The Danish Twin Registry

Abstract

Introduction

The Danish Twin Registry is a unique source for studies of genetic, familial and environmental factors on life events, health conditions and diseases.

Content

More than 85,000 twin pairs born 1870–2008 in Denmark.

Validity and coverage

Four main ascertainment methods have been employed. Completeness of ascertainment varies according to birth cohorts. For birth cohorts 1870–1930 both twins should survive to age 6 years. From 1931–1968 72% of all twin pairs has been ascertained, with complete ascertainment of all live born twins since 1968.

Conclusion

Because twins have been identified independent of traits and on a population basis, the Danish Twin Registry is well suited for studies to understand the influence of genetic and environmental factors for a wide variety of diseases and traits.

Introduction

The main purpose of the Danish Twin Registry (DTR) is to study and disentangle the importance of genetic, familial and environmental factors on life events, health conditions and diseases.

With more than 85,000 twin pairs, the DTR is one of the largest registers in the world containing multiple births, and it is the oldest nationwide twin register in the world, initiated in 1954 by the late professors M Hauge and B Harvald. During the past 55 years, the register has grown from initially including twin cohorts from 1870–1910 to now covering all birth cohorts of twins since 1870.

Content

The ascertainment of twins has taken place using four different methods, depending on the birth cohorts. However, despite different methods the ascertainment involves at least two common steps: identification of the twins, i.e. date of birth and names of the twins, and tracing of the twins, i.e. their vital status (alive, emigrated, or dead) and addresses, if alive. Figure 1 illustrates the period covered by DTR and the four different ascertainment methods, which are briefly described below. A detailed description can be found in references [1–4].

Figure 1.

Lexis diagram illustrating coverage of Danish Twin Registry and the four different ascertainment methods. Birth cohorts 1911–1930 include only same-sexed twin pairs, while all other birth cohorts include both same-sexed and opposite-sexed twin pairs. Cohorts: Dark grey = 1870–1930; medium grey = 1931–1952; grey = 1953–1982; light grey = 1983–2008.

Birth cohorts 1870–1930

Twins from the birth cohorts 1870–1930 were ascertained by professors M Hauge and B Harvald in the 1950s and 1960s. The twins were identified by searching the birth registers in every parish in Denmark. The search was carried out by the local clergy who were asked to report twin births where both twins had survived to the age of 6 years. For each twin pair, date of birth, names of the twins, names of their parents and their address were requested. Subsequently, all reported twins were searched through available population registers and other public sources, and whenever needed, their closest relatives were sought; an effort that continued for more than 30 years.

In the birth cohorts 1870–1910 twin pairs of same sex and opposite sex were reported, while only same-sexed twin pairs were reported for the cohorts 1911–1930. Although Nordschleswig (until 1921 part of Germany) became a part of Denmark in 1921, the search for twins was restricted to the same area as before 1920.

Birth cohorts 1931–1952

The nationwide ascertainment of twins from the birth cohorts 1931–1952 was done from 1996–1998. Twins were ascertained mainly based on information from the Civil Registration System (CRS) and the fact that most twins are born on the same date, born in the same parish and given the same surname at birth. A search in CRS extracted all sets of persons fulfilling these three conditions and being born in Denmark during 1931 to 1952, and confirmation of twin status was obtained by a one–page questionnaire; approximately half of those responding were twins. This method required both twins to have survived to 2 April 1968, when CRS was established. In order to estimate the loss due to the death of one or both twins prior to this date, a number of birth registers were scrutinized to identify all twin births in the period.

In addition, some groups of twins had previously been identified by other means and were used as secondary sources, e.g. a manually kept register of twin conscripts providing same-sexed male twin pairs as well as opposite-sexed twin pairs. A twin panel from 1969 identified from CRS as persons born on the same date and having the same address provided twins from the Copenhagen area.

Birth cohorts 1953–1982

Twins born between 1953 and 1982 were ascertained in the beginning of the 1990s, primarily using CRS as the source for identification. The primary key to the identification was the link between mother and children. Pairs of people with the same mother and born within three consecutive days were assumed to be twins. Due to the method of establishing the links between parents and children, it was only possible to perform a reliable search from 1953. The number of twin pairs identified from the earliest cohorts was small, but all twin pairs born in the period from 2 April 1968, to 31 December 1982, were ascertained by this method if both twins were live born. Pairs with one or both stillborn were not identified, because stillborn children do not get a unique personal identification number (CPR-number). Since the identification source was CRS, all additional information needed for tracing was immediately at hand. Additional twins were found in secondary sources mentioned above.

Birth cohorts since 1983

The youngest twins from birth cohorts 1983 and onwards have been identified from the Medical Birth Register where all births, including stillbirths, have been registered since 1973. Multiple births are marked by a special code, and information about the twins and their mother has been extracted, including the CPR-number. Tracing the twins was done by linkage to CRS.

The Medical Birth Register has also been used to supplement the birth cohorts from 1973 to 1982 with stillborn twins, so that the DTR has a complete registration of all twin births in Denmark since 1973.

Through the ascertainment procedure, basic data on each twin individual and twin pair such as gender, name, zygosity, and date and place of birth have been collected. Besides the basic data, the twins are followed by yearly updates of vital status, addresses, and marital status. A special variable indicates whether a twin is available for participation in surveys. Information about other family members, like parents, spouses and children, are also collected.

Furthermore, the register is linked to other national health-related registers on a regular basis. These include the Danish Cancer Registry, the Danish Register of Causes of Death, the Danish National Patient Register, and the Medical Birth Register.

In addition to register linkage, surveys are used to obtain information that is available only from the twins themselves, such as physical and cognitive function, occupation and lifestyle habits. Large-scaled self-administered questionnaire surveys have been performed in 1966, 1994, 2002 and 2003, while interview-based surveys have been conducted at regular intervals since 1995 [5]. Biological material has been collected as part of the interview-based surveys, either in form of blood spots on filter paper or blood samples.

Validity and coverage

Table I shows the number of twin pairs in the DTR identified by name and date of birth along with the number of twin births in Denmark since 1870 [6].

Table I.

Number of twin pairs in Danish Twin Registry by zygosity, and twin pairs born in Denmark 1870–2008 [6].

Birth cohorts	Zygosity					Twin pairs born
	MZ	DZ-SS	DZ-OS	UZ	Total
1870–1910	1,834	3,342	5,142	2,919	13,237	37,914
1911–1930	1,635	3,326	1	1,188	6,150	22,700
1931–1952	2,451	5,406	6,626	2,598	17,081	24,574
1953–1967a	2,470	3,871	3,825	729	10,895	14,332
1968–1982a	2,788	2,887	2,921	1,616	10,212	9,856
1983–2000	1,769	4,182	5,555	4,692	16,198	16,195
2001–2008	–	–	4,304	6,809	11,113	11,113
Totalb	12,947	23,014	28,374	20,551	84,886	136,684

MZ = monozygotic twin pairs, DZ-SS = same-sexed dizygotic twin pairs, DZ-OS = opposite-sexed dizygotic twin pairs, UZ = unknown zygosity.

^aTwin pairs born outside Denmark are included in DTR (678 twin pairs).

^bIn addition 775 triplets and 22 quadruplets are included in Danish Twin Registry.

A number of comments are necessary to understand the seemingly low coverage for the older birth cohorts. First, for the period 1870–1930 only twin pairs in which both twins had survived to 6 years of age, were reported. Secondly, for the birth cohorts 1911–1930 only same-sexed twin pairs were reported. Thirdly, infant mortality is three to five times higher for twins than singletons, which especially influences the identification procedures involving CRS (birth cohorts 1931–1967) – for the birth cohorts 1931–1952 70% of all twin births have been ascertained. Based on a thorough search in selected parish registers it has been estimated that almost 90% of the eligible twin pairs were ascertained taking into account infant mortality and survival to age 6 years [4]. From 1968 there is complete ascertainment of twin pairs with both members live born, and since 1973 complete ascertainment of all multiple births due to the use of the Medical Birth Register.

The central variable in the twin register is zygosity, i.e. whether the twin pair originates from one fertilized egg, thus having identical genes, or from two fertilized eggs, thus having half of their segregating genes in common, like ordinary siblings. Opposite-sexed twin pairs are always dizygotic, while same-sexed twin pairs can be either monozygotic (MZ) or dizygotic (DZ).

Throughout the entire period of existence of the DTR, zygosity determination among same-sexed twins has been based on questions about the degree of similarity between co-twins. All initial questionnaires to the twins (or their relatives, if the twin was dead or had emigrated) included three or four questions about this. Based on the answers to these questions, the zygosity has been established as MZ or DZ. In case of disagreement between the two twins or inconsistent answers, zygosity has been classified as unknown (UZ). If only one twin answered, zygosity diagnosis was based on that answer alone.

When biological material has been sampled, the zygosity can be further assessed by means of genetic markers.

The questionnaire method has been proved to assign correct zygosity in 95% of all twin pairs, compared to zygosity determined by genetic markers [7].

The assessment of zygosity depends on the cooperation of the twins and, for the younger cohorts, their parents. In addition, both twins have to survive to at least 6 years of age before the similarity method can be used, which may explain the relatively high number of twin pairs with unknown zygosity for the birth cohorts 1931–1952.

Conclusion

Despite the various limitations in the completeness and truncations due to the different ascertainment methods, the Danish Twin Registry constitutes a valuable research resource, spanning a period of almost 140 years. This opens possibilities for the study of diseases and health-related behaviour in all ages. Besides the classical twin study of the relative importance of genetic and environmental factors and twin-control of environmental exposures, it also makes it possible to further study the influence of age in relation to genes, environment and the interaction of these.