Identifying homozygous sickle cell disease when ... - SAGE Journals

3 downloads 0 Views 182KB Size Report
Aug 20, 2004 - Black River) at least once during the same 27-year period. A number ... Jamaican cohort study of sickle cell disease.20 The Jamaican. Ministry ...
04 Hambleton cpp 5/11/04 12:03 PM Page 175

175

ORIGINAL PAPER

Identifying homozygous sickle cell disease when neonatal screening is not available: a clinic-based observational study IR Hambleton and KJJ Wierenga ................................................................................................... J Med Screen 2004;11:175–179

Objectives: Life-threatening clinical complications can occur in the first years of life in people with

See end of article for authors’ affiliations

...............

Correspondence to: Ian R Hambleton, Sickle cell Unit, University of the West Indies, Mona, Kingston, Jamaica; [email protected] du.jm Accepted for publication 20 August 2004

...............

homozygous sickle cell disease. There is consensus that a clinical care programme comanaged by a specialist clinic should follow early-life disease identification. In a setting without widespread neonatal screening for this disease, we predict the percentage of affected births that enrol in specialist clinics during childhood, and the percentage that enrol early enough to benefit from penicillin prophylaxis (which is offered until five years of age). Setting: A retrospective study of enrolment between 1973 and 1999 at three clinics in Jamaica, the country’s only referral centres for sickle cell disease. Results: Among enrolees not screened at birth, observed enrolment by age five was 10.1% (95% confidence interval [CI] 5.7–16.7%) among 1974 births, which is predicted to rise to 35.7% (95% CI 35.0–36.4%) among 1999 births. Observed enrolment by 18 years of age was 45.9% (95% CI 35.7–58.2%) among 1974 births, which is predicted to peak at 61.9% (95% CI 60.5–63.2%) among 1984 births, and fall to 48.9% (95% CI 40.9–56.9%) among 1999 births. Median age at enrolment was 10.5 years (95% CI 10.0–11.3). Conclusions: Based on 1999 estimates, almost 65% of children affected by homozygous sickle cell disease not identified at birth will not benefit from important early-life clinical intervention, and half will not enrol for specialised care by their 18th birthday. Among patients that enrol, half do so in adolescence when management is less focused on preventive care.

INTRODUCTION

H

omozygous sickle cell disease is an autosomal recessive disorder caused by the inheritance of the sickle βglobin allele from both parents.1 It is the most common of all haemoglobin disorders, with 230,000 new cases per year in sub-Saharan Africa,2 a region that accounts for 70% of worldwide prevalence.3 This lifelong condition is characterised by the predominance of sickle haemoglobin, leading to chronic haemolytic anaemia and recurrent episodes of vaso-occlusion. Patients with homozygous sickle cell disease have special health care needs. Preventive measures are especially important in early childhood,4,5 while the frequent and unpredictable nature of complications extends these health care needs beyond adolescence. There is consensus that the cornerstone of managing homozygous sickle cell disease patients should be comprehensive clinical care, comanaged by a specialist clinic.2,6–9 Simple, reliable and inexpensive screening procedures to detect homozygous sickle cell disease at birth are available10 and there is evidence from observational studies that newborn screening programmes can lead to substantial reductions in morbidity and mortality if linked to subsequent and appropriate patient care and parental education. A mortality rate of 1.8% in the first 10 years of life among neonatally screened patients versus 8% among clinically ascertained controls has been reported.11 Based on the proven efficacy of penicillin prophylaxis in the prevention of pneumococcal infections in young children with homozygous sickle cell disease,5 screening for all neonates in the US irrespective of ethnicity has been

www.jmedscreen.com

advocated by the National Institutes of Health since 1987.7 In the UK, reports commissioned by the National Health Service advise universal neonatal screening in regions with >0.5–1.8 cases of sickle cell disease per 1000 live births, with targeted screening elsewhere.8,9 The World Health Organisation working group on haemoglobinopathies recommended national sickle cell programmes including neonatal screening in countries where the disease constitutes a ‘common public health problem’ (defined as >0.5 affected patients per 1000 live births in countries where the infant mortality rate has fallen to 60% at 18 years old. We note the incomplete information due to the younger age ranges in recent years. The most striking example of this bias is for enrolment by 18 years of age, where we have no information among people born after 1982. Without observed information, we use a prediction model to obtain estimates of enrolment.

Predicting patients that will eventually enrol Observing patients that have already enrolled We present the proportion of expected unscreened deliveries that have enrolled in the clinic by five years of age in Table 1, and by other selected ages in Table 2 and Figure 2. Enrolment proportions increase with age and over time and reach values >10% at 1 year old, >40% at 5 years old, >50%

www.jmedscreen.com

We present predicted unscreened clinic coverage in homozygous sickle cell disease in Table 2 for selected years of birth between 1974 and 1999. By December 1979 (when the patients born in 1974 were 5 years old), 10.1% (95% confidence interval [CI] 5.7–16.7%) of unscreened affected infants had enrolled in Journal of Medical Screening

2004

Volume 11

Number 4

04 Hambleton cpp 5/11/04 12:03 PM Page 178

178

Hambleton and Wierenga

Observed and predicted proportion of SS disease births who were not screened at birth and who were identified on arrival at a clinic at selected ages and in selected years between 1974 and 1979 in Jamaica

Table 2

Age Year of birth

1 year (95% CI)

5 years (95% CI)

10 years (95% CI)

18 years (95% CI)

Observed 1974 1979 1984 1989 1994 1999

2.7 (0.7–6.9) 7.7 (3.8–13.8) 4.7 (2.0–9.2) 12.4 (7.8–18.8) 11.9 (7.3–18.1) 2.5 (0.5–7.4)

10.1 28.0 34.3 37.3 47.5 –

(5.7–16.7) (20.0–38.1) (26.1–44.2) (28.8–47.4) (37.9–58.8)

24.3 37.8 50.0 52.0 – –

(17.0–33.7) (28.4–49.3) (40.0–61.7) (41.9–63.7)

45.9 (35.7–58.2) 53.8 (42.5–67.3) – – – –

Predicted 1974 1979 1984 1989 1994 1999

1.4 (1.1–1.7) 4.5 (4.4–4.6) 7.1 (7.0–7.1) 8.8 (8.8–8.9) 9.9 (9.8–10.0) 10.3 (10.2–10.5)

13.4 26.6 34.1 37.1 37.3 35.7

(12.6–14.3) (26.2–27.0) (33.8–34.4) (36.8–37.3) (37.0–37.6) (35.0–36.4)

26.8 43.4 50.5 51.4 49.2 45.1

(25.4–28.2) (42.7–44.1) (50.0–51.1) (50.8–51.9) (48.3–50.0) (43.0–47.1)

42.5 57.8 61.9 59.5 54.8 48.9

(40.4–44.7) (56.6–59.1) (60.5–63.2) (57.3–61.7) (51.2–58.4) (40.9–56.9)

CI=confidence interval; SS=homozygous sickle cell disease

the clinics. This 5-year enrolment level rose to around 40% through most of the 1980s and early 1990s, and is predicted to fall slightly to 35.7% (95% CI 35.0–36.4%) among 1999 births. By December 1992 (when the patients born in 1974 were 18 years old), 45.9% (95% CI 35.7–58.2) of unscreened infants had enrolled at the clinics. This 18-year enrolment level is predicted to rise to around 60% among patients born in the 1980s, before falling to 48.9% (95% CI 40.9–56.9%) among 1999 births. Uncertainty associated with the estimation of patients is summarised in the associated 95% confidence interval, and this uncertainty is large among patients born in recent years.

Age at clinic enrolment We present the age at enrolment among unscreened enrolees in Figure 3. Enrolment was slightly earlier among boys (hazard ratio 1.12, 95% CI 1.03–1.23, p=0.01). This gender difference was mainly seen in early childhood and enrolment proportions were roughly equivalent by adolescence: the median enrolment age among boys was 10.5 years (95% CI 9.7–11.4 years), while the median enrolment age among girls was 10.6 years (95% CI 9.8–11.6 years).

DISCUSSION Evidence about the benefits of early disease detection for homozygous sickle cell disease patients comes from a

Boys

Girls

Proportion enrolled

1.0

0.8

0.5

0.3

0.0 0

4

12

8

16

Age at enrolment

Figure 3 Age at enrolment to clinic among boys and girls with homozygous sickle cell disease who were not identified at birth. Journal of Medical Screening

2004

Volume 11

Number 4

randomised clinical trial5 supported by observational studies.11 This evidence is widely accepted7,12 and neonatal screening is considered the gold-standard procedure for detecting the disease. The extent of the benefits stemming from neonatal screening is not well documented, and a study withholding intervention among a prospectively randomised subgroup to measure the size of these benefits would be unthinkable given the proven benefit of treatment. Owing to this dilemma, we use the Jamaican reality of a largely unscreened population as an opportunity to document the effects of not screening for homozygous sickle cell disease. The corollary is the beneficial effect of neonatal screening for the disease. Without neonatal screening, patient presentation to a health care provider can be delayed at each of three stages: the onset of clinical symptoms, disease diagnosis, and clinic enrolment. In an unscreened homozygous sickle cell disease population, almost all patients have become symptomatic by their eighth birthday,27 of which patients only two-thirds have been diagnosed.28 We now show that only one-third of patients have enrolled in the clinics by that age, implying an equal and considerable delay at these two stages (diagnosis and enrolment). Males enrol at a younger age than females; a feature that has been observed previously.29,30 Delays in diagnosis could be shortened with training programmes for health care workers, and delays in enrolment could be avoided by mandatory referral and a central patient registration facility. It is inevitable, however, that in the absence of a universal neonatal screening programme, a proportion of the homozygous sickle cell disease population will never present to the clinic while the remainder will arrive with varying lengths of delay. It remains unknown whether the enrolled and non-enrolled patients have different morbidity and mortality risks stemming from either an inherent heterogeneity in disease expression, the absence of comprehensive and prophylactic health care, or both. Unscreened clinic coverage describes the likelihood of an unscreened homozygous sickle cell disease patient presenting to a clinic. This unscreened coverage by 18 years old is 60% among births in the early 1980s and decreases thereafter. Newborn screening has been performed exclusively in Kingston during two clinic periods (1973–1981 and 1995–present). Outreach clinics were initiated in western Jamaica between 1973 and 1981, and these clinics probably fuelled the continued increase in enrolment of unscreened patients despite Kingston-based screening reducing the pool www.jmedscreen.com

04 Hambleton cpp 5/11/04 12:03 PM Page 179

Identifying sickle cell disease

of unscreened individuals available in the island’s capital. During the second and ongoing wave of neonatal screening (1995-present), there has been no new homozygous sickle cell disease population tapped to offset the decreased pool of unscreened patients. The remaining pool of unscreened births is from outside of Kingston, and the increased distance from the main clinic facility in Kingston will be a defining factor in the decreased patient referral. Other possible causes of non-enrolment include good health, non-referral, death or migration. The substantial levels of delayed-enrolment and nonenrolment have important consequences for the clinical care of homozygous sickle cell disease patients, for public health policymaking, and for clinical research. First, early-life preventive management and emergency intervention are vital aspects of clinical care and are only possible within a regime of early disease detection and comprehensive followup. We have shown that by the fifth year of life, before which penicillin prophylaxis plays a crucial role in this preventive management, between 53–90% of available homozygous sickle cell disease patients will not have enrolled to a specialist clinic. Second, incomplete and delayed patient enrolment invalidates assessment of health care infrastructure, by underestimating the size, morbidity and mortality of the patient population (‘the burden of disease’). This results in a low visibility among homozygous sickle cell disease populations12 that contributes to a failure to provide timely and adequate care, perpetuating inequality in health. Third, this artificially low burden of disease introduces an important bias in clinical research. Neonatal screening is the only method for achieving a representative patient sample. Despite this, observational research commonly employs cross-sectional clinic-based populations subject to delayed-enrolment and non-enrolment bias. The failure to realise the limitations of studying these biased populations could present a significant problem in the advancement of knowledge concerning homozygous sickle cell disease. Improvements in hygiene, nutrition and control of infection are increasing the relative importance of homozygous sickle cell disease (and other inherited haemoglobin disorders) as a global health problem.2 Our quantification of the low visibility of homozygous sickle cell disease, using unscreened patient coverage at various ages and age at enrolment, has shown that without early, comprehensive and systematic identification of patients, the large burden of homozygous sickle cell disease – suggested by the known prevalence of the disease in Jamaica – remains largely unrecognised. The true disease prevalence remains unknown in many countries and national studies to obtain this information must be a priority. Agencies and governments of countries with a substantial burden of homozygous sickle cell disease must then consider programmes for systematic disease identification, control and management.

ACKNOWLEDGEMENTS Contributors: IH carried out all statistical analyses and prepared the initial draft of the paper. Both authors were responsible for study inception and design and finalised the draft of the paper. Funding: IH was supported by a Fellowship from the Medical Research Council (UK). .................

Authors’ affiliations Ian R Hambleton, Senior Lecturer in Medical Statistics, Sickle cell Unit, Tropical Medicine Research Institute, University of the West Indies, Mona, Kingston 7, Jamaica

www.jmedscreen.com

179

Klaas JJ Wierenga, Research Scientist, Sickle cell Unit, Tropical Medicine Research Institute, University of the West Indies, Mona, Kingston 7, Jamaica; and The Dr. John T. Macdonald Foundation Center for Medical Genetics and the Division of Medical Genetics, Department of Pediatrics, University of Miami, Miami, Florida, USA

REFERENCES 1 Embury SH, Hebbel RP, Mohandas N, Steinberg MH. Sickle cell disease. Basic principles and clinical practice. New York: Raven Press, 1994. 2 Weatherall DJ, Clegg JB. Inherited haemoglobin disorders: an increasing global health problem. Bull World Health Org 2001;79:704–12. 3 World Health Organisation. Joint WHO/TIF Meeting on the Control of Haemoglobinopathies. Report of the VIIth Meeting of the WHO Working Group on the Control of Hereditary Anaemias. Nicosia, Cyprus, 3–4 April 1994. (unpublished document WHO/HDP/TIF/HA/93.1). Geneva, WHO: 1994. 4 Emond AM, Collis R, Darvill D, Higgs DR, Maude GH, Serjeant GR. Acute splenic sequestration in homozygous sickle cell disease: natural history and management. J Pediatr 1985;107:201–06. 5 Gaston MH, Verter JI, Woods G, et al. Prophylaxis with oral penicillin in children with sickle cell anemia. A randomized trial. N Engl J Med 1986;314:1593–99. 6 American Academy of Pediatrics.Section on Hematology/Oncology Committee on Genetics. Health supervision for children with sickle cell disease. Pediatrics 2002;109:526–35. 7 Newborn screening for sickle cell disease and other hemoglobinopathies. NIH Consensus statement Apr 6–8 1987;6:1–22. 8 Zeuner D, Ades AE, Karnon J, Brown J, Dezateux C, Anionwu EN. Antenatal and neonatal haemoglobinopathy screening in the UK: review and economic analysis. Health Technol Assess 1999;3(11):1–186. 9 Davies SC, Cronin E, Gill M, Greengross P, Hickman M, Normand C. Screening for sickle cell disease and thalassaemia: a systematic review with supplementary research. Health Technol Assess 2000;4(3):i–99. 10 Henderson SJ, Fishlock K, Horn ME, Oni L, Bellingham AJ. Neonatal screening for haemoglobin variants using filter paper-dried blood specimens. Clin Lab Haematol 1991;13:327–34. 11 Vichinsky E, Hurst D, Earles A, Kleman K, Lubin B. Newborn screening for sickle cell disease: effect on mortality. Pediatrics 1988;81:749–55. 12 World Health Organisation. Guidelines for the control of haemoglobin disorders. Report of the VIth Annual Meeting of the WHO Working Group on Haemoglobinopathies. Caglieri, Sardinia, 8–9 April 1989. (unpublished document WHO/HDP/WG/HA/89.2). Geneva, WHO: 1989. 13 Ashley-Koch A, Yang Q, Olney RS. Sickle hemoglobin (HbS) allele and sickle cell disease: A HuGE review. Am J Epidemiol 2000;151:839–45. 14 Streetly A, Grant C, Bickler G, Eldridge P, Bird S, Griffiths W. Variation in coverage by ethnic group of neonatal (Guthrie) screening programme in south London. BMJ 1994;309(6951):372–74. 15 Serjeant GR, Serjeant BE, Forbes M, Hayes RJ, Higgs DR, Lehmann H. Haemoglobin gene frequencies in the Jamaican population: a study in 100,000 newborns. Br J Haematol 1986;64:253–62. 16 UN-ONU. United Nations World Population Prospect; the 1996 revision: WHO/HST. 1997. 17 World Bank. World Development Indicators 2001. 2001. 18 Handa S, Neitzert M. Chronic Illness and Retirement in Jamaica. World Bank LSMS Working Paper No. 131. 1998. 19 Serjeant GR. Sickle Cell Disease. 2nd ed. Oxford: Oxford University Press, 1992. 20 Serjeant GR, Grandison Y, Lowrie Y, et al. The development of haematological changes in homozygous sickle cell disease: a cohort study from birth to 6 years. Br J Haematol 1981;48:533–43. 21 Lindsey JK. Fitting bivariate intensity functions, with an application to modelling delays in reporting acquired immune deficiency syndrome. Journal Of The Royal Statistical Society Series A – Statistics In Society 1996;159:125–31. 22 Chong MY, Hamilton DC, Cole DEC. Retrospective estimation of the birth prevalence for delayed onset disorders: application to cystic fibrosis in Nova Scotia. Stat Med 2000;19:743–51. 23 Esbjerg S, Keiding N, Koch-Henriksen N. Reporting delay and corrected incidence of multiple sclerosis. Stat Med 1999;18:1691–706. 24 STATIN. Statistical Institute of Jamaica: Demographic Statistics (Annual editions: 1973–1999). 2000. 25 Akaike H. A new look at statistical model identification. IEEE Transactions on Automatic Control 1974;AC-19:716–23. 26 Falletta JM, Woods GM, Verter JI, et al. Discontinuing penicillin prophylaxis in children with sickle cell anemia. Prophylactic Penicillin Study II. J Pediatr 1995;127:685–90. 27 Bainbridge R, Higgs DR, Maude GH, Serjeant GR. Clinical presentation of homozygous sickle cell disease. J Pediatr 1985;106:881–85. 28 Singhal A, Jones R, David S, Serjeant G. Delayed diagnosis of homozygous sickle cell disease in Jamaica. In: Berchel C, Papiernik E, De Caunes F (eds). Perinatal problems of islands in relation to the prevention of handicaps. France: Editions Inserm, 1992:268–75. 29 Platt OS, Brambilla DJ, Rosse WF, et al. Mortality in sickle cell disease. Life expectancy and risk factors for early death. N Engl J Med 1994;330:1639–44. 30 Wierenga KJ, Hambleton IR, Lewis NA. Survival estimates for patients with homozygous sickle cell disease in Jamaica: a clinic-based population study. Lancet 2001;357(9257):680–83.

Journal of Medical Screening

2004

Volume 11

Number 4