2.1. EPIDEMIOLOGY AND SERVICE DEVELOPMENT

Epidemiology, which is the study of the distribution of a disease in human populations as well as the factors influencing this distribution (3) has, beyond any academic interests, very practical applications in the planning of services. The questions posed by an epidemiologic investigation should lead to concrete information which will result in rational planning aimed at both clinical care and prevention of disease. Developing appropriate health policies depends to a great extent on the availability of accurate epidemiological information concerning the size and distribution of the problem.

There is great variability in the prevalence of haemoglobin disorders across the globe but also in their distribution within a country. The recognition of the effect of the problem depends partly on prevalence, the number of affected individuals, but also on the overall health picture of each country and its socio-economic development. Low resource countries for example with several other serious health agendas, such malnutrition and infectious diseases, often with a high infant mortality from these conditions, may not see the hereditary disorders as a priority even though the affected children will be more susceptible and be the first to die from an infection (4). Visibility of hereditary disorders is often more clear when the infant mortality falls, and by experience this is when the level is around 50 per thousand live births.

In order to understand the impact of the hereditary anaemias on a population, especially with service development in mind, simple numbers are not enough. More complex data must be collected and used to both assess the country situation and promote the planning of services. Such data include:

• Data which indicate the the impact of haemoglobin disorders in terms of numbers: prevalence data such as patient numbers and their location; incidence, which is the number of new affected births which in turn depends on carrier frequency; data on the frequency of consanguinity which may have an impact on birth frequency. These figures are usually given as the overall data for a country but are more useful, if available, for various locations within a country or among the various population groups – this is known micromapping, which is particularly relevant to large populations with divergent ethnic groups. Information on incidence depends also on knowing the crude birth rate, the total number of births per annum, the overall life expectancy etc.
• Data which will indicate the overall health status of a country: basic data includes the infant mortality rate, the under-five mortality rate, the health expenditure per capita and knowledge of the way health provision is financed and the need for out-of-pocket expenditure, which is particularly relevant to chronically sick patients.
• Service indicators: these are the figures derived from the basic data, which will more directly lead to planning services either for patient care or for prevention. The term ‘service indicators’ was coined by Prof Bernadette Modell (1) and these will be further discussed below.
• Changing epidemiology: none of the information mentioned above is constant and changes may occur in the short term but also in the long term. The most important short term change in recent years has been migrations from high prevalence regions of the world, such as Africa, the Mediterranean basin and Asia to low prevalence regions such as northern Europe and the Americas. Long term effects may come about by the reduction of habitual consanguineous marriage (5), by the falling crude birth rate of many populations and by the removal of environmental factors said to favour the carrier rate such as the eradication of malaria (6).
• Outcome measures: these include complication rates, survival rates, causes of death, and quality of life studies, the number of annual births of affected individuals and the effects of prevention measures. These are part of necessary motoring and evaluation processes which will audit the whole service provision and its dynamic evolution.
• Cost effectiveness studies: these are necessary public health components of service provision which should be coupled with Health Technology Assessment studies for the best possible use of resources which will determine the viability and sustainability of these national programmes. These will not be considered in detail in this chapter since they are part of a separate science of health economics. However some comparisons in the past have been made between the cost of patient care and the cost of prevention which are useful indicators for health planners (7-9).

It is with this service orientated perspective that we approach the study of the epidemiology of the haemoglobin disorders.

2.2. EPIDEMIOLOGICAL INFORMATION: METHODOLOGY

Epidemiology provides the foundation of public policy and the means for monitoring and evaluating these policies, and studying the changes over time and place.

2.3. CALCULATING THE EXPECTED NUMBER OF HOMOZYGOTE BIRTHS

From the carrier rates a calculation of the number of affected conceptions per 1000 live births can be made based on the Hardy-Weinberg equation for a recessively inherited single gene disorder. This depends on certain assumptions:

1. That the population is mating randomly i.e. there are no consanguineous marriages.
2. There is no selection process (e.g. an ongoing prevention program) or genetic drift or frequent spontaneous mutations.

With these assumptions satisfied, the equation is as follows: p² + 2pq + q² =1

p = thalassaemia gene frequency (½ carrier frequency)

q = Hb A gene frequency = 1-p

p² = the frequency homozygotes at birth

pq= the frequency of heterozygotes

q² = the frequency for homozygote normals

Example:

In a country with a carrier frequency of 3% the gene frequency is approximately 1.5% or 0.015

• p = 0.015 q = 1-0.015=0.985
• The birth rate of homozygotes = p²
  p² = 0.000225 = 0.0225% or 0.225/1000
• 2pq = proportion of carriers born 0.02955 ~ 3%
• The proportion of normals at birth = q² = 97%
• p²+2pq+q² = 0.000225 + 0.02955 + 0.970225 = 1

If the country has 500,000 births/yr, then 112.5 births/yr will be homozygotes.

A more rough calculation is this:

Carriers - 3% of the population or 1/33

Marriages between carriers 1/33 × 1/33 = 1/1089

Affected births = 1/1089 × ¼ = 1/4356

$500,000\times \frac{1}{4356}=114.8\text{births / year}$

These calculations must be modified in two situations:

1. If two haemoglobin disorders co-exist in a population, which may interact: e.g. β-thalassaemia with HbS, or with HbE.
   In the case of co-existence of interacting haemoglobin disorders the Hardy-Weinberg equation is still applied. The sum of the carrier frequencies is used. For example where β-thalassaemia and HbE coexist then p=the combined β-thalassaemia and HbE frequency (e.g. if β-thal=3% and HbE=3.5% then the total is 6.5% and so p= 0.035). A nomogram, derived from this calculation is here reproduced from the WHO publication “Guidelines for the control of Haemoglobin disorders”, B. Modell, 1994).
2. Where consanguinity is frequent.
   Where consanguineous marriage is frequent the frequency of homozygote births increases for a given carrier frequency. The importance of consanguinity is inversely related to the gene frequency. If a mutation is rare, a carrier who marries another member of his/her own family group is more likely to have married another carrier since both may have inherited the rare trait from a common ancestor. A partner from the general population is unlikely to carry the same rare gene (15, 16). The contribution of consanguinity is expressed as the coefficient of consanguinity (or inbreeding), which is half of the probability that a couple share an allele (F). For a population the coefficient of consanguinity is equal to the mean of individual coefficients (a). The population coefficient is included in the Hardy-Weinberg equation when calculating the annual affected births:
   (p² + apq) + 2(pq - apq) + (q² + apq) = 1
   Values for population coefficients of various populations and sub-groups can be viewed on line (in www.consang.net).
   Finally in order to calculate the annual affected births the demographics of each location must be known, which includes the total population, the general birth rate (per 1000 live births), the infant mortality rate, and the total number of births.
   The number of expected births of children with clinically significant syndromes is derived from the calculations discussed above. This will also provide an indicator [3] for planning services for patient care if there is no prevention, since it is the number of patients expected to be added to the existing patients who will need blood and other resources.

2.4. THE REQUIREMENTS FOR HEALTH PLANNING – THE SERVICE INDICATORS

The needs for effective planning of prevention and control programmes can only be based on epidemiological information and this are summarised in the “service indicators”, already mentioned above. These were first described by Professor Bernadette Model as practical measures to be considered for service developed (1). These indicators are the following:

• Indicator for patient care: the annual affected new births in the absence of prevention – the potential increase in patients and their needs in blood, medications, laboratory and other services, including manpower planning.
• Indicator for carrier screening: this requires prior epidemiological information on carrier rates and carrier distribution. A screening policy is derived from such information such as premarital screening of the whole population or ante-natal screening according to the annual number of pregnancies in at-risk groups. This indicator will help in planning appropriate laboratory services.
• Indicator for the number of carriers expected for counselling and information. This information is derived from annual number of carriers detected in screening.
• Indicator for the need of expert services such as risk assessment and genetic counselling. This is the annual number of carrier couples detected and the pregnancies to carrier couples.
• Indicator for prenatal diagnosis: the annual number of at-risk pregnancies.

2.5. NEONATAL SCREENING

Neonatal screening, discussed in detail in Chapter 10, will provide information on the size of the problem in a given population and also is an indicator for patient care. In haemoglobin disorders there are three situations in which it may be usefully applied:

• The detection of sickle cell disease
• The detection of alpha thalassaemia
• The detection of other haemoglobin variants

2.5.1. SICKLE CELL DISEASE

Neonatal screening is relevant to areas where sickle cell disease is prevalent and where the at-risk couples are not detected by a population-based carrier screening programme. The rationale for its adoption historically was that early detection of affected children will lead to timely interventions which will reduce the likelihood of life threatening complications such as pneumococcal infections. These interventions include penicillin prophylaxis, vaccinations, education of parents in early detection and response to complications such as infections or splenic sequestration, Transcranial Doppler to detect intracranial vascular stenosis; these methods have been shown to reduce both morbidity and mortality (17, 18). In addition, neonatal screening provides invaluable epidemiological information regarding the frequency and geographical distribution of homozygous sickle cell disease, sickle cell thalassaemia, HbS/D, HbSC disease and HbS/O-Arab (19 - 22). Despite its usefulness, practical application is limited to few countries which include the USA, Jamaica and some European countries (see Table 2.1, provided by M de Montalembert and B. Gulbis). In many high prevalence areas, including Africa and South America, the service is either targeted to high-risk groups, limited to parts of a country or not provided at all. Targeted screening has led to the possibility of discrimination, especially in countries where sickle cell disease is prevalent in ethnic minorities and universal screening may be preferable. The success of neonatal screening as a means to timely and effective patient care depends to a great extent on follow up of cases detected. This has been reported to be a problem in several African settings (23).

Table 2.1

Neonatal screening for sickle cell disease (SCD) in Europe.

2.6. CHANGING EPIDEMIOLOGY - THE EFFECT OF MIGRATIONS

Changes in the epidemiology of inherited disease can be brought about by several possible factors which include natural selection, new mutational events, founder effects, genetic drift and migrations, and reproductive behaviour which includes customary consanguineous marriage and the reduction of crude birth rate in a population. It is not the purpose of this chapter to discuss the field of population genetics but to emphasise the practical consequences of these events, especially the migrations of the recent past. Migrations have been going on for some centuries and have led to population mixes in many situations, as are the migrations from south to north Italy. The same can be said of Georgia where over the centuries Azeris, Armenians, Greeks and Jews have mixed with indigenous Georgians so that it is difficult today to identify genetically distinct groups as far thalassaemia is concerned (31). In other situations racial differences and minimal intermarriage have kept haemoglobinopathy genes within ethnic or racial groups, as is the case of sickle cell genes in north America. However the more recent migrations of the last two centuries have been to destinations in Europe and the Americas. These are mostly economic migrations and their effects in terms of genetic disease, are difficult to estimate in terms of numbers with accuracy, since many factors need to be considered. Such factors include the permanency of the migration, whether it is a migration of single people or of families, whether the migrant will marry locally or from the country of origin, whether consanguineous marriage will still be practiced in the host country, whether there will be free choice partner or an arranged marriage, whether birth rate of immigrant families will be that of the home or the host country and whether there is a second generation of migrants and the customs that they have adopted. The effects of these many variables are difficult to estimate and so in the TIF estimates of the effect of migration we consider country reports on the total number of migrants and their countries of origin, the carrier rate in the home country, and at the same time assume intermarriage between members of the same community. All calculated estimates are therefore based on assumptions which can lead only to approximations. However the size of the problem in each country can be assessed more accurately through surveys (32 - 35) and in the case of variants through neonatal screening programmes and through national registries.

2.7. OUTCOME STUDIES

Beyond the studies aimed at health planning, epidemiological techniques must be employed in outcome studies to monitor and evaluate the effectiveness and quality of services. Such studies include patient survival rates and age distribution of patients (36, 37), quality of life studies (38), complication rates (39) and health economic studies. Knowledge of the causes of death provides an important indicator to direct clinical research and therapeutic interventions aimed at improving survival of thalassaemia patients. The identification of cardiac complications of iron overload as the cause of 60%-70% of deaths (40), led eventually to the early detection of heart iron deposition by cardiac MRI and early intensification of iron chelation to increase survival. Surveillance of prevention programs is also important and is discussed fully in chapter 3. Continual surveillance of outcomes is, therefore, an important epidemiological exercise.

2.8. THE CURRENT SITUATION

Accurate epidemiological data are difficult to collect and a major challenge is collecting data from non-indigenous population groups in Europe (35). From these groups much information is derived from epidemiological information in the country of origin and not always from direct surveying of local ethnic minority. It is assumed that the migrant group is a representative sample of the population of the country of origin, with the same birth rate, etc. and calculations are made on these assumptions. Much of the regional and global data may not be accurate but are the best estimates on available information.

Based on such data, the most recently published global estimates are the March of Dimes Global Report on Birth Defects published in 2006 (41) with a more detailed focus on haemoglobin disorders, presented in the WHO bulletin in 2008 (1). In their comments Modell and Darlison suggest that policies for treatment and prevention are required in 71% of 229 countries globally: this means 162 countries worldwide.

Of the 42,409 annual β-thalassaemia conceptions that the authors estimate is the global incidence, how many survive to receive treatment? Of the estimated 1.33 million at-risk pregnancies, how many are aware of their risk, how many are counselled and given choices? This is difficult to know. In Southeast Asia - which includes some of the poorer countries – of 9983 expected transfusion dependent patients, 962 (10%) are actually transfused. It is clear that socio-economic development is not the only factor - or even the most important one – as to why patients do not receive appropriate treatment. It seems that the attention paid by health services to such rare (in some populations) and chronic disorders, which require lifelong and expensive medical care, is not adequate and the concept of using epidemiological data to plan and locate services has not yet sufficiently penetrated bureaucratic health systems. Another factor, as stated earlier, is the difficulty in catering for “immigrant” diseases in many parts of Europe. Some countries in Europe have tackled the problem in an organised way, partly because they host several generations of immigrants. One example is the United Kingdom where 7% of all residents are from minorities at-risk and 366 pregnancies per year are affected (42, 43). There are 900 registered patients, centres of expertise in London and a developing network with physicians who see less than 10 patients. Neonatal screening, ante-natal clinic screening and counselling services carried out by trained personnel who speak the language and share the culture of the immigrant groups. Such services need to be established in many countries across Europe.

In the meantime, more detailed epidemiological information is required at national level, which should include national registers of patients, carrier rates among immigrant populations, micromapping and the calculation of service indicators. It is on such an epidemiological basis that rational planning can lead to effective and equitable services both for the treatment of patients and for disease prevention.

REFERENCES

1.

Modell B, Darlison M. Global epidemiology of haemoglobin disorders and derived service indicators. Bull World Health Organ. 2008;86(6):480–487. [PMC free article: PMC2647473] [PubMed: 18568278]

2.

Weatherall DJ. The inherited diseases of haemoglobin are an emerging global health burden. Blood. 2010;115(22):4331–4336. [PMC free article: PMC2881491] [PubMed: 20233970]

3.

Lilienfeld AM, Lilienfeld DE. Foundation of Epidemiology. 2nd edition. 1980.

4.

Alwan A, Modell B. Recommendations for introducing genetics services in developing countries. Nat Rev Genet. 4(1):61–68. 200. [PubMed: 12509754]

5.

Bittles AH, Black ML. Consanguinity, human evolution and complex diseases. Proc Nat Acad Sci. 2010;107(S1):1779–1786. [PMC free article: PMC2868287] [PubMed: 19805052]

6.

Firdous N, Gibbons S, Modell B. Falling prevalence of beta-thalassaemia and eradication of malaria in the Maldives. J Community Genet. 2011;2(3):173–189. [PMC free article: PMC3186032] [PubMed: 22109823]

7.

Ginsberg G, Tulchinsky T, Filon D, et al. Cost-benefit analysis of a national thalassaemia prevention programme in Israel. J Med Screen. 1998;5(3):120–126. [PubMed: 9795870]

8.

Karnon J, Zeuner D, Brown J, et al. Lifetime costs of β-thalassaemia major. 1999;21:37–385. [PubMed: 10671989]

9.

Ghobti N, Tsukatani T. Evaluation of the national health policy of thalassaemia screening in the Islamic Republic of Iran. East Mediterr Health J. 2005;11(3):308–318. [PubMed: 16602449]

10.

Thuret I, Pondarre C, Loundou A, et al. Complications and treatment of patients with β-thalassaemia in France: results of the National Registry. Haematologica. 2010;95(5):724–729. [PMC free article: PMC2864377] [PubMed: 20007138]

11.

Voskaridou E, Ladis V, Kattamis A, et al. A national registry of haemoglobinopathies in Greece: Deducted demographics, trends in mortality and affected births. Ann Hematol. 2012;91(9):1451–1458. [PubMed: 22526366]

12.

Arts DGT, De Kaizer NF, Scheffer GJ. Defining and improving data quality in medical registries: a literature review, case study and generic framework. J Am Med Inform Assoc. 2002;9:600–611. [PMC free article: PMC349377] [PubMed: 12386111]

13.

Levine MK, Julian JA. Registries that show efficacy: good, but not good enough. J Clin Oncol. 2008;26(23):5316–5319. [PubMed: 18854560]

14.

Negro Calduch E, Diaz A, Diaz M. Ethical and legal issues related to health access for migrant populations in the Euro-Mediterranean area. Eurosurveillance. 2008;13(50) [PubMed: 19087864]

15.

Modell B, Kuliev AM. Social and genetic implications of customary consanguineous marriage among British Pakistanis. Galton Institute Occasional Papers. 1992. p. 4. (Second Series).

16.

Bittles AH. The role and significance of consanguinity as a demographic variable. Population and Development Review. 1994;20:561–568.

17.

Vichinsky E, Hurst D, Earles A, Kleman K, Lubin B. Newborn screening for sickle cell disease: effect on mortality. Pediatrics. 1988;81(6):749–755. [PubMed: 3368274]

18.

Lê PQ, Ferster A, Cotton F, et al. Sickle cell disease from Africa to Belgium, from neonatal screening to clinical management. Med Trop (Mars). 2010;70(5-6):467–470. [PubMed: 21516988]

19.

Streetly A, Latinovic R, Henthorn J. Positive screening and carrier results for the England-wide universal newborn sickle cell screening programme by ethnicity and area for 2005-07. J Clin Pathol. 2010;63(7):626–629. [PMC free article: PMC2990147] [PubMed: 20591912]

20.

Gulbis B, Tshilolo L, Cotton F, Lin C, Vertongen F. Newborn screening for haemoglobinopathies: the Brussels experience. J Med Screen. 1999;6(1):11–15. [PubMed: 10321364]

21.

Thuret I, Sarles J, Merono F, et al. Neonatal screening for sickle cell disease in France: evaluation of the selective process. J Clin Pathol. 2010;63(6):548–551. [PubMed: 20498028]

22.

Bouva MJ, Mohrmann K, Brinkman HB, et al. Implementing neonatal screening for haemoglobinopathies in the Netherlands. J Med Screen. 2010;17(2):58–65. [PubMed: 20660432]

23.

Management of Haemoglobin disorders; Report of a joint WHO-TIF meeting; Cyprus: Nicosia; Nov 16-18, 2007.

24.

Siala H, Ouali F, Messaoud T, Bibi A, Fattoum S. alpha-Thalassaemia in Tunisia: some epidemiological and molecular data. J Genet. 2008;87(3):229–234. [PubMed: 19147907]

25.

Kyriacou K, Kyrri A, Kalogirou E, et al. Hb Bart’s levels in cord blood and alpha-thalassemia mutations in Cyprus. Hemoglobin. 2000;24(3):171–180. [PubMed: 10975437]

26.

Alkindi S, Al Zadjali S, Al Madhani A, et al. Forecasting hemoglobinopathy burden through neonatal screening in Omani neonates. Hemoglobin. 2010;34(2):135–144. [PubMed: 20353348]

27.

Bouva MJ, Sollaino C, Perseu L, et al. Relationship between neonatal screening results by HPLC and the number of α-thalassaemia gene mutations; consequences for the cut-off value. J Med Screen. 2011;18(4):182–186. [PubMed: 22184733]

28.

Hoppe CC. Newborn screening for non-sickling hemoglobinopathies. Hematology Am Soc Hematol Educ Program. 2009:19–25. [PubMed: 20008178]

29.

Benz EJ Jr. Newborn screening for α-thalassemia--keeping up with globalization. N Engl J Med. 2011;364(8):770–771. [PubMed: 21345108]

30.

Tang HS, Zhou JY, Xie XM, et al. Screening for common nondeletional α-thalassemias in Chinese newborns by determination of Hb Bart’s using the Sebia Capillarys 2 electrophoresis system. Hemoglobin. 2012;36(2):196–199. [PubMed: 22239481]

31.

Mestiashvili I. Heterogeneity and gene-geography of β-thalassemia in Georgia. Georgian Med News. 2010 Nov;(188):45–51. [PubMed: 21178203]

32.

Amato A, Grisanti P, Lerone M, et al. Prevention strategies for severe hemoglobinopathies in endemic and nonendemic immigration countries: the Latium example. Prenat Diagn. 2009;29(12):1171–1174. [PubMed: 19816879]

33.

Hickman M, Modell B, Greengross P, et al. Mapping the prevalence of sickle cell and beta thalassaemia in England: estimating and validating ethnic-specific rates. Br J Haematol. 1999;104(4):860–867. [PubMed: 10192451]

34.

Henderson S, Timbs A, McCarthy J, et al. Incidence of haemoglobinopathies in various populations - the impact of immigration. Clin Biochem. 2009;42(18):1745–1756. [PubMed: 19497317]

35.

Modell B, Darlison M, Birgens H, et al. Epidemiology of haemoglobin disorders in Europe: an overview. Scand J Clin Lab Invest. 2007;67:39–70. [PubMed: 17365984]

36.

Zurlo MG, De Stefano P, Borgna-Pignati C, et al. Survival and causes of death in thalassaemia major. Lancet. 1989;2(8653):27–30. [PubMed: 2567801]

37.

Telfer P, Coen PG, Christou S, et al. Survival in medically treated thalassaemia patients in Cyprus. Trends and risk factors over the period 1980-2004. Haematologica. 2006;91(9):1187–1192. [PubMed: 16956817]

38.

Telfer P, Constantinidou G, Andreou P, et al. Quality of life in thalassemia. Ann N Y Acad Sci. 2005;1054:273–282. [PubMed: 16339675]

39.

Borgna-Pignatti C, Cappellini MD, De Stefano P, et al. Survival and complications in thalassaemia. Ann NY Acad Sci. 2005;1054:40–47. [PubMed: 16339650]

40.

Anderson LJ, Holden S, Davis B, et al. Cardiovascular T2-star (T2*) magnetic resonance for the early detection of myocardial iron overload. Eur Heart J. 2001;22(23):2171–2179. [PubMed: 11913479]

41.

Christianson A, Howson CP, Modell B. March of Dimes Birth Defects Foundation. White Plains; New York: 2006. March of Dimes Global Report on Birth Defects.

42.

Modell B, Khan M, Darlison M. Survival in beta-thalassaemia major in the UK: data from the UK Thalassaemia register. Lancet. 2000;355(9220):2051–2052. [PubMed: 10885361]

43.

Hickmann M, Modell B, Greengross P, et al. Mapping the prevalence of sickle cell and beta-thalassaemia in England: estimating and validating ethnic specific rates. Br J Haematol. 1999;104:860–867. [PubMed: 10192451]