575193 research-article2015
SJP0010.1177/1403494815575193A. Erlangsen and I. FedyszynDanish population-based registers
Scandinavian Journal of Public Health, 2015; 43: 333–339
Review Article
Danish nationwide registers for public health and health-related research
Annette Erlangsen1,2 & Izabela Fedyszyn1 1Suicide
Prevention Research, Research Unit, Mental Health Centre Copenhagen, Capital Region of Denmark, Denmark, and 2Department of Mental Health, Johns Hopkins School of Public Health, Baltimore, MD, USA
Abstract Aims: The Nordic countries have a strong tradition of using nationwide social and health registers for research purposes. The aim of the current paper is to provide an overview of the Danish population-based registers in public health and healthrelated research, and to discuss their strengths and limitations. Methods: Danish registers on somatic and psychiatric hospital contacts as well as care provided by general practitioners were reviewed. The availability of demographic, individual-level variables of relevance for health-related research was summarized. Results: Since 1968, every person living in Denmark has a unique identifier. This identifier is listed in Danish registers enabling linkage of information from a range of registers on an individual level. The nationwide coverage of all patient contacts at somatic and psychiatric hospitals, consultations with general practitioners, purchases of prescribed medications, and a complete follow-up with respect to causes of death support public health studies surveying trends of prevalence and incidence. Historical data on psychiatric and somatic hospitalizations since 1969 and 1977, respectively, allow an in-depth assessment of the burden of disease and time trends. Demographic characteristics of individuals and family units, together with information on education, employment, income, place of residence and migration, are provided by social registers. Conclusions: Register data are fully representative of the entire population with no loss to follow-up. Nationwide coverage also ensures a large sample to investigate events and conditions with low base rates. Clinical observations are limited and often only available for select patient populations. However, the opportunities available for public health research through linkage of register data with the increasing number of nationwide clinical databases, bio-banks and surveys entail promising perspectives for future research. Key Words: Register data, public health, mental health, population-based study
Introduction
Population registers
Nordic nationwide social and health registers have been established largely for administrative purposes. However, they have since become a vital data source and are now extensively used for research in the Nordic countries [1,2]. The aim of the current paper is to provide an overview of the type of information most commonly retrieved from the Danish population-based registers for public health and healthrelated research, and to discuss the strengths as well as limitations of the register data.
One person, one number The Danish history of collecting population data dates back to 1769 and the first population census [3]. National registration of all Danish residents was introduced in 1924, when information pertaining to members of each family was recorded and updated manually on index cards [3]. As a means of lessening the administrative burden of maintaining the register, in 1968, the Danish parliament decided to employ
Correspondence: Annette Erlangsen, Mental Health Centre Copenhagen, Bispebjerg Bakke 23, indgang 13A, 3.sal, 2400 Copenhagen NV, Denmark. E-mail: [email protected] (Accepted 6 February 2015) © 2015 the Nordic Societies of Public Health DOI: 10.1177/1403494815575193
334 A. Erlangsen and I. Fedyszyn electronic recording of the population, and the Civil Registration System (CRS) was introduced, including a unique identifier known as the CRS number [4]. The CRS number consists of the person’s birth date and a four-digit number, where the last digit indicates the gender of the holder; an odd number denotes male gender and an even number refers to female gender. Other digits provide information on the century of birth [5]. In October 1968, all individuals living in Denmark received their CRS number in the mail [4]. Since then, every new member of the population in Denmark has been assigned the CRS number, either at birth or upon first immigration into the country [3]. The number is unique; thus, migrants who leave and re-enter Denmark retain their original CRS number. Under rare circumstances, a person might be given a new CRS number, for example after identity theft, or following gender reassignment surgery, although the majority of changes to the CRS number are related to errors in the original number [6]. The CRS includes records of the CRS numbers together with name, gender, the date and place of birth, citizenship, nationality of both parents, continuously updated residential address, migration in and out of Denmark, and marital status with relevant dates. The Danish CRS was possibly the first of the Nordic countries to implement a distinction between households at the same address, i.e. people living in apartments in the same house number. By listing a floor number and an apartment designation, persons living in the same household could be linked, facilitating identification of, for instance, cohabiting couples [3,7]. The validity of the CRS is considered to be good because the CRS numbers are used on a regular basis in daily life. Linkage of information from public and private authorities via the CRS number further encourages people living in Denmark to ensure that the CRS has their current address; among others, the CRS advises authorities where to direct by post salary and bank statements, tax forms, invoices and household bills. Change of address is, by law, required to be reported within 2 weeks of moving [8]. Nowadays, this can be done either in person or online. Moreover, any errors pertaining to individual information that might be detected by municipal case workers are corrected retrospectively [4]. Family demographics The CRS lists important data on family units. Information on the person’s marital status has been updated on a continuous basis since the register was established in 1968 [3] and, from 1989, it has incorporated same-sex unions, i.e. registered partnerships [9]. Data on children and siblings are available for younger birth cohorts in Denmark. Children can be
linked to legal parents based on the record of them living in the same household, which has been listed in the CRS since 1968. Children born in 1953 or later would have been 15 years or younger in 1968 and, thus, still living with their legal parents at the time when the CRS number was introduced. It is estimated that 98.7% and 95.7% of children born in 1960 can be linked to their legal mother and father, respectively [3]. For children born after 1969, the linkage is almost perfect for both parents [3]. These data are limited to persons born in Denmark. The Lexis diagram in Figure 1 illustrates the population that is available for research using register data also in terms of linkage to parents. Biological mothers can be identified through the Medical Birth Register for children born after 1973, whereas biological fathers were included from 1991 onwards [10]. Siblings can be identified through the CRS by using parental CSR numbers [3]. It is estimated that for women born after 1935, complete linkage of her children as siblings is possible [3]. Further information relevant to family demographics can be accessed from other registers, such as data on adoptions since 1924 [10] and twins born since 1930 [11]. However, only persons who were alive in 1968 were assigned the CRS number and can be linked to information from other registers [11]. Information on social and economic variables Danish nationwide registers enable access to a wide range of information on social and economic variables that can be examined in the aetiology, course and outcomes of different disorders. While most variables used in health-related studies are individual-level data, e.g. education, income, savings and employment status, there is also a range of contextual variables to be found on different levels, e.g. household, municipal, company or regional level. Individual-level register data on education are estimated to be available for 97% of people born after 1945 and are maintained by the Ministry of Education across several databases, with the Population’s Education Register being most widely used [12]. Each year, educational institutions provide the Ministry of Education individual-level information on the enrolment status, completed level of education and exam results for all students [12]. High-quality information concerning individual salaries, capital income, taxes, as well as benefits (sickness and unemployment) and pensions (disability, old age and private pension contributions) can be sourced from the Income Statistics Register, which was established in 1970 [13]. All information can furthermore be linked with history of residence. Employment status and job category are recorded prospectively in different labour market registers [14].
Danish population-based registers 335
Figure 1. Lexi diagram of linkage between parents and child.
Given that the data collection is longitudinal, it is possible to study changes in employment over time for each individual. Based on these and other data, the socio-economic status is coded on a yearly basis for each individual in the population. Affiliation by employees to employer can be identified through an employer ID while companies are classified using industrial codes [14,15], hereby enabling complete studies of people employed in specific companies or sectors as well as people’s job changes. Contextual data, such as regional unemployment rates, municipal tax levels, regional number of hospital beds, rates of diseases by job code can be obtained from specific or combined registers; many of these are available through Statistics Denmark’s registers [16]. Health registers Denmark has a free public health care system [17]. Primary care is provided by general practitioners who can refer patients for tests, treatments, hospitalizations and to specialist care providers [18]. General practitioners are the patients’ first point of contact with the health care system, except for emergencies when the emergency phone line and emergency departments take over this role.
the number of consultations provided to each person on a weekly basis. These data have been captured by the National Health Service Register since 1990 [19]. All types of consultations are listed, including personal contacts at the clinic, home visits, consultations by phone and email (the last one was introduced in 2003) [19]. Dentists and physiotherapists, as well as psychiatrists and psychologists in private practice, also report the number of individual treatments to the National Health Service Register, but only for treatments that are subsidized by the state. This is determined by the guidelines from the Danish National Board of Health, which specify, for example, when it is appropriate for the general practitioners to refer a person to see a psychologist [20]. All treatments provided to children below the age of 16 years used to be recorded under the CRS number of one of their parents, but since 1996 they are listed under their own CRS number [19]. Since 2006, electronic patient records (e-journals) have replaced the paper copies of patient medical histories [21]. The e-journals can be written in only by the general practitioners and hospital medical staff. Currently, the e-journals are not widely used in public health or health-related research, but could constitute a useful data source for future studies.
Contacts with the general practitioners
Somatic hospital treatment
With the purpose of reimbursement, the general practitioners report to the National Board of Health
All persons admitted to somatic hospitals as inpatients, part-time patients, outpatients and emergency
336 A. Erlangsen and I. Fedyszyn department patients are recorded on an individual basis in the National Patient Register. Dating back to 1977, this appears to be the oldest somatic hospital register in the Nordic countries [22,23]. Dates of admission and discharge are listed along with the diagnoses and an identifier for the hospital ward where the patient was treated. Before 1994, the diagnoses were recorded according to the eighth revision of the International Classification of Diseases (ICD8) and since then, the tenth revision of ICD has been employed [24,25]. The ninth revision was never used in Denmark. Both the main and the supplementary diagnoses are included, and transfers to other hospital wards are generally recorded as new entries. Contacts with emergency departments and outpatient treatments as well as performed procedures have been included in the register from 1995 onwards [26]. The National Patient Register is considered to be a reliable source of information on somatic hospital treatment. Since admission can be linked to individuals, the register data are used to address research questions both on the person-level as well as pertaining to hospital sectors. However, an audit of the patient diagnoses using the ICD-8 from different hospital wards found that the three-digit level of the hierarchical diagnosis system is more reliable than the four- and five-digit diagnoses [23]. More recent studies of validity for specific disorders have evaluated the registration favourably, for instance with respect to chronic obstructive pulmonary disorders, epilepsy and febrile seizures [27–29]. Private hospital treatment Although public health care is free for Danish residents and citizens, there may be waiting times for certain treatments [17]. Individuals and companies on behalf of their employees may therefore choose to purchase health insurance that covers treatment in private hospitals [22]. A study from 2005 reported that approximately 1% of hospital beds are located in private hospitals [18]. Since 2003, all private hospitals are required to provide data on treated patients to the National Patient Register, although it has taken some years for all treatments from the private health sector to be reported [22]. Psychiatric hospital treatment The CRS numbers of individuals seeking psychiatric treatment have been listed in the Psychiatric Central Research Register since 1969 for the entire country along with an array of information on the provided care, such as admission date, discharge date, main diagnosis, supplementary diagnoses and patient type
[30]. As part of public health care, psychiatric hospitalization and psychiatric outpatient treatment are free of charge. However, self-referral is generally only possible when one seeks help for mental health concerns at emergency departments. Other forms of psychiatric treatment require referrals from primary or specialist health care providers. As in the National Patient Register, all inpatient treatments and partial hospitalizations have been included since the beginning of the Psychiatric Central Register, while emergency department and outpatient contacts were added in 1995 [31]. In addition, ICD-10 diagnostic codes have been utilized since 1994, replacing the ICD-8 system. Among information listed in the register are also supplementary information, such as the reason for referral, and, an indication of whether incidents of self-harm occurred prior to or during hospitalization [31]. Studies on specific psychiatric disorders have indicated a good validity of the diagnoses listed in the register [31,32]. Prescribed medication Certain medications, including antidepressants and antipsychotics, are only available on prescription in Denmark [33]. All prescriptions redeemed at Danish pharmacies are recorded in the National Prescription Register [33,34]. This practice commenced in 1995, following trial versions of the register first established on the Island of Funen and in Northern Jutland in 1992 [33]. The National Prescription Register contains information on the CRS number; the date when medication was redeemed; the active agents, coded according to the Anatomical Therapeutic Chemical code [35]; the number of pills; the strength of the drug; and an identifier for the issuing general practitioner [34]. Nationwide clinical databases All of the above-described registers were established for administrative purposes and have been made accessible for research. In addition to these, national clinical databases have been created with the purpose of surveying health trends and treatment progress as well as for research. Since 2001, the Danish National Board of Health has approved, and supervised, 50+ nationwide clinical databases [36,37]. The clinical databases are typically initiated and maintained by research units of different hospitals, and their overarching aim is to monitor and improve the quality of clinical care [36]. Although the differences between the health registers and the clinical databases are at times blurry, the latter tend to
Danish population-based registers 337 contain supplementary information about diagnostic evaluations, treatments and outcomes [36]. Clinical databases concern specific diseases, disorders or conditions (e.g. chronic obstructive pulmonary disease, schizophrenia, hip fracture), interventions (e.g. interventions for acute stroke or diabetes) and activities (e.g. information on all transfusions or anaesthetic procedures) [36]. Because the data in the nationwide clinical databases are recorded using the patients’ CRS numbers, these can be linked to the nationwide register data. In addition, there exist numerous clinical databases that are regional or local [36]. Causes of death The Register of Causes of Death dates back to 1970 [38]. It records the date of death and distinguishes between deaths from natural causes, accidents, suicide, homicide and unknown causes. Specific information on the cause of death is listed under the primary, secondary and tertiary causes of death, which are classified according to the ICD system. Similar to the hospital registers; the Register of Causes of Death used ICD-8 codes until 1994, followed by ICD-10. The data are based on the information from the death certificates, which are issued by a medical doctor or by the regional central doctor’s office. The latter certifies deaths attributed to accidents, suicide, homicide or when the cause is unknown. Death certificates are nowadays submitted electronically to the Danish National Board of Health which is responsible for maintaining the Register of Causes of Death [39]. Strengths and limitations of register data Strengths Nationwide registers bring a number of strengths to the Danish public health and health-related research. The significance of having a unique CRS number for the entire nation is evident. Using the CRS number included in each of the registers, data from different sources can be linked on an individual level. Another major advantage is that studies are representative of the entire population with no loss to follow-up. An additional advantage is that nationwide coverage ensures large samples to investigate events and conditions with relatively low base rates (e.g. suicide, Huntington’s disease). The detailed information obtained through hospital admission dates allows for large scale cohort studies with longitudinal followups that can span over decades, and for multivariate analyses with time-varying covariates [40,41]. Also, the data are retrospectively collected and not depending on consent of participants. It is collected
in a uniform manner for all individuals. Although vehement security restrictions and the need for permissions to conduct research using register data, it is relatively simple for researchers with affiliations to Danish institutions to obtain access to data. Limitations Although the registers capture a substantial amount of information on the entire population, the available information on an individual level is limited, for example weight, body mass index, blood pressure, scores on depression and anxiety measures or personality traits. It is not feasible to collect such data in the current administrative setting of the registers apart from clinical samples. However, since such information might be interesting from a research and prevention perspective, it can be obtained for a subset of the general population from different population-based sample studies, allowing for a partial linkage with register data. In terms of illnesses and diseases, diagnostic information is only available from hospital records [26]. At current, there is no registration by general practitioner of diagnosis or reason for contact at the provided consultation, which is a major drawback from a research perspective. Consequently, diseases of people who present to, and are treated exclusively in primary care settings, are not captured in the registers. Some studies have addressed this limitation by using data on redeemed prescriptions from the Danish Prescription Register as an indicator for patient contact with a general practitioner. This is on the grounds that a consultation would be required in order to issue a prescription [42]. Date of onset also has to be interpreted with caution, both for somatic and psychiatric disorders; for this piece of information the accuracy of the register data largely depends on the probability of a hospital contact shortly after onset of the disorder. Coronary heart disease, manifested through heart attacks or heart failure, as well as other acute conditions are likely to lead to immediate hospitalization and, consequently, prove to have good and valid registrations for date of onset [43]. While psychiatric disorders, such as first psychosis, it is generally agreed that hospital treatment is likely to be initiated within a substantial longer period of time. [44] On the other hand, date of first onset could escape registration entirely for disorders, which might first be treated by the general practitioner. Examples for this are chronic pain, depression or dementia where the date of onset could escape registration and require sample data for studies [26,45]. While the CRS lists information on the country of birth as well as citizenship, there is no record of
338 A. Erlangsen and I. Fedyszyn ethnicity. It seems there are no easy solutions to obtain complete data on ethnicity as; self-reporting might entail other bias. Using the register data, one may apply assumptions as that people who are born abroad, whose parents are born abroad or with a different citizenship may belong to a different ethnic group. As Denmark previously did not allow for multiple citizenships after the age of 18, second-generation immigrants might have chosen to have a Danish citizenship. Thus, in order to study this generation of immigrants, one would have to identify the group based on parental country of birth or citizenship. Privacy concerns are taken seriously In order to obtain data extracts from the Danish registers for research, approval from the Danish Data Protection Agency is required. Certain registers, such as the Register of Causes of Death, necessitates an additional approval from the administrator, in this case the ‘Statens Serum Institut’. Data extracts can be obtained either through Statistics Denmark who can provide access to via a secure web portal, or at the institution. Under the current legislation, only researchers living in Denmark or affiliated with Danish institutions are allowed to access individual level data. This has put a barrier to international studies, as these would require merging data sets from across countries. In theory, such projects could be carried out if the register data from collaborating parties were stored and analysed in Denmark (e.g. at Statistics Denmark). What the future entails The number of electronic records on each individual is increasing over time, as are the options for computational research and Internet-based data collections. In Denmark, the focus of a new, large research study are genetic markers, derived from heel-blood samples of children born after 1982, as predictors and moderators of mental disorders [46,47]. Electronic patient journals and e-journals stored centrally allow general practitioners and hospital staff to access and record information on a treated patient in a joint e-journal. Using text-scanning tools, researchers would be able to collect more sensitive data material on patients over future years, which also could help improve diagnostic criteria. There are many relatively unexplored adjacent data collections, which could be explored. For instance, all men aged 18 years and over living in Denmark are obliged to present for conscription including various physical and psychological exams. Although data are collected, so far this information seems to have been used for research aims at a local rather than national level [48,49]. Another example
is local individual-level databases, such as municipal services of care for older and frail individuals in Denmark, which currently are not collected nationally. Here, the reporting to national register takes place on an aggregate level, i.e. the number of services provided by each municipality [50]. Given that trends in comorbidity of the growing and longer living cohorts are essential information for much of tomorrow’s health care [51], this area will likely be further explored in the years to come. Conclusion Danish register data are a valuable source for public health and health-related research. Newer research opportunities, in terms of linkage with clinical databases and bio-banks ensure that register data will be an important source of information also in the years to come. Acknowledgements None. Conflict of interest The authors declare that there is no conflict of interest. Funding This research received no specific grant from any funding agency in the public, commercial or notfor-profit sectors. References [1] Frank L. When an entire country is a cohort. Science 2000;287:2398–9. [2] Rosen M. National Health Data Registers: A Nordic heritage to public health. Scand J Public Health 2002;30:81–5. [3] Pedersen CB, Gøtzsche H, Møller JØ, et al. The Danish Civil Registration System. Dan Med Bull 2006;53:441–9. [4] Nielsen H. CPR - Danmarks Folkeregister [The Centralised Civil Register - The population register of Denmark]. Koebenhavn: Indenrigsministeriet; 1991. [5] Pedersen CB. The Danish Civil Registration System. Scand J Public Health 2011;39:22–5. [6] Det Centrale Person Register [The Civil Register System]. CPR 2014 May 14, https://cpr.dk/borgere/kan-jeg-faa-etnyt-personnummer (accessed 16 February 2015). [7] Standarder for opbygning af navne og adresser i CPR [Standards for listing of names and addresses in CRS]. CPR 2014 May 14, https://cpr.dk/cpr-systemet/tegnsaet-ogstandarder/#standarder (accessed 16 February 2015). [8] Borgerservice: Når du skal flytte[Citizen services: When you are moving]. Borger dk 2014 May 12, https://www.borger.dk/ Sider/Naar-du-skal-flytte.aspx (accessed 16 February 2015). [9] Lov om registreret partnerskab [Law on registered partnership], LOV nr 372 af 07/06/1989, Danish Parliament (1989). [10] Petersen L and Sorensen TI. The Danish Adoption Register. Scand J Public Health 2011;39:83–6. [11] Skytthe A, Kyvik KO, Holm NV, et al. The Danish Twin Registry. Scand J Public Health 2011;39:75–8.
Danish population-based registers 339 [12] Jensen VM and Rasmussen AW. Danish education registers. Scand J Public Health 2011;39:91–4. [13] Baadsgaard M and Quitzau J. Danish registers on personal income and transfer payments. Scand J Public Health 2011;39:103–5. [14] Petersson F, Baadsgaard M and Thygesen LC. Danish registers on personal labour market affiliation. Scand J Public Health 2011;39:95–8. [15] Zeuthen HE, Emerek R, Hansen PV, et al. Integreret Database for Arbejdsmarkedsforskning [The Integrated Database for Longitudinal Labour Market Research]. Statistics Denmark 1990, http://www.dst.dk/upload/hovedrapport.pdf (accessed 30November 2009). [16] Graversen B and Nielsen J. Oversigt over databaser. Med relevans for overvågning, udredning og forskning på det sociale område [Overview over databases with relevance for surveillance, and research of social aspects]. Copenhagen: The Danish National Institute of Social Research; 2001. [17] Davis K. The Danish health system through an American lens. Health Policy 2002;59:119–32. [18] Pedersen KM, Christiansen T and Bech M. The Danish health care system: evolution–not revolution—in a decentralized system. Health Econ 2005;14:S41-S57. [19] Andersen JS, Olivarius NF and Krasnik A. The Danish National Health Service Register. Scand J Public Health 2011;39:34–7. [20] Henvisning til psykolog [Referal to psychologist]. Danish National Board of Health 2014, http://sundhedsstyrelsen. dk/da/sundhed/planlaegning-og-beredskab/praksisomraadet/ psykologer/henvisning-til-psykolog (accessed 16 February 2015). [21] Historien om elektronisk patientjournal (EPD) [The history of the electronic patient records]. Danish Regions 2014 May 5, http://www.regioner.dk/Sundhed/Sundheds-IT/~/media/ Filer/IT%20og%20Kvalitet/Diverse/EPJ%20historien.ashx (accessed 16 February 2015) [22] Lynge E, Sandegaard JL and Rebolj M. The Danish National Patient Register. Scand J Public Health 2011;39:30–3. [23] Mosbech J, J¢rgensen J, Madsen M, et al. Landspatientregisteret. Ugeskr Laeger 1995;157:3741–5. [24] World Health Organization. International Statistical Classification of Diseases and Related Health Problems. 8th revision ed. Geneva: WHO; 1965. [25] World Health Organization. International Statistical Classification of Diseases and Related Health Problems. World Health Organization 2007, http://apps.who.int/classifications/apps/ icd/icd10online/ (accessed 24 November 2009). [26] Andersen TF, Madsen M, Jørgensen J, et al. The Danish National Hospital Register. Dan Med Bull 46, 263–268. 1999. [27] Vestergaard M, Obel C, Henriksen TB, et al. The Danish National Hospital Register is a valuable study base for epidemiologic research in febrile seizures. J Clin Epidemiol 2006;59:61–6. [28] Thomsen RW, Lange P, Hellquist B, et al. Validity and underrecording of diagnosis of COPD in the Danish National Patient Registry. Respir Med 2011;105:1063–8. [29] Christensen J, Vestergaard M, Olsen J, et al. Validation of epilepsy diagnoses in the Danish National Hospital Register. Epilepsy Res 2007;75:162–70. [30] Munk-Jørgensen P and Mortensen PB. The Danish Psychiatric Central Register. Dan Med Bull 1997;44:82–4. [31] Mors O, Perto GP and Mortensen PB. The Danish Psychiatric Central Research Register. Scand J Public Health 2011;39:54–7. [32] Phung TK, Andersen BB, Hogh P, et al. Validity of dementia diagnoses in the Danish hospital registers. Dement Geriatr Cogn Disord 2007;24:220–8.
[33] Gaist D, Sørensen HT and Hallas J. The Danish prescription registers. Dan Med Bull 1997;44:445–8. [34] Kildemoes HW, Sorensen HT and Hallas J. The Danish National Prescription Registry. Scand J Public Health 2011;39:38–41. [35] The WHO Collaborating Centre for Drug Statistics Methodology. Norwegian Institute of Public Health 2008, http:// www.whocc.no/atcddd/ (accessed 25 April 2008). [36] Green A. Danish clinical databases: an overview. Scand J Public Health 2011;39:68–71. [37] Godkendte kliniske kvalitetsdatabaser [Approved clinical quality databases]. Serum Inst 14 A.D. May 13, http:// www.ssi.dk/Sundhedsdataogit/Indberetning%20og%20 patientregistrering/~/media/Indhold/DK%20-%20dansk/ Sundhedsdata%20og%20it/NSF/Indberetning/Kliniske%20kvalitetsdatabaser/ListeOverGodkendteKKDB.ashx (accessed 16 February 2015). [38] Juel K and Helweg-Larsen K. The Danish registers of causes of death. Dan Med Bull 1999;46:354–7. [39] Helweg-Larsen K. The Danish Register of Causes of Death. Scand J Public Health 2011;39:26–9. [40] Qin P, Mortensen PB, Waltoft BL, et al. Allergy is associated with suicide completion with a possible mediating role of mood disorder - a population-based study. Allergy 2011;66:658–64. [41] Agerbo E, Sterne JA and Gunnell DJ. Combining individual and ecological data to determine compositional and contextual socio-economic risk factors for suicide. Soc Sci Med 2007;64:451–61. [42] Andersen UA, Andersen M, Rosholm JU, et al. Contacts to the health care system prior to suicide: a comprehensive analysis using registers for general and psychiatric hospital admissions, contacts to general practitioners and practicing specialists and drug prescriptions. Acta Psychiatr Scand 2000;102:126–34. [43] Baker JL, Olsen LW and Sorensen TI. Childhood bodymass index and the risk of coronary heart disease in adulthood. N Engl J Med 2007;357:2329–37. [44] Thorup A, Petersen L, Jeppesen P, et al. Gender differences in young adults with firstepisode schizophrenia specturm disorders at baseline in the Danish OPUS study. J Nerv Ment Dis 2007;195:396–405. [45] Eriksen J, Jensen MK, Sjogren P, et al. Epidemiology of chronic non-malignant pain in Denmark. Pain 2003;106:221–8. [46] Ipsych. University of Aarhus, Denmark 2014, http://ipsych. au.dk/ [47] The Danish Neonatal Screening Biobank. Serum Institute, Denmark 2014 May 18, http://www.ssi.dk/diagnostik/klinisk %20information/~/media/Indhold/DK%20-%20dansk/ Diagnostik/Klinisk%20information/Blodproeve%20fra %20nyfoedte/The%20Danish%20Neonatal%20Screening %20Biobank%2028092012.ashx (accessed 16 February 20115). [48] Jacobsen N, Jensen H and Goldschmidt E. Prevalence of myopia in Danish conscripts. Acta Ophthalmol Scand 2007;85:165–70. [49] Ehrenstein V, Pedersen L, Grijota M, et al. Association of Apgar score at five minutes with long-term neurologic disability and cognitive function in a prevalence study of Danish conscripts. BMC Pregnancy Childbirth 2009;9:14. [50] Ældreområdet - indikatorer [Indicators - Senior Citizen care]. Statistics Denmark 2014, http://www.dst.dk/da/Statistik/dokumentation/kvalitetsdeklarationer/aeldreomraadet— indikatorer.aspx (accessed 16 February 2015). [51] Christensen K, Doblhammer G, Rau R, et al. Ageing populations: the challenges ahead. Lancet 2009;374:1196–208.
Copyright of Scandinavian Journal of Public Health is the property of Sage Publications, Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
SJP0010.1177/1403494815575193A. Erlangsen and I. FedyszynDanish population-based registers
Scandinavian Journal of Public Health, 2015; 43: 333–339
Review Article
Danish nationwide registers for public health and health-related research
Annette Erlangsen1,2 & Izabela Fedyszyn1 1Suicide
Prevention Research, Research Unit, Mental Health Centre Copenhagen, Capital Region of Denmark, Denmark, and 2Department of Mental Health, Johns Hopkins School of Public Health, Baltimore, MD, USA
Abstract Aims: The Nordic countries have a strong tradition of using nationwide social and health registers for research purposes. The aim of the current paper is to provide an overview of the Danish population-based registers in public health and healthrelated research, and to discuss their strengths and limitations. Methods: Danish registers on somatic and psychiatric hospital contacts as well as care provided by general practitioners were reviewed. The availability of demographic, individual-level variables of relevance for health-related research was summarized. Results: Since 1968, every person living in Denmark has a unique identifier. This identifier is listed in Danish registers enabling linkage of information from a range of registers on an individual level. The nationwide coverage of all patient contacts at somatic and psychiatric hospitals, consultations with general practitioners, purchases of prescribed medications, and a complete follow-up with respect to causes of death support public health studies surveying trends of prevalence and incidence. Historical data on psychiatric and somatic hospitalizations since 1969 and 1977, respectively, allow an in-depth assessment of the burden of disease and time trends. Demographic characteristics of individuals and family units, together with information on education, employment, income, place of residence and migration, are provided by social registers. Conclusions: Register data are fully representative of the entire population with no loss to follow-up. Nationwide coverage also ensures a large sample to investigate events and conditions with low base rates. Clinical observations are limited and often only available for select patient populations. However, the opportunities available for public health research through linkage of register data with the increasing number of nationwide clinical databases, bio-banks and surveys entail promising perspectives for future research. Key Words: Register data, public health, mental health, population-based study
Introduction
Population registers
Nordic nationwide social and health registers have been established largely for administrative purposes. However, they have since become a vital data source and are now extensively used for research in the Nordic countries [1,2]. The aim of the current paper is to provide an overview of the type of information most commonly retrieved from the Danish population-based registers for public health and healthrelated research, and to discuss the strengths as well as limitations of the register data.
One person, one number The Danish history of collecting population data dates back to 1769 and the first population census [3]. National registration of all Danish residents was introduced in 1924, when information pertaining to members of each family was recorded and updated manually on index cards [3]. As a means of lessening the administrative burden of maintaining the register, in 1968, the Danish parliament decided to employ
Correspondence: Annette Erlangsen, Mental Health Centre Copenhagen, Bispebjerg Bakke 23, indgang 13A, 3.sal, 2400 Copenhagen NV, Denmark. E-mail: [email protected] (Accepted 6 February 2015) © 2015 the Nordic Societies of Public Health DOI: 10.1177/1403494815575193
334 A. Erlangsen and I. Fedyszyn electronic recording of the population, and the Civil Registration System (CRS) was introduced, including a unique identifier known as the CRS number [4]. The CRS number consists of the person’s birth date and a four-digit number, where the last digit indicates the gender of the holder; an odd number denotes male gender and an even number refers to female gender. Other digits provide information on the century of birth [5]. In October 1968, all individuals living in Denmark received their CRS number in the mail [4]. Since then, every new member of the population in Denmark has been assigned the CRS number, either at birth or upon first immigration into the country [3]. The number is unique; thus, migrants who leave and re-enter Denmark retain their original CRS number. Under rare circumstances, a person might be given a new CRS number, for example after identity theft, or following gender reassignment surgery, although the majority of changes to the CRS number are related to errors in the original number [6]. The CRS includes records of the CRS numbers together with name, gender, the date and place of birth, citizenship, nationality of both parents, continuously updated residential address, migration in and out of Denmark, and marital status with relevant dates. The Danish CRS was possibly the first of the Nordic countries to implement a distinction between households at the same address, i.e. people living in apartments in the same house number. By listing a floor number and an apartment designation, persons living in the same household could be linked, facilitating identification of, for instance, cohabiting couples [3,7]. The validity of the CRS is considered to be good because the CRS numbers are used on a regular basis in daily life. Linkage of information from public and private authorities via the CRS number further encourages people living in Denmark to ensure that the CRS has their current address; among others, the CRS advises authorities where to direct by post salary and bank statements, tax forms, invoices and household bills. Change of address is, by law, required to be reported within 2 weeks of moving [8]. Nowadays, this can be done either in person or online. Moreover, any errors pertaining to individual information that might be detected by municipal case workers are corrected retrospectively [4]. Family demographics The CRS lists important data on family units. Information on the person’s marital status has been updated on a continuous basis since the register was established in 1968 [3] and, from 1989, it has incorporated same-sex unions, i.e. registered partnerships [9]. Data on children and siblings are available for younger birth cohorts in Denmark. Children can be
linked to legal parents based on the record of them living in the same household, which has been listed in the CRS since 1968. Children born in 1953 or later would have been 15 years or younger in 1968 and, thus, still living with their legal parents at the time when the CRS number was introduced. It is estimated that 98.7% and 95.7% of children born in 1960 can be linked to their legal mother and father, respectively [3]. For children born after 1969, the linkage is almost perfect for both parents [3]. These data are limited to persons born in Denmark. The Lexis diagram in Figure 1 illustrates the population that is available for research using register data also in terms of linkage to parents. Biological mothers can be identified through the Medical Birth Register for children born after 1973, whereas biological fathers were included from 1991 onwards [10]. Siblings can be identified through the CRS by using parental CSR numbers [3]. It is estimated that for women born after 1935, complete linkage of her children as siblings is possible [3]. Further information relevant to family demographics can be accessed from other registers, such as data on adoptions since 1924 [10] and twins born since 1930 [11]. However, only persons who were alive in 1968 were assigned the CRS number and can be linked to information from other registers [11]. Information on social and economic variables Danish nationwide registers enable access to a wide range of information on social and economic variables that can be examined in the aetiology, course and outcomes of different disorders. While most variables used in health-related studies are individual-level data, e.g. education, income, savings and employment status, there is also a range of contextual variables to be found on different levels, e.g. household, municipal, company or regional level. Individual-level register data on education are estimated to be available for 97% of people born after 1945 and are maintained by the Ministry of Education across several databases, with the Population’s Education Register being most widely used [12]. Each year, educational institutions provide the Ministry of Education individual-level information on the enrolment status, completed level of education and exam results for all students [12]. High-quality information concerning individual salaries, capital income, taxes, as well as benefits (sickness and unemployment) and pensions (disability, old age and private pension contributions) can be sourced from the Income Statistics Register, which was established in 1970 [13]. All information can furthermore be linked with history of residence. Employment status and job category are recorded prospectively in different labour market registers [14].
Danish population-based registers 335
Figure 1. Lexi diagram of linkage between parents and child.
Given that the data collection is longitudinal, it is possible to study changes in employment over time for each individual. Based on these and other data, the socio-economic status is coded on a yearly basis for each individual in the population. Affiliation by employees to employer can be identified through an employer ID while companies are classified using industrial codes [14,15], hereby enabling complete studies of people employed in specific companies or sectors as well as people’s job changes. Contextual data, such as regional unemployment rates, municipal tax levels, regional number of hospital beds, rates of diseases by job code can be obtained from specific or combined registers; many of these are available through Statistics Denmark’s registers [16]. Health registers Denmark has a free public health care system [17]. Primary care is provided by general practitioners who can refer patients for tests, treatments, hospitalizations and to specialist care providers [18]. General practitioners are the patients’ first point of contact with the health care system, except for emergencies when the emergency phone line and emergency departments take over this role.
the number of consultations provided to each person on a weekly basis. These data have been captured by the National Health Service Register since 1990 [19]. All types of consultations are listed, including personal contacts at the clinic, home visits, consultations by phone and email (the last one was introduced in 2003) [19]. Dentists and physiotherapists, as well as psychiatrists and psychologists in private practice, also report the number of individual treatments to the National Health Service Register, but only for treatments that are subsidized by the state. This is determined by the guidelines from the Danish National Board of Health, which specify, for example, when it is appropriate for the general practitioners to refer a person to see a psychologist [20]. All treatments provided to children below the age of 16 years used to be recorded under the CRS number of one of their parents, but since 1996 they are listed under their own CRS number [19]. Since 2006, electronic patient records (e-journals) have replaced the paper copies of patient medical histories [21]. The e-journals can be written in only by the general practitioners and hospital medical staff. Currently, the e-journals are not widely used in public health or health-related research, but could constitute a useful data source for future studies.
Contacts with the general practitioners
Somatic hospital treatment
With the purpose of reimbursement, the general practitioners report to the National Board of Health
All persons admitted to somatic hospitals as inpatients, part-time patients, outpatients and emergency
336 A. Erlangsen and I. Fedyszyn department patients are recorded on an individual basis in the National Patient Register. Dating back to 1977, this appears to be the oldest somatic hospital register in the Nordic countries [22,23]. Dates of admission and discharge are listed along with the diagnoses and an identifier for the hospital ward where the patient was treated. Before 1994, the diagnoses were recorded according to the eighth revision of the International Classification of Diseases (ICD8) and since then, the tenth revision of ICD has been employed [24,25]. The ninth revision was never used in Denmark. Both the main and the supplementary diagnoses are included, and transfers to other hospital wards are generally recorded as new entries. Contacts with emergency departments and outpatient treatments as well as performed procedures have been included in the register from 1995 onwards [26]. The National Patient Register is considered to be a reliable source of information on somatic hospital treatment. Since admission can be linked to individuals, the register data are used to address research questions both on the person-level as well as pertaining to hospital sectors. However, an audit of the patient diagnoses using the ICD-8 from different hospital wards found that the three-digit level of the hierarchical diagnosis system is more reliable than the four- and five-digit diagnoses [23]. More recent studies of validity for specific disorders have evaluated the registration favourably, for instance with respect to chronic obstructive pulmonary disorders, epilepsy and febrile seizures [27–29]. Private hospital treatment Although public health care is free for Danish residents and citizens, there may be waiting times for certain treatments [17]. Individuals and companies on behalf of their employees may therefore choose to purchase health insurance that covers treatment in private hospitals [22]. A study from 2005 reported that approximately 1% of hospital beds are located in private hospitals [18]. Since 2003, all private hospitals are required to provide data on treated patients to the National Patient Register, although it has taken some years for all treatments from the private health sector to be reported [22]. Psychiatric hospital treatment The CRS numbers of individuals seeking psychiatric treatment have been listed in the Psychiatric Central Research Register since 1969 for the entire country along with an array of information on the provided care, such as admission date, discharge date, main diagnosis, supplementary diagnoses and patient type
[30]. As part of public health care, psychiatric hospitalization and psychiatric outpatient treatment are free of charge. However, self-referral is generally only possible when one seeks help for mental health concerns at emergency departments. Other forms of psychiatric treatment require referrals from primary or specialist health care providers. As in the National Patient Register, all inpatient treatments and partial hospitalizations have been included since the beginning of the Psychiatric Central Register, while emergency department and outpatient contacts were added in 1995 [31]. In addition, ICD-10 diagnostic codes have been utilized since 1994, replacing the ICD-8 system. Among information listed in the register are also supplementary information, such as the reason for referral, and, an indication of whether incidents of self-harm occurred prior to or during hospitalization [31]. Studies on specific psychiatric disorders have indicated a good validity of the diagnoses listed in the register [31,32]. Prescribed medication Certain medications, including antidepressants and antipsychotics, are only available on prescription in Denmark [33]. All prescriptions redeemed at Danish pharmacies are recorded in the National Prescription Register [33,34]. This practice commenced in 1995, following trial versions of the register first established on the Island of Funen and in Northern Jutland in 1992 [33]. The National Prescription Register contains information on the CRS number; the date when medication was redeemed; the active agents, coded according to the Anatomical Therapeutic Chemical code [35]; the number of pills; the strength of the drug; and an identifier for the issuing general practitioner [34]. Nationwide clinical databases All of the above-described registers were established for administrative purposes and have been made accessible for research. In addition to these, national clinical databases have been created with the purpose of surveying health trends and treatment progress as well as for research. Since 2001, the Danish National Board of Health has approved, and supervised, 50+ nationwide clinical databases [36,37]. The clinical databases are typically initiated and maintained by research units of different hospitals, and their overarching aim is to monitor and improve the quality of clinical care [36]. Although the differences between the health registers and the clinical databases are at times blurry, the latter tend to
Danish population-based registers 337 contain supplementary information about diagnostic evaluations, treatments and outcomes [36]. Clinical databases concern specific diseases, disorders or conditions (e.g. chronic obstructive pulmonary disease, schizophrenia, hip fracture), interventions (e.g. interventions for acute stroke or diabetes) and activities (e.g. information on all transfusions or anaesthetic procedures) [36]. Because the data in the nationwide clinical databases are recorded using the patients’ CRS numbers, these can be linked to the nationwide register data. In addition, there exist numerous clinical databases that are regional or local [36]. Causes of death The Register of Causes of Death dates back to 1970 [38]. It records the date of death and distinguishes between deaths from natural causes, accidents, suicide, homicide and unknown causes. Specific information on the cause of death is listed under the primary, secondary and tertiary causes of death, which are classified according to the ICD system. Similar to the hospital registers; the Register of Causes of Death used ICD-8 codes until 1994, followed by ICD-10. The data are based on the information from the death certificates, which are issued by a medical doctor or by the regional central doctor’s office. The latter certifies deaths attributed to accidents, suicide, homicide or when the cause is unknown. Death certificates are nowadays submitted electronically to the Danish National Board of Health which is responsible for maintaining the Register of Causes of Death [39]. Strengths and limitations of register data Strengths Nationwide registers bring a number of strengths to the Danish public health and health-related research. The significance of having a unique CRS number for the entire nation is evident. Using the CRS number included in each of the registers, data from different sources can be linked on an individual level. Another major advantage is that studies are representative of the entire population with no loss to follow-up. An additional advantage is that nationwide coverage ensures large samples to investigate events and conditions with relatively low base rates (e.g. suicide, Huntington’s disease). The detailed information obtained through hospital admission dates allows for large scale cohort studies with longitudinal followups that can span over decades, and for multivariate analyses with time-varying covariates [40,41]. Also, the data are retrospectively collected and not depending on consent of participants. It is collected
in a uniform manner for all individuals. Although vehement security restrictions and the need for permissions to conduct research using register data, it is relatively simple for researchers with affiliations to Danish institutions to obtain access to data. Limitations Although the registers capture a substantial amount of information on the entire population, the available information on an individual level is limited, for example weight, body mass index, blood pressure, scores on depression and anxiety measures or personality traits. It is not feasible to collect such data in the current administrative setting of the registers apart from clinical samples. However, since such information might be interesting from a research and prevention perspective, it can be obtained for a subset of the general population from different population-based sample studies, allowing for a partial linkage with register data. In terms of illnesses and diseases, diagnostic information is only available from hospital records [26]. At current, there is no registration by general practitioner of diagnosis or reason for contact at the provided consultation, which is a major drawback from a research perspective. Consequently, diseases of people who present to, and are treated exclusively in primary care settings, are not captured in the registers. Some studies have addressed this limitation by using data on redeemed prescriptions from the Danish Prescription Register as an indicator for patient contact with a general practitioner. This is on the grounds that a consultation would be required in order to issue a prescription [42]. Date of onset also has to be interpreted with caution, both for somatic and psychiatric disorders; for this piece of information the accuracy of the register data largely depends on the probability of a hospital contact shortly after onset of the disorder. Coronary heart disease, manifested through heart attacks or heart failure, as well as other acute conditions are likely to lead to immediate hospitalization and, consequently, prove to have good and valid registrations for date of onset [43]. While psychiatric disorders, such as first psychosis, it is generally agreed that hospital treatment is likely to be initiated within a substantial longer period of time. [44] On the other hand, date of first onset could escape registration entirely for disorders, which might first be treated by the general practitioner. Examples for this are chronic pain, depression or dementia where the date of onset could escape registration and require sample data for studies [26,45]. While the CRS lists information on the country of birth as well as citizenship, there is no record of
338 A. Erlangsen and I. Fedyszyn ethnicity. It seems there are no easy solutions to obtain complete data on ethnicity as; self-reporting might entail other bias. Using the register data, one may apply assumptions as that people who are born abroad, whose parents are born abroad or with a different citizenship may belong to a different ethnic group. As Denmark previously did not allow for multiple citizenships after the age of 18, second-generation immigrants might have chosen to have a Danish citizenship. Thus, in order to study this generation of immigrants, one would have to identify the group based on parental country of birth or citizenship. Privacy concerns are taken seriously In order to obtain data extracts from the Danish registers for research, approval from the Danish Data Protection Agency is required. Certain registers, such as the Register of Causes of Death, necessitates an additional approval from the administrator, in this case the ‘Statens Serum Institut’. Data extracts can be obtained either through Statistics Denmark who can provide access to via a secure web portal, or at the institution. Under the current legislation, only researchers living in Denmark or affiliated with Danish institutions are allowed to access individual level data. This has put a barrier to international studies, as these would require merging data sets from across countries. In theory, such projects could be carried out if the register data from collaborating parties were stored and analysed in Denmark (e.g. at Statistics Denmark). What the future entails The number of electronic records on each individual is increasing over time, as are the options for computational research and Internet-based data collections. In Denmark, the focus of a new, large research study are genetic markers, derived from heel-blood samples of children born after 1982, as predictors and moderators of mental disorders [46,47]. Electronic patient journals and e-journals stored centrally allow general practitioners and hospital staff to access and record information on a treated patient in a joint e-journal. Using text-scanning tools, researchers would be able to collect more sensitive data material on patients over future years, which also could help improve diagnostic criteria. There are many relatively unexplored adjacent data collections, which could be explored. For instance, all men aged 18 years and over living in Denmark are obliged to present for conscription including various physical and psychological exams. Although data are collected, so far this information seems to have been used for research aims at a local rather than national level [48,49]. Another example
is local individual-level databases, such as municipal services of care for older and frail individuals in Denmark, which currently are not collected nationally. Here, the reporting to national register takes place on an aggregate level, i.e. the number of services provided by each municipality [50]. Given that trends in comorbidity of the growing and longer living cohorts are essential information for much of tomorrow’s health care [51], this area will likely be further explored in the years to come. Conclusion Danish register data are a valuable source for public health and health-related research. Newer research opportunities, in terms of linkage with clinical databases and bio-banks ensure that register data will be an important source of information also in the years to come. Acknowledgements None. Conflict of interest The authors declare that there is no conflict of interest. Funding This research received no specific grant from any funding agency in the public, commercial or notfor-profit sectors. References [1] Frank L. When an entire country is a cohort. Science 2000;287:2398–9. [2] Rosen M. National Health Data Registers: A Nordic heritage to public health. Scand J Public Health 2002;30:81–5. [3] Pedersen CB, Gøtzsche H, Møller JØ, et al. The Danish Civil Registration System. Dan Med Bull 2006;53:441–9. [4] Nielsen H. CPR - Danmarks Folkeregister [The Centralised Civil Register - The population register of Denmark]. Koebenhavn: Indenrigsministeriet; 1991. [5] Pedersen CB. The Danish Civil Registration System. Scand J Public Health 2011;39:22–5. [6] Det Centrale Person Register [The Civil Register System]. CPR 2014 May 14, https://cpr.dk/borgere/kan-jeg-faa-etnyt-personnummer (accessed 16 February 2015). [7] Standarder for opbygning af navne og adresser i CPR [Standards for listing of names and addresses in CRS]. CPR 2014 May 14, https://cpr.dk/cpr-systemet/tegnsaet-ogstandarder/#standarder (accessed 16 February 2015). [8] Borgerservice: Når du skal flytte[Citizen services: When you are moving]. Borger dk 2014 May 12, https://www.borger.dk/ Sider/Naar-du-skal-flytte.aspx (accessed 16 February 2015). [9] Lov om registreret partnerskab [Law on registered partnership], LOV nr 372 af 07/06/1989, Danish Parliament (1989). [10] Petersen L and Sorensen TI. The Danish Adoption Register. Scand J Public Health 2011;39:83–6. [11] Skytthe A, Kyvik KO, Holm NV, et al. The Danish Twin Registry. Scand J Public Health 2011;39:75–8.
Danish population-based registers 339 [12] Jensen VM and Rasmussen AW. Danish education registers. Scand J Public Health 2011;39:91–4. [13] Baadsgaard M and Quitzau J. Danish registers on personal income and transfer payments. Scand J Public Health 2011;39:103–5. [14] Petersson F, Baadsgaard M and Thygesen LC. Danish registers on personal labour market affiliation. Scand J Public Health 2011;39:95–8. [15] Zeuthen HE, Emerek R, Hansen PV, et al. Integreret Database for Arbejdsmarkedsforskning [The Integrated Database for Longitudinal Labour Market Research]. Statistics Denmark 1990, http://www.dst.dk/upload/hovedrapport.pdf (accessed 30November 2009). [16] Graversen B and Nielsen J. Oversigt over databaser. Med relevans for overvågning, udredning og forskning på det sociale område [Overview over databases with relevance for surveillance, and research of social aspects]. Copenhagen: The Danish National Institute of Social Research; 2001. [17] Davis K. The Danish health system through an American lens. Health Policy 2002;59:119–32. [18] Pedersen KM, Christiansen T and Bech M. The Danish health care system: evolution–not revolution—in a decentralized system. Health Econ 2005;14:S41-S57. [19] Andersen JS, Olivarius NF and Krasnik A. The Danish National Health Service Register. Scand J Public Health 2011;39:34–7. [20] Henvisning til psykolog [Referal to psychologist]. Danish National Board of Health 2014, http://sundhedsstyrelsen. dk/da/sundhed/planlaegning-og-beredskab/praksisomraadet/ psykologer/henvisning-til-psykolog (accessed 16 February 2015). [21] Historien om elektronisk patientjournal (EPD) [The history of the electronic patient records]. Danish Regions 2014 May 5, http://www.regioner.dk/Sundhed/Sundheds-IT/~/media/ Filer/IT%20og%20Kvalitet/Diverse/EPJ%20historien.ashx (accessed 16 February 2015) [22] Lynge E, Sandegaard JL and Rebolj M. The Danish National Patient Register. Scand J Public Health 2011;39:30–3. [23] Mosbech J, J¢rgensen J, Madsen M, et al. Landspatientregisteret. Ugeskr Laeger 1995;157:3741–5. [24] World Health Organization. International Statistical Classification of Diseases and Related Health Problems. 8th revision ed. Geneva: WHO; 1965. [25] World Health Organization. International Statistical Classification of Diseases and Related Health Problems. World Health Organization 2007, http://apps.who.int/classifications/apps/ icd/icd10online/ (accessed 24 November 2009). [26] Andersen TF, Madsen M, Jørgensen J, et al. The Danish National Hospital Register. Dan Med Bull 46, 263–268. 1999. [27] Vestergaard M, Obel C, Henriksen TB, et al. The Danish National Hospital Register is a valuable study base for epidemiologic research in febrile seizures. J Clin Epidemiol 2006;59:61–6. [28] Thomsen RW, Lange P, Hellquist B, et al. Validity and underrecording of diagnosis of COPD in the Danish National Patient Registry. Respir Med 2011;105:1063–8. [29] Christensen J, Vestergaard M, Olsen J, et al. Validation of epilepsy diagnoses in the Danish National Hospital Register. Epilepsy Res 2007;75:162–70. [30] Munk-Jørgensen P and Mortensen PB. The Danish Psychiatric Central Register. Dan Med Bull 1997;44:82–4. [31] Mors O, Perto GP and Mortensen PB. The Danish Psychiatric Central Research Register. Scand J Public Health 2011;39:54–7. [32] Phung TK, Andersen BB, Hogh P, et al. Validity of dementia diagnoses in the Danish hospital registers. Dement Geriatr Cogn Disord 2007;24:220–8.
[33] Gaist D, Sørensen HT and Hallas J. The Danish prescription registers. Dan Med Bull 1997;44:445–8. [34] Kildemoes HW, Sorensen HT and Hallas J. The Danish National Prescription Registry. Scand J Public Health 2011;39:38–41. [35] The WHO Collaborating Centre for Drug Statistics Methodology. Norwegian Institute of Public Health 2008, http:// www.whocc.no/atcddd/ (accessed 25 April 2008). [36] Green A. Danish clinical databases: an overview. Scand J Public Health 2011;39:68–71. [37] Godkendte kliniske kvalitetsdatabaser [Approved clinical quality databases]. Serum Inst 14 A.D. May 13, http:// www.ssi.dk/Sundhedsdataogit/Indberetning%20og%20 patientregistrering/~/media/Indhold/DK%20-%20dansk/ Sundhedsdata%20og%20it/NSF/Indberetning/Kliniske%20kvalitetsdatabaser/ListeOverGodkendteKKDB.ashx (accessed 16 February 2015). [38] Juel K and Helweg-Larsen K. The Danish registers of causes of death. Dan Med Bull 1999;46:354–7. [39] Helweg-Larsen K. The Danish Register of Causes of Death. Scand J Public Health 2011;39:26–9. [40] Qin P, Mortensen PB, Waltoft BL, et al. Allergy is associated with suicide completion with a possible mediating role of mood disorder - a population-based study. Allergy 2011;66:658–64. [41] Agerbo E, Sterne JA and Gunnell DJ. Combining individual and ecological data to determine compositional and contextual socio-economic risk factors for suicide. Soc Sci Med 2007;64:451–61. [42] Andersen UA, Andersen M, Rosholm JU, et al. Contacts to the health care system prior to suicide: a comprehensive analysis using registers for general and psychiatric hospital admissions, contacts to general practitioners and practicing specialists and drug prescriptions. Acta Psychiatr Scand 2000;102:126–34. [43] Baker JL, Olsen LW and Sorensen TI. Childhood bodymass index and the risk of coronary heart disease in adulthood. N Engl J Med 2007;357:2329–37. [44] Thorup A, Petersen L, Jeppesen P, et al. Gender differences in young adults with firstepisode schizophrenia specturm disorders at baseline in the Danish OPUS study. J Nerv Ment Dis 2007;195:396–405. [45] Eriksen J, Jensen MK, Sjogren P, et al. Epidemiology of chronic non-malignant pain in Denmark. Pain 2003;106:221–8. [46] Ipsych. University of Aarhus, Denmark 2014, http://ipsych. au.dk/ [47] The Danish Neonatal Screening Biobank. Serum Institute, Denmark 2014 May 18, http://www.ssi.dk/diagnostik/klinisk %20information/~/media/Indhold/DK%20-%20dansk/ Diagnostik/Klinisk%20information/Blodproeve%20fra %20nyfoedte/The%20Danish%20Neonatal%20Screening %20Biobank%2028092012.ashx (accessed 16 February 20115). [48] Jacobsen N, Jensen H and Goldschmidt E. Prevalence of myopia in Danish conscripts. Acta Ophthalmol Scand 2007;85:165–70. [49] Ehrenstein V, Pedersen L, Grijota M, et al. Association of Apgar score at five minutes with long-term neurologic disability and cognitive function in a prevalence study of Danish conscripts. BMC Pregnancy Childbirth 2009;9:14. [50] Ældreområdet - indikatorer [Indicators - Senior Citizen care]. Statistics Denmark 2014, http://www.dst.dk/da/Statistik/dokumentation/kvalitetsdeklarationer/aeldreomraadet— indikatorer.aspx (accessed 16 February 2015). [51] Christensen K, Doblhammer G, Rau R, et al. Ageing populations: the challenges ahead. Lancet 2009;374:1196–208.
Copyright of Scandinavian Journal of Public Health is the property of Sage Publications, Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
