Cause of death
Contact: Martijn Huisman
Mortality of LASA respondents is reported by the Municipal Personal Records Database (BRP formerly: GBA) or through family members. By using date of birth, LASA mortality data is linked with mortality data of Statistics Netherlands (Centraal Bureau voor de Statistiek, CBS) to check the LASA mortality data and attain the cause of death. In addition, cause of death is obtained from Statistics Netherlands. Statistics Netherlands obtains this through an obligatory reporting system in the Netherlands, through which the treating doctor or the municipal coroner of a deceased person is obligated to send a cause of death form (B-statement) to the Register of Births, Marriages, and Deaths of the municipality where the death occurred. This is then sent directly to CBS, since the B-statement is collected for statistical purposes only. The causes of death recorded in the B-statement are translated into codes according to the International Classification of Diseases and Related Health Problems (ICD) of the World Health Organization (WHO). When the cause is not recoded clearly, written or telephone enquiries are made by Statistics Netherlands.
Primary and Secondary cause of death
Statistics Netherlands documents one primary cause of death and up to three secondary causes. Therefore, LASA too registers one primary and up to three secondary causes of death. The primary cause of death is defined as the disease or event that started the chain of events that led to death. The secondary cause of death is either a consequence or complication of the primary cause, or another disease which might have contributed to the death of the person. For some respondents, no municipality data are available because they either moved abroad or have a secret municipality code.
Ninth and tenth version ICD
Before 1996, cause of death is defined according to ICD-9 (WHO, 1977). From 1996 onwards, the ICD-10 is used (WHO, 1992). ICD-10 codes can be looked up easily on the website of the World Health Organization. The ICD-9 codes are no longer available online. ICD-10 differs from the 9th revision in several ways. Categories have been added, chapters have been rearranged, titles have been changed and conditions have been regrouped. ICD-9 codes are divided into 19 chapters and ICD-10 into 22 chapters. The chapters are further divided into smaller groups. Furthermore, ICD-9 has numeric categories, whilst ICD-10 has alphanumeric categories.
Matching ICD-9 and ICD-10
If interested in one specific cause of death, one option is to define the codes corresponding to a certain cause of death for ICD-9 and ICD-10, separately, by looking it up by hand. This is illustrated in the following two examples. First, mortality caused by cardiovascular diseases, the codes 3900 to 4599 from ICD-9 represent cardiovascular disease and the codes I000 to I999 of ICD-10 represent cardiovascular disease, as applied by Schalk, Visser, Bremmer, Penninx, Bouter and Deeg (2006). Another example is mortality caused by vascular dementia. Up until 1995, code 2904 from ICD-9 is defined as vascular dementia. From 1996, codes F010 to F019 of ICD-10 are defined as vascular dementia. Although there are cases in which ICD-9 can be precisely matched with ICD-10, ICD-10 is a completely different code set compared to ICD-9. Comparisons between ICD-9 and ICD-10 are not always possible. De Boo, Bijlsma and Hoogenboezem (1998) reported the most important differences between ICD-9 and ICD-10.
A second option is to use the Beldo list (BELangrijke DOodsoorzaken), created by Statistics Netherlands (CBS, 1993). This is a shortlist of causes of death in the Netherlands, which are matched with ICD-8, ICD-9 and ICD-10. They aimed to include a selection of causes of death, based on their relevance for the Netherlands. Those causes that are common in the Netherlands are considered relevant. The Beldo list includes 17 chapters (i.e. 17 causes of death), further divided into subgroups of causes of death. However, the Beldo list does not contain all important causes of death when considering the Dutch situation. For instance, the Beldo list does not include unspecified dementia (ICD-10 code F03) (Verweij, 2002). These diseases might be of particular importance for older persons. Still, the Beldo list might be useful if a quick overview is needed of the cause of death (Verweij, 2002).
Based on the first version of the Beldo list, Eurostat created the ‘European Shortlist’ in august 1998 (Eurostat, 1998; Verweij, 2002). In the European Shortlist, the most important causes of death of the European Union are included. These diseases and events are chosen based on the mortality patterns in the European Union, the relevance for national health programs, the relevance for mortality at regional level, special interest in mortality trends and mortality prognoses and high request of users. In addition, the comparability with ICD-8, ICD-9 and ICD-10 is guaranteed, making longitudinal and international studies possible.
It is concluded that the ICD-codes may be applied either by hand or by using the Beldo list or the European shortlist, depending on the research question. In addition, the selection of the method needs to be carefully thought through since none of the methods seems to be ideal.
Previous use in LASA
Mortality data have been used in numerous LASA studies (e.g. Wijnhoven et al 2010; Holwerda et al., 2007; De Jongh et al 2011). Cause of death has been studied by Van Bunderen and colleagues (2010). They showed an increased risk of all-cause mortality for older persons with IGF-I values in the lowest quintile as compared to the middle quintile. A more than 2-fold increased risk of CVD mortality was revealed for both low-normal and high-normal IGF-I values.
- Boo, A.J. de, Bijlsma, F. & Hoogenboezem, J. (1998). Sterfte in 1996 naar de doodsoorzaak volgens ICD-10 (EN: Mortality in 1996 according to cause of death by ICD-10). Maandbericht Gezondheidsstatistiek, august, 4-11.
- CBS (1993). Wijzigingen in presentatie van demografische gegevens en sterfte naar doodsoorzaak (EN: Alterations in the presentation of demographic data and cause of death). Maandbericht Gezondheidsstatistiek, juli 1993, 34–35.
- Eurostat (1998). Eurostat Working Paper on Population and Social Conditions, nr 22, Causes of Death 1994-1995 statistics. Eurostat, Luxemburg.
- Holwerda, T.J., Schoevers, R.A., Dekker, J., Deeg, D.J.H., Jonker, C., Beekman, A.T.F. (2007). The relationship between generalized anxiety disorder, depression and mortality in old age. International Journal of Geriatric Psychiatry, 22, 241-249.
- De Jongh, R., Lips, P., Rijs, K.J., Van Schoor, N.M., Kramer, M.H.H., Vandenbroucke, J.P., Dekkers, O.M. (2011). Associations between vitamin D receptor genotypes and mortality in a cohort of older Dutch individuals. Journal of Endocrinology, 164 (1), 75-82.
- Schalk, B.W.M., Visser, M., Bremmer, M.A., Penninx, B.W.J.H., Bouter, L.M., and Deeg, D.J.H. (2006). Change of Serum Albumin and Risk of Cardiovascular Disease and All-Cause Mortality. Longitudinal Aging Study Amsterdam. American Journal of Epidemiology, 164(10), 969-977.
- van Bunderen, C.C., van Nieuwpoort, I.C., van Schoor , N.M., Deeg, D.J.H., Lips, P.T.A., Drent, M.L. (2010). The association of serum insulin-like growth factor-I with mortality, cardiovascular disease, and cancer in the elderly: a population-based study. The Journal of Clinical Endocrinology & Metabolism, 95, 4616-4624.
- Verweij, G. (2002). De lijst van belangrijke doodsoorzaken nader bekeken. (EN: A closer look at the list of important causes of death.) In: Maandstatistiek van de bevolking, Jaargang 50 – maart 2002.
- Wijnhoven, H.A.H., van Bokhorst-de van der Schueren, M.A.E., Heymans, M.W., de Vet, H.C.W., Kruizenga, H.M., Twisk, J.W.R., Visser, M. (2010). Low mid-upper arm circumference, calf circumference, and Body Mass Index and mortality in older persons. Journal of Gerontology: Medical Sciences.
- WHO, 1977, International Classification of Diseases. 1975 revision. Volume 1. Manual of the international statistical classification of diseases, injuries, and causes of death. World Health Organization, Genève.
- WHO, 1992, International Statistical Classification of Diseases and Related Health Problems. Tenth Revision. World Health Organization, Genève.