent past, par-ticularly in relation to sickle cell carrier screen-ing in the USA.'333Some evidence is emerging of the distress

caused by carrier testing for cystic fibrosis inthe healthy adult sibs of affected subjects.34The limited experience with population screen-ing for cystic fibrosis in Britain indicates thatmany adults, even some of those related toknown carriers,35 decline the offer of carrierscreening tests. Uptake is influenced pre-dominantly by how the test is offered,3637 sopublic enthusiasm for the tests may be sub-stantially less than the enthusiasm of the healthor health research professionals, and the futureadult enthusiasm of the present child cannotbe taken for granted.

Carrier testing in childhood avoids urgent cariertesting in young teenage pregnancies (oftenconcealed until late pregnancy)Sexual activity, however, is not a reason forgenetic testing, and it should not be pre-supposed that such teenagers will all want car-rier testing and possible prenatal diagnosis,

792


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


calling for a decision about a possible ter-mination of pregnancy (in fact, could some

teenage pregnancies in at risk families be con-

cealed so as to avoid such tests?) A number ofalternative practical approaches can be ad-opted. First, the question of their possible gen-etic risk can be put to the adolescent withinthe family setting or by their family doctor, andreferral for genetic counselling can be arrangedif and when they choose. To push such coun-

selling or testing upon an adolescent (or any-one) is unacceptable, and it is likely to generatemore problems than it "solves". Furthermore,adolescents who may be carriers of a geneticdisorder, like other adolescents, should be givenproper information about human reproductionand contraception.

Testing in childhood ensures that the testing is atleast performed, and may absolve the doctor ofhis responsibility in this regardHowever, imparting the information to the fam-ily does not ensure that the correct informationwill be given to the child in an appropriatefashion and at an appropriate age. If parentsin practice find it difficult to impart the factsof sexual intercourse and human reproductionto their children, will they find it easier todiscuss genetic disease? Anecdotes reported tous suggest not.When carrier tests are carried out in child-

hood, we believe that there is still a re-

sponsibility on the medical profession to ensurethat the future adult is offered genetic coun-

selling at an appropriate age. This is both toensure that the adults have the opportunity todiscuss the issues relevant to themselves, andbecause the facts given to the parents, in con-

nection with the particular disease prognosisor diagnostic testing, may no longer apply be-cause of advances in medical knowledge.

Carrying out carrier tests in childhood, then,does not absolve the doctor of responsibilityfor ensuring that the future adult is offeredgenetic counselling at an appropriate age. Whilethe initial scenario may work well in some

families, it must be remembered that manyfamilies do not conform to the stereotype nuc-lear family structure. In a social context withparental fluidity and serial step parenting, andbecause of the emotional responses of guilt,blame, responsiblity, and denial in relation totheir genetic disease in many families, it isnot possible to rely upon the effectiveness andsensitivity ofintergenerational communication.In addition, to set up testing of children in thisway runs counter to a core principle of clinicalgenetic practice, by separating the testing fromthe counselling by a gap of years. Finally, thecarrier testing methods may change, so thatfurther analysis may be appropriate in any case

once the child is adult and beginning to plana family.

If testing is not performed in childhood, how may

a doctor fuifil his responsibility to offer testing toall relevant persons in the family?There are several possible ways ofensuring that

a child is offered testing at an appropriate age.These include: (1) a clear transfer of re-sponsibility from doctor to family, so that theonus of offering genetic counselling and testingto the child (once adolescent or adult) lies withthem. This may be the only reasonable optionif the family is planning to emigrate; (2) a cleartransfer of responsibility to the child's familypractitioner, who can prompt a genetics referralwhen the child is old enough to be directlyinterested in the issues; (3) continuing in-volvement with the family in the form of anactive genetic register, with regular, long termfollow up to keep in touch with family memberseven when they move area.

In general, we would favour the third option,of involvement on an active follow up register.However, it is clear that this is the most ex-pensive option, because the resources requiredto establish and maintain such registers areconsiderable. We must also emphasise that test-ing in childhood does not in itself remove thephysician's responsibility to ensure that thechild is offered counselling when older.

SummaryThe above arguments tend towards the viewthat carrier testing for genetic disease in child-hood may not be the most appropriate ap-proach, at least if practical alternatives canbe devised and established. Inadvertent carriertesting (from research or from prenatal diag-noses that indicate an unaffected, but carrierfetus) should be avoided wherever possible.Research is needed to ascertain the psycho-social consequences, the adverse effects or thebenefits, of carrier testing in childhood.

Unsought information in research andclinical practiceIt is possible for unsought information to begenerated about a family in the course of re-search, for which consent has not been ob-tained, and which is unlikely to be of use inthe research. This is particularly likely to occurwhen blood samples have been collected on allavailable family members as part of a pre-liminary gene localisation (linkage) study, whenthese samples are stored in a research unit, andwhen tests are carried out on the stored researchsamples some years later once a specific testhas been developed. This amounts to genetictesting without counselling or consent, and islikely to arise when samples have been collectedindiscriminately from all available family mem-bers, and when the samples are tested wholesalein a manner for which consent was not obtainedfrom the family (because the test was then notavailable). This can generate difficult practicaland ethical problems, but these can often beavoided by careful advance planning, attentionto detail in research consent procedures, anda clear separation of research work from serviceresults and clinical records. This area has beendiscussed by Harper,42 and guidelines havebeen proposed. A specific effort on the partof clinical and laboratory staff will often berequired to ensure that information is not gen-

793


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


erated on family members which has not beenrequested by them; once such unsought in-formation is available, it may be difficult toprevent its being passed to the family, perhapsinadvertently and possibly resulting in seriousfamily distress.

Predictive testing carried out as part of aresearch study into the consequences of earlydiagnosis may well be legitimate, but attentionshould be paid to the guidelines of the MedicalResearch Council Working Party on Researchon Children43 and the British Paediatric As-sociation's Guidelines for the Ethical Conductof Medical Research Involving Children."Unsought information may also result from

the genetic testing of children in standard clin-ical practice, and we report one scenario thatillustrates some of the problems that can result.In a family in which some members carrieda balanced chromosomal rearrangement, thechildren were tested cytogenetically when veryyoung, and at least one child was found tocarry the rearrangement in balanced form. Thefamily subsequently attended for counsellingbecause they did not know how to give theirteenage daughter her (positive) tests results; itwas decided to offer her a test once she ex-pressed interest, as if she had never been testedin early childhood, rather than tell her that shehad in fact been tested before and that herparents had known her test result for years.

Who has the right or responsibility toarrange for the genetic testing ofchildren?PARENTSThere is no clear consensus on this issue amongBritish professionals. One legal view fromNorth America would hold that parents arelikely to have the right to the full disclosure ofany genetic information about the child towhich they want access,45 although this has notyet been tested in the courts. It has also beensuggested that geneticists may have an ob-ligation to carry out such testing, and it hasbeen emphasised that each child's case mustbe considered individually; no general policywould be defensible.46 The legal situation inthis country is different, and to our minds moresatisfactory, in that any decisions must be madeon the basis of the best interests of the child,although again no blanket policy would bedefensible. Given that the genetic test resultswill have no medical management implicationsfor the child, and that the uptake of carrier andlate onset predictive tests by adults in Britainis low, the arguments in favour of parentalrights to demand this information are weak,although the parents may have the right todemand alternative means ofensuring that test-ing and counselling are offered when the childis older.

ADOPTION AGENCIESAdoption agencies are recognised by pae-diatricians as having a legitimate interest in thegenetic status of prospective adoptive children.This view is reasonable, in so far as the result

of such testing will influence the likelihood ofa child being accepted by suitable adoptiveparents. However, it may be better for a childto be adopted by parents whose willingness toadopt is not dependent upon the results ofpredictive or carrier genetic tests.

MEDICAL PRACTITIONERSWhere testing is to the possible medical ad-vantage of the child, it is clearly the duty ofmedical practitioners to ensure that it is carriedout. Where testing may be of interest to thefuture adult, for health reasons or to permitinformed reproductive decision making, theoffer of counselling (and possible testing)should be made once the person is matureor in early adult life; this may require theestablishment of an active genetic register.With respect to carrier testing, or to pre-

dictive testing for a late onset disorder, themedical profession needs not assume any ob-ligation to initiate such tests. When such testingof children is requested by others, however,doctors have a right to refuse to carry it out ifthey consider that it may cause harm to thechild or is not in the child's best interests.

We thank all those who responded to our various questionnairesand letters, professionals and lay societies and support groups.We thank Miss Jean Dunscombe for her hard work and or-ganisational ability in ensuring the dispatch of so many doc-uments and for coordinating the work of the working party. Wethank the professional bodies who distributed questionnaires totheir members in their own mailings. We also thank Michaeland Sue Aldred for their assistance in designing the ques-tionnaires, the data preparation staff of the University of WalesCollege of Medicine for imprinting data onto floppy disks, andIain Fenton for his assistance in extracting this infornationonce again. We would also like to thank Professor MartinBobrow for his diligent reading of several drafts of the reportand his very constructive comments upon them. We are gratefulto the Marie Stopes Research Fund, administered by the GaltonInstitute, for its financial support of the questionnaire studyincorporated into this report.

1 Craufurd D, Harris R. Ethics of predictive testing for Hun-tington's chorea: the need for more information. BMJ1986;293:249-5 1.

2 World Federation of Neurology. Research Committee Re-search Group Ethical issues policy statement on Hun-tington's disease molecular genetics predictive test. JNeurol Sci 1989;94:327-32 and J Med Genet 1990;27:34-8.

3 Bloch M, Hayden MR. Opinion. Predictive testing forHuuntington disease in childhood: challenges and im-plications. Am J Hum Genet 1990;46:1-4.

4 Huggins M, Bloch M, Wiggins S, et al. Predictive testingfor Huntington disease in Canada: adverse effects andunexpected results in those receiving a decreased risk. AmJMed Genet 1992;42:508-15.

5 Craufurd D, Dodge A, Kerzin-Storrar L, Harris R. Uptakeofpresymptomatic testing for Huntington's disease. Lancet1989;ii:603-5.

6 TylerA, Morris M, Lazarou L, Meredith L, MyringJ, HarperPS. Presymptomatic testing for Huntington's disease inWales 1987-1990. Br J Psychiatry 1992;61;481-9.

7 Bloch M, Adam S, Wiggins S, Huggins M, Hayden MR.Predictive testing for Huntington disease in Canada: theexperience of those receiving an increased risk. Am J MedGenet 1992;42:499-507.

8 Morris M, Tyler A, Harper PS. Adoption and genetic pre-diction for Huntington's disease. Lancet 1988;ii: 1069-70.

9 Morris M, Tyler A, Lazarou L, Meredith L, Harper PS.Problems in genetic prediction for Huntington's disease.Lancet 1989;ii:601-3.

10 Harper PS, Clarke A. Should we test children for "adult"genetic diseases? Lancet 1990;335: 1205-6.

11 Collinge J, Poulter M, Davis MB, et al. Presymptomaticdetection or exclusion of prion protein gene defects infamilies with inherited prion diseases. Am J Hum Genet1991;49:1351-4.

12 Billings PR, Kohn MA, de Cuevas M, Beckwith J, AlperJS, Natowicz MR. Discrimination as a consequence ofgenetic testing. Am J Hum Genet 1992;30:476-82.

13 Bowman JE. Invited editorial. Prenatal screening for hae-moglobinopathies. Am J Hum Genet 1991;48:433-8.

14 Harper PS. Genetic testing and insurance. Jf R Coil PhysLond 1992;26:184-7.

794


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


15 Harper PS. Insurance and genetic testing. Lancet 1993;341:224-7.

16 Gillick v W. Norfolk & Wisbech HA (1985). 3 ALL ER402.

17 Re D (1976). Fam 185.18 Re J (1990). 3 All ER 930.19 Re R (1991). 4 All ER 177.20 Re W (1992). 4 All ER 627.21 Montgomery J. Consent to health care for children. Child

Law 1993;5:177-24.22 Re J (1992). The Times, 12 June.23 Bolam v Friern HMC (1957). 2 All ER 118.24 Payne H. Genetic disease and adoption. In preparation.25 Turnpenny PD, Simpson SA, McWhinnie AM. Adoption,

genetic disease and DNA. Arch Dis Child 1993;69: 411-13.26 Kessler S. Invited essay on the pscyhological aspects of

genetic counselling. V. Preselection: a family coping strat-egy in Huntington disease. Am Med Genet 1988;31:617-21.

27 Thelin T, McNeil TF, Aspergren-Jansson E, Sveger T.Psychological consequences of neonatal screening foralpha-1-antitrypsin deficiency. Acta Paediatr Scand 1985;74:787-93.

28 Craufurd D, Donnai D, Kerzin-Storrar L, Osborn M. Test-ing of children for "adult" genetic diseases. Lancet 1990;335:1406.

29 Stamatoyannopoulos G. Problems of screening and coun-selling in the haemoglobinopathies. In: Motulsky AG,Ebling FJB, eds. Birth Defects: Proceedings of the FourthInternational Conference. Amsterdam: Excerpta Medica,1974.

30 Marteau TM, van Duijn M, Ellis I. Effects of geneticscreening on perceptions of health: a pilot study. MedGenet 1992;29:24-6.

31 Zeesman S, Clow CL, Cartier L, Scriver CR. A private viewof heterozygosity: eight-year follow-up study on carriersof the Tay-Sachs gene detected by high school screeningin Montreal. Am Med Genet 1984;18:769-78.

32 Loader S, Sutera CJ, Segelman SG, Kozyra A, Rowley PT.Prenatal hemoglobinopathy screening. IV. Follow-up ofwomen at risk for a child with a clinically significanthemoglobinopathy. Am Hum Genet 1991;49:1292-9.

33 Kenen RH, Schmidt RM. Stigmatisation of carrier status:social implications of heterozygote screening programmes.Am Pub Health 1978;68:1116-20.

34 Fanos JH, Johnson JP. Barriers to carrier testingfor CF siblings.American Society of Human Genetics Meeting, New Or-leans, October 1993, Session 22, Abstract 51.

35 Trembath RC, Green J, McMahon C, Malcolm S, PembreyME. Camer testing for CFTR gene mutations in relativesidentified through a cystic fibrosis prenatal testing service.Response rates and determinants of uptake. Clinical GeneticsSociety, Belfast, 1991.

36 Watson EK, Mayall E, Chapple J, et al. Screening for carriersof cystic fibrosis through primary health care services.BMJ 1991;303:504-7.

37 Bekker H, Modell M, Denniss G, et al. Uptake of cysticfibrosis testing in primary care: supply push or demandpull? BMJ 1993;306:1584-6.

38 Harper PS, Tyler A, Smith S, Jones P, Newcombe RG,McBroom V. A genetic register for Huntington's choreain south Wales. Med Genet 1982;19:241-5.

39 Read AP, Kerzin-StorrarL, Mountford RC, Elles RG, HarrisR. A register-based system for gene tracking in Duchennemuscular dystrophy. Med Genet 1986;24:84-7.

40 Norman AM, Rogers C, Sibert JR, Harper PS. Duchennemuscular dystrophy in Wales: a 15 year study, 1971 to1986. Med Genet 1989;26:560-4.

41 Burn J, Church W, Chapman PD, Gunn A, Delhanty J,Roberts DF. A regional register for familial adenomatouspolyposis: congenital hypertrophy of the retinal pigmentepithelium as a means of carrier detection. Med Genet1989;26:207A.

42 Harper PS. Research samples from families with geneticdiseases: a proposed code of conduct. BMJ 1993;306:1391-4.

Table 2 Clinical questionnaire

Specialties of responding clinicians Responses Questionnaires sent

Consultant clinical geneticists 32 (58%) 55Genetic co-workers 17 (23%) 73Consultant paediatricians 337 (34%) 990Consultant paediatrician members of the British

Paediatric AssociationOther members of CGS 14 470

(Clinical Genetics Society)"10%" group of paediatricians 50 (42%) 118

(of all consultant members of British PaediatricAssociation)

"FAP" surgeons 14 (32%) 44Paediatric surgeons 14 (16%) 87Ophthalmologists 7 (64%) 11Haematologists (45 adult, 6 paediatric) 51Other physicians 30

The proportion of respondents for each professional group is not precise because of some overlapof memberships between different groups (for example, CGS and the Genetic Nurses and SocialWorkers Association; CGS and the British Paediatric Association).The low response rate from "other" CGS members appears to have resulted from the reluctanceof non-consultant clinical geneticists to reply to a questionnaire that required them to list "their"cases in which childhood testing for genetic disorders was proposed; such cases, it was felt, wouldbe best reported by the appropriate consultant so as to avoid a duplication of reports.

43 Medical Research Council. The ethical conduct of research onchildren. Report of working party. London: MRC, 1991.

44 British Paediatric Association. Guidelines for the ethical con-duct of medical research involving children. BPA Ethics Ad-visory Committee, 1992.

45 Pelias MZ. Duty to disclose in medical genetics: a legalperspective. Am J Med Genet 199 1;39:347-54.

46 Sharpe NF. Presymptomatic testing for Huntington disease:is there a duty to test those under the age of eighteenyears? Am J Med Genet 1993;46:250-3.

Appendix 1. Results of the questionnaire study

Molecular genetics laboratoriesWe sent questionnaires to 35 of the laboratories listed in theClinical Molecular Genetics Society handbook, targeting thoseones that we thought most likely to be involved in diagnosticwork and including all Regional NHS laboratories. We receivedreplies from 16 laboratories, of which half did not carry outtesting that would fall within our definition as being potentiallyproblematical.Some laboratories carry out the genetic testing of young

children to determine their disease status, and generate carrierstatus information incidentally. This information is not thenusually withheld from the family. Several laboratories carry outpredictive testing for disorders such as familial adenomatouspolyposis, myotonic dystrophy, and adult polycystic kidneydisease. Testing is also carried out for some X linked disordersthat can affect females (OCT deficiency, fragile X mental re-tardation). In these conditions there is a distinction to be drawnbetween testing for a child who may be clinically affected andone in whom evidence of carrier status is being sought. Six ofthe laboratories regularly test young children specifically todetermine their carrier status for recessive conditions such ascystic fibrosis or sex linked muscular dystrophy (Duchenne orBecker). Presymptomatic diagnosis of Leber's optic atrophy hasbeen carried out in a young girl. Two centres specifically statedthat it was their policy not to carry out any such predictive orcarrier tests on children.

Cytogenetics laboratoriesTen of the 23 regional or subregional cytogenetics laboratoriesto which questionnaires were sent provided us with detailedinformation about their testing of children, with particularemphasis on the determination of their carrier status for familialchromosomal rearrangements. Several other laboratories re-sponded with general statements and observations. the picturethat emerges is that most laboratories examine samples fromseveral children each year in order to determine whether or notthey carry a familial chromosomal rearrangement. However,several reasons were given to account for some of these teststhat do not necessarily apply in all situations. First, some testsare carried out at birth to confirm a prenatal test result. Second,the results of the test may be used to assist with the interpretationof test results in other family members. However, there isprobably still a substantial number of children tested wherethese considerations do not apply, and where the questionsraised by our working party do need to be addressed.

It was put to us that cytogenetics laboratories generally acceptsamples referred from paediatricians, geneticists, or other cli-nicians in good fairth, without questioning the ethical or clinicaljudgements involved; it is primarily with the clinicians that theseissues need to be explored. The same argument could be usedby molecular laboratory staff. Whether or not this argumentis an adequate response to these issues may deserve widerdiscussion.

Clinicians and genetic co-workers (nurses, counsellors,etc)The questionnaire was circulated to members of several pro-fessional groups, including members of the Clinical GeneticsSociety (550, including some paediatricians and genetic co-workers), members of the Genetic Nurses and Social WorkersAssociation (73), consultant members of the British PaediatricAssociation (990), members of the British Paediatric CardiacAssociation (100), members of the British Society for Haem-atology (780), the Society for Endocrinology (1000), the BritishPaediatric Neurology Association (20 non-BPA members), As-sociation of British Neurologists (500), the British Associationof Paediatric Surgeons (87), general surgeons with an interestin familial adenomatous polyposis (44), and to selected oph-thalmologists (11). Respondents are shown in table 2. Becauseof the importance of obtaining responses from a representativegroup of paediatricians, and because of the low response ratefrom paediatricians as a whole, a second copy of the ques-tionnaire was distributed to the non-respondents among a ran-domly selected group of 10% of consultant paediatricians.

PREDICTIVE TESTINGOnly 184 of the 512 respondents (36%) had received requeststo carry out predictive testing of 902 children. In addition, somerespondents were involved in newbom screening programmesfor haemoglobinopathies and other disorders. Many of the testsdiscussed were potentially of health benefit to the child (forexample, hyperlipidaemias, medium chain acyl-CoA de-hydrogenase deficiency, family cancer syndromes, haemo-globinopathies, cystic fibrosis, alpha-l-antitrypsin deficiency)or were for disorders likely to become manifest in childhood

795


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


(adrenoleucodystrophy, Becker or Duchenne muscular dys-trophy). Requests for predictive testing were declined or de-ferred by a number of persons, principally for neurodegenerativedisorders (HD, Charcot-Marie-Tooth disease, myotonic dys-trophy) and also for retinitis pigmentosa (RP), autosomal dom-inant polycystic kidney disease (APKD), and familialadenomatous polyposis coli (FAP). Some requests for the pre-dictive testing of children for myotonic dystrophy, RP, APKD,and hypertrophic obstructive cardiomyopathy (HOCM) wereaccepted and carried out.Only a few of these requests originated from outside the

nuclear family, typically from adoption agencies (preplacement).

CARRIER TESTINGA greater proportion of respondents (41%) received requeststo determine the carrier status of healthy children for inherited

Table 3 Differences in attitudes between professionalgroups

A "Do you agree that the following have the right to requestthat genetic testing (predictive or carrier status) be performedon a child, even if the result will have no direct health benefitfor the child?"

Affirmative replies(% of respondents in each professional groups)

Parents Extended Adoptionfamily agencies

Medicalpractitioner

Geneticists 57* 6 28 21(n= 46)Co-workers 29 0 12 12(n= 17)Paediatricians 75 10 55 42(n= 337)"10%" Paeds 72 16 48 40(n= 50)Haematologists 76 14 37 61(n= 51)All (n=512) 71 11 47 41

* Of the 27 geneticists (59%) who thought that parents shouldbe able to arrange for genetic testing of their child(ren), 18 (39%of all the geneticists) would permit this for certain disorders only.B "It is up to the family to decide whether or not theirchild(ren) should be tested for genetic disorders (forpredictive or carrier tests)".

Agree (%)

Geneticists 46Co-workers 35Paediatricians 79"10% Paeds" 78Haematologists 84All 74

Table 4 Responses to supplementary questionnaire

"Would you test a 5 year old child whose parents wanted to know the genetic status for thefollowing disorder(s)?"

Disorders

Predictive testingAdult polycystic kidney diseaseBecker muscular dystrophyCharcot-Marie-Tooth diseaseFacioscapulohumeral muscular dystrophyHuntington's diseaseHyperlipidaemiasHypertrophic obstructive cardiomyopathyLeber's optic atrophyMarfan syndromeMultiple endocrine neoplasiaMyotonic dystrophyNeurofibromatosis IPolyposis coliPrion protein dementiaRetinitis pigmentosaSpinal muscular atrophyvon Hippel-Lindau diseaseUnaffected carrier status testingAlpha-l-antitrypsin deficiencyCystic fibrosisHaematological diseaseSpinal muscular atrophyTay-Sachs diseaseAdrenoleucodystrophyBecker muscular dystrophyChoroideraemiaDuchenne muscular dystrophyG6PD deficiencyHaemophilia (A and B)Immune defectsNorrie's diseaseRetinitis pigmentosa(Balanced) chromosomal translocation

Geneticists(n= 49)

Yes No

Paediatmcians(n = 260)

Yes No

27 17 198 2930 15 189 3811 28 154 468 30 156 502 47 100 8940 7 223 1637 8 201 1320 18 152 4043 4 199 2432 9 160 2219 26 182 4230 13 207 2928 16 214 140 42 66 4820 22 177 3516 23 163 4536 8 143 22

19 27 132 9222 27 170 7422 28 155 7417 30 130 9820 30 142 8812 32 132 7411 33 136 8211 32 75 8114 32 163 7513 32 139 9013 33 159 7412 31 139 7912 30 81 7711 32 130 7827 20 146 83

disorders. As in predictive testing, most of these professionalswere clinical geneticists, paediatricians, or haematologists. Thebulk of such testing involved familial chromosomal re-arrangements, Duchenne and Becker dystrophies, cystic fibrosis,and the haemoglobinopathies; almost all of these tests werecarried out, but some respondents who discouraged such testsdid not report the number of tests requested. Carrier statustests for X linked visual disorders, coagulopathies, and fragileX mental retardation were also carried out regularly.A number of requests for carrier testing were declined or

deferred, for example, for CF, DMD, chromosome trans-locations, etc, although only a minority of respondents (8%)had a firm age limit below which they did not carry out suchtesting, and the age limit varied from 1 to 18 years. Of the 276respondents who had received no requests, 28 discourage suchtests in childhood, 123 refer all such requests elsewhere, and130 had never been asked. Only a few requests were receivedfrom outside the family. The numbers of children on whomtesting was performed or requested was reported as 1706 chil-dren annually. Although this is probably only a proportion ofthe tests carried out, it is clear that a substantial number ofchildren are being tested for their carrier status.

AttitudesPREDICTIVE TESTSPredictive tests in childhood were thought to be justified undersome circumstances by 66% of respondents on the grounds ofa health benefit to the subject, while 16% of the respondentsthought that such testing was not justified. Those who con-sidered such testing to be helpful gave many examples ofdisorders where this could be so. These disorders included thehyperlipidaemias, FAP, Wilms' tumour, retinoblastoma, andMarfan syndrome. Some respondents stated that they wouldonly carry out such testing if they had available a useful in-tervention (for example, treatment or surveillance for com-plications). A minority of respondents wished to offer testingfor disorders where there are no clearly established usefulinterventions, including HD, HOCM, Leber's optic atrophy,and myotonic dystrophy.

Attitudes to predictive genetic testing in childhood wereexplored further in very general terms. Respondents were askedto indicate the strength of their agreement or disagreement withthe set of statements about genetic testing (table 1). The sevenpoints given as possible advantages of predictive testing inchildhood were all regarded positively by a large majority ofrespondents, although not so many supported the notion thatthe child's attitude to reproduction as an adult would be "moreresponsible" as a result of testing in childhood. On this question,most geneticists and all fieldworkers expressed disagreement,whereas a majority of paediatricians and others supported thisjudgement. A majority thought that testing in childhood wouldalso permit more accurate genetic counselling of other familymembers.

RIGHT TO REQUEST GENETIC TESTINGThe next question, conceming who had the right to request thegenetic testing of a child, showed marked differences in attitudebetween the members of different professional groups (table 3).Because the response rate to the questionnaire was poor (only58% of consultant geneticists and 34% of consultant pae-diatricians), it is not clear how representative the respondentsare of their professional groups. However, the attitudes of thegroup of 10% of consultant paediatricians who were given asecond mailing if they had failed to respond to the first, werevery similar to those of the other responding paediatricians. Itis therefore reasonable to interpret the questionnaire results asif the respondents' views are representative of their professionalgroups'.A majority of genetics co-workers stated that not even parents

should have the right to request the genetic testing of theirchild. While clinical geneticists were divided on this issue,they thought that this right should apply to testing for certaindisorders only. In contrast, a majority of respondents overall(364 = 71%), and of paediatricians and haematologists in par-ticular, thought that parents should have the right to requestthe genetic testing of their child. This was confirmed by theresponses to a separate question (table 3), suggesting that the"right to request" was generally interpreted as a right to havethe test performed. A very substantial majority of all groupsthought that social services (unless having parental re-sponsibility), legal authorities, the extended family, and theeducation authorities should not have this right. Some gen-eralisations concerning differences between professional groupscan be made. Most haematologists who responded thought thatmedical practitioners but not adoption agencies should havethis right. Paediatricians were divided on the question ofdoctors'rights and of adoption agencies' rights, but were more in favourof adoption agencies' rights than were other groups; clinicalgeneticists and fieldworkers would restrict the rights of boththese groups of professionals.

POTENTIAL DISADVANTAGES OF TESTING IN CHILDHOODTurning to the potential disadvantages of carrying out tests inchildhood, a majority also agreed with the statements thatsuch tests limited the child's future autonomy, could lead todifficulties obtaining life insurance, and could have an adverseeffect upon parental expectations of the child's future re-productive behaviour (table 1). Additional potential dis-advantages that were put forward included future difficulty inobtaining training and employment, and the disturbance of the

796


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/


parent-child relationship leading to impaired "bonding" andthe possible rejection of the child by the parent.

Despite the lack of firm evidence on which to base theirviews, and the broad range of opinion found within eachprofessional group, we did find significant differences ofopinionbetween professional groups. Paediatricians were evenly dividedas to whether or not the child's sense of self-esteem could bedamaged by testing, whereas a small majority of geneticists anda large majority of fieldworkers thought that this could bean important disadvantage of such testing. A majority of allrespondents also considered that a diagnosis in childhood couldresult in excessive anxiety about possible early signs of thedisorder, just as a majority of respondents also considered thatit could lead to the resolution of such anxiety.The questions we had identified as issues to be considered

were recognised as potential issues, and additional distinct issueswere not identified. There was some evidence that respondentswere more confident of recognising potential advantages oftesting than they were ofrecognising the potential disadvantages(fewer responses to the statements of potential advantages oftesting indicated neutrality, and substantially more responsesindicated a strength of feeling).One set of respondents thought that children might adjust to

unwelcome genetic information more readily in middle child-hood than in adolescence. Others expounded the opposite view,that such a process could be harmful or misleading. A largemajority of all groups thought that children should be consultedon these questions once they are sufficiently mature.There were several reasons for considering the process as

potentially harmful. The disorder may be variable, so that theprevious experience in the family merely causes confusion.Children may be singled out to their own detriment or that oftheir sibs. Healthy children may be "medicalised" when theyare still very impressionable. The test results may be givenan excessive emotional weight in some families, generatingdisproportionate levels of anxiety. Some respondents thoughtthat, by performing the test only at the specific request of the(now adult) persons, one is allowing them a degree of controlthat may be helpful in their coming to terms with unwelcomeinformation.

Supplementary questionnaireSome months later a second questionnaire was distributed tomembers of the Clinical Genetics Society and to consultantmembers of the British Paediatric Association. This askedwhether or not the respondent would be willing to arrangegenetic testing of a healthy 5 year old child for a number ofdifferent disorders (predictive tests for some dominant and sexlinked disorders; carrier status tests for some autosomal recessiveand sex linked disorders and for balanced chromosomal trans-locations). It was hoped that differences in the acceptability of

Appendix 2

Table S Prospective study of testing in molecular geneticslaboratories 1992

Genetics tests Performed Age range Deferred

Alpha-l-antitrypsindeficiency(carrier test) 1 9y 0

BCR 13 Infancy-14y 2BMD predictive 4 1-12y 0CF carrier 68 Infancy-15y 9DMD carrier 5 8-15y 4Fabry's carrier 1 3y 0FAP 22 Infancy-1 Iy 0FSH dystrophy 1 1Oy 0Fragile X carrier 10 4-14y IFragile X NTM 5 4-14y 0Haemoglobinopathy

carrier 10 1-14y 5Haemophilia A carrier 8 3-14y 0HMSN I 1 Birth 1HOCM predictive 0 1 (age 4)Huntington's disease 0 2 (ages 5,

14)Kennedy's disease

(predictive) 0 1 (infant)Myotonic dystrophy 13 Infancy-14y 7Spinal muscularatrophy (SMA),intermediate 1 3mth 0SMA3 1 1Oy 0

Carrier and predictive tests for adrenoleucodystrophy, Wilson'sdisease, and a few other conditions were deferred for a varietyof reasons, sometimes including technical as well as ethical andcounselling factors.

BCR=balanced chromosomal rearrangementBMD = Becker muscular dystrophyCF =cystic fibrosisDMD = Duchenne muscular dystrophyFAP = familial adenomatous polyposis coliFSH = facioscapulohumeralHMSN = hereditary motor sensory neuropathyNTM = normal transmitting male

797

testing for different disorders might allow inferences to be drawnabout the factors that influence professional attitudes.

INTERPRETATION OF SUPPLEMENTARY QUESTIONNAIRE RESULTSResponses were received from 260 paediatricians, 49 geneticists,seven co-workers, and eight haematologists -324 in all. Thesame trends appear in these data as in the longer, originalquestionnaire: paediatricians were generally much more willingfor genetic tests to be performed in childhood than eithergeneticists or co-workers. Fewer firm opinions were expressedconcerning the more uncommon disorders. There was little tosuggest that the availability of an early treatment or surveillancefor complications for a disorder made any difference to thereadiness to test for it.Respondents fell into two broad groups, that in which most

tests were approved, and that in which most would be refused.The most important factor influencing responses seems to havebeen the respondents' attitudes to such testing in general, andthe nature of the various disorders seems to have had relativelylittle influence. Most respondents answered similarly for all thediseases for which they stated a view (although the proporitionof questions answered with a "Don't know" did vary betweenpersons). Only with Huntington's disease and prion dementiawas there a suggestion ofa widespread unwillingness to carry outpredictive testing on a young child. Even with these conditions, itwas striking that many paediatricians expressed a readiness tocarry out presymptomatic predictive testing. The few paediatrichaematologists surveyed expressed uncertainty about testing formost disorders, and were divided in their willingness to testchildren to deternine their carrier status for haematologicaldisorders. The further analysis of these responses, and theexploration of the reasons for this pattern of results, will needto be studied as a future piece of research, and is beyond thescope of this report.

Appendix 3 Attitudes offamily and patient supportComments on the replies from the group which favoured apolicy of not performing predictive or carrier tests in childhoodincluded remarks such as "at least wait until child able to giveinformed consent", and "once of an age to be sexually active".This group included all six replies from adults with spinalmuscular atrophy. Those opposed to the principle of predictiveor carrier testing in childhood generally favoured the right ofhealth professionals to make decisions about whether or not toproceed with tests requested by parents. Those in favour ofcarrying out such tests generally opposed this type ofprofessionalcontrol, regarding it as excessively paternalistic.Remarks written in opposition to the principle of genetic

testing in childhood included:"Testing will destroy the innocence of childhood.""Good counselling - yes; testing - no.""If there is some preventive measure ... then tests should be

done, but otherwise no.""I don't think it should be done just for the parents' peace

of mind....""We feel it is not the parents' right, but the 'child's' right in

adulthood. Doctors should be able to refuse the tests in orderto protect the child."

"I wanted my son tested to see if he was a carrier of PKU...but after long discussions my husband and I decided it wouldserve no purpose other than labelling him... Other more seriousdisorders would need a lot more thought but I think the samedecisions should be made."

"It is important, I think, for a child to grow up knowing thatthere is a chance that he/she is a carrier for something-but notuntil she/he is at the older age when they are thinking ofpartnersand producing offspring should it be necessary to do anytesting..."

"Predictive testing would be just another example ofprojectedunethical genetic cleansing."

"If as a result of trying to wipe out the disease, there werefewer remaining, then society may become less tolerant."Remarks written by those in general support of childhood

testing included:"A parent's job is hard enough. If there is the possibility that

their child will become seriously ill at whatever age, then parentsshould have the opportunity. to either allay their fears and thestress on the family or alternatively to make provision for thesupport - moral, emotional, financial - of their child. A parentshould have the right to ask for testing but (should) not be"pressured" in any way. The decision to test or not should bewith the parents."

"Doctors should have no say in the matter.""I think it (predictive testing) is essential, and that persons

with a potential disorder in later life should be made aware sothat they can plan for their futures, and be prepared (for) whatto expect."

"This issue (carrier testing in childhood) is simpler (thanpredictive testing). I feel that testing of children in this casewould be not only right, but vital. An informed decision onreproduction could be made."There were two further comments that provide interesting

perspectives:"Decisions on carrier testing should be left to the child when

they are old enough... GPs should ensure that such childrenfully understand the risks before the decision to start a family.Support and counselling are needed and cannot be left toparents."

"As with all matters relevant to a child, the fundamentalcriterion has to be what is in the best interests of the child.This may be a decision capable of being made by the parentsor doctors, but not necessarily either."


http://jmg.bm

j.com/

J Med G


ctober 1994. Dow

nloaded from

http://jmg.bmj.com/

J Clinical practice in medical genetics Thegenetic testing ... · JMedGenet 1994;31:785-797...

Documents

Transcript of J Clinical practice in medical genetics Thegenetic testing ... · JMedGenet 1994;31:785-797...