A new Swedish twin registry containing environmental and medical base line data from about 14,000 same-sexed pairs born 1926-58.

Since the beginning of the 1960's the Department of Environmental Hygiene of the Karolinska Institute and of the National Swedish Environment Protection Board has maintained a registry on about 10.000 same-sexed twin-pairs born 1886-1925. The registry has been used for studies of morbidity and mortality against the background of certain external risk factors, especially smoking. Supported by the Research Committee of the National Swedish Environment Protection Board, the establishment of a new twin registry covering younger age cohorts was initiated in 1970. The main purposes of this registry are to evaluate individuals' adaptation to changes in the environment and to study the effects of the environment on human health. Moreover, the registry is intended to be usable in obtaining specific target groups with a certain exposure. The main reason for the establishment of a twin registry instead of using a representative sample from the entire population is that a twin population offers some additional analytical possibilities. With twins composing the study group, it is possible to assess the role of the genetic factor for different variables. In addition, the twin approach provides opportunities of analyzing the subjects as matched pairs--the twin control method. This method involves the evaluation of the effect of one factor, to which one twin in the pair is exposed and the other not, while other variables are kept constant in a far-reaching way, e.g. sex, age, genetic composition, childhood, environment, etc. A twin registry can also be employed as a general epidemiological base line registry. For such purposes, the twin methodology is ignored, and the effects of some agent, for example having to do with environmental hygiene, are studied on all of the individuals, independently of their twin pair qualities. The registry is then in many regards comparable with a registry of non-twins. The first step in creating the Swedish twin registry was to compile information on names and addresses. By means of searches in birth records for the years 1926-1949 and birth notices for the years 1950-1967, registration was made of ca 110 000 twin individuals, which constitute virtually all twins accounted for in official statistics. Of these, ca 13 800 were stillbirths or infant deaths. Before the search for the twins' current addresses was begun, all pairs in which at least one individual was dead were eliminated. Such was the case in ca 10 250 pairs. The address search hence covered about 89 500 individuals and was carried out with the assistance of two location procedures. The first was applied to those born between 1926 and 1949 and the second to those born between 1950 and 1967. The result was that in ca 96% of the pairs both members (37 590 pairs) or one member (4 853 pairs) could be found. It could be anticipated that of the latter pairs 1 450 would not be useable due to the death of one of the twins. The data collection has been carried out via a mail questionnaire which consists of the following main areas of inquiry: medication, annoyance experienced because of factors in the general and occupational environments, smoking and drinking habits, physical activity, food habits, psychosocial status, occupational and educational history, conditions of place of dwelling and certain background data. Before the questionnaire was sent out to the twin population, it was tested in a pilot investigation, which purported to study the effectiveness of the data collection procedure and to elucidate the validity of some of the individual questions. The pilot investigation showed an external nonresponse rate of 18%. Regarding the ability of the subjects to answer the questions, it was revealed that only a few of the question categories presented problems which gave rise to minor changes in the questionnaire. Comparison between the answers on the questionnaire and a parallel interview showed a high degree of agreement. Thus the questionnaire was considered satisfactory. The principal data collection was carried out during the period January - May 1973 and during the period December 1973 - February 1974. Of the 42 294 individuals included in the data collection, 3 339 could not be contacted because of death, illness etc. Responses have been received from 32 374 individuals, which corresponds to 83% of the subjects who could be contacted. Regarded as a population of twin pairs, 3 262 pairs have not been utilizable for similar reasons. The total number of pairs in which both have answered the questionnaire is 13 811, corresponding to a little over 77%. There are three versions of the twin registry; a birth registry, an address registry and a questionnaire registry. The birth registry contains birth data from about 110,000 twin individuals. The address registry contains in addition current information as to name and address as well as current and past census registration. The questionnaire registry, which comprises twins born 1926-1958, moreover contains information obtained on the basis of the mail questionnaire.