The Swedish National Inpatient Register (IPR), also called the Hospital Discharge Register, is a principal source of data for numerous research projects. The IPR is part of the National Patient Register. The Swedish IPR was launched in 1964 (psychiatric diagnoses from 1973) but complete coverage did not begin until 1987. Currently, more than 99% of all somatic (including surgery) and psychiatric hospital discharges are registered in the IPR. A previous validation of the IPR by the National Board of Health and Welfare showed that 85-95% of all diagnoses in the IPR are valid. The current paper describes the history, structure, coverage and quality of the Swedish IPR.
Methods and results
In January 2010, we searched the medical databases, Medline and HighWire, using the search algorithm "validat* (inpatient or hospital discharge) Sweden". We also contacted 218 members of the Swedish Society of Epidemiology and an additional 201 medical researchers to identify papers that had validated the IPR. In total, 132 papers were reviewed. The positive predictive value (PPV) was found to differ between diagnoses in the IPR, but is generally 85-95%.
Conclusions
In conclusion, the validity of the Swedish IPR is high for many but not all diagnoses. The long follow-up makes the register particularly suitable for large-scale population-based research, but for certain research areas the use of other health registers, such as the Swedish Cancer Register, may be more suitable.
Keywords
Classification of diseasesdiseaseepidemiologymorbidityregister
Background
The Swedish National Inpatient Register (IPR; Swedish: slutenvårdsregistret), also called the Hospital Discharge Register, was established in 1964 (Figure 1). The IPR has complete national coverage since 1987. The IPR is part of the National Patient Register (Swedish: patientregistret). Currently, more than 99% of all somatic and psychiatric hospital discharges are registered in the IPR. Diagnoses in the IPR are coded according to the Swedish international classification of disease (ICD) system, first introduced in 1964 (adapted from the WHO ICD classification system) (Figure 1). A history of the Swedish and Nordic ICD system has been published elsewhere [1]. It is mandatory for all physicians, private and publicly funded, to deliver data to the IPR (except for visits in primary care). A detailed description of the regulations relevant to the IPR has been given in the Appendix (Additional file 1).
Figure 1
Timeline of the Swedish Inpatient Register. Years inside the arrow indicate the first year when an ICD classification was in use. ICD-10 was introduced in 1997, with the exception of the county of Skåne where ICD-9 was still in use throughout 1997. The one-year delay in introducing ICD-10 in Skåne has some implications when identifying patients with a certain disease/disorder in this county because about 8-9% of the Swedish population live in Skåne.
History and coverage of the IPR
The IPR was founded in 1964 when the NBHW (National Board of Health and Welfare; Swedish: Socialstyrelsen) began collecting data on somatic inpatient care in six Swedish counties (roughly the Uppsala region)(Figure 2, red line)[2] (for the population statistics underlying Figures 2 and 3, please see Additional file 2). In fact, the NBHW started to collect data on psychiatric care in 1962 but when the IPR was reconstructed in the 1990s, all psychiatric data originating before 1973 were removed (Figure 3). Beginning in about 1970, data collection for the IPR went from a pilot project to an all-inclusive effort to cover the entire country. In 1983, approximately 85% of all somatic care and almost all psychiatric care were reported to the NBHW [2]. In 1984, the NBHW asked permission from the National Data Inspection Board to link individual data to the personal identity number (PIN) (Swedish: personnummer) [3] of each individual. Although granted permission, the NBHW postponed the introduction of a PIN-based register because the Swedish attorney general objected to the use of the PIN in the IPR. Only in 1993 did the Swedish government declare that the IPR should use the PIN as the unique identifier in all hospital discharges. After 1993, all counties have collaborated on reconstructing earlier hospital discharges linked to the PIN for the years 1984-91. This linkage was possible for all but three counties: two counties were unable to reconstruct data for the year 1985 while the third did not enter the IPR until 1987.
Figure 2
Somatic care: coverage of the Swedish population. Red = Proportion of the Swedish population living in counties that had started to report somatic hospital discharges to the Swedish Inpatient Register. Blue = Proportion of the Swedish population living in counties where all somatic hospital discharges were reported to the Swedish Inpatient Register (1964: 6%; 1972: 36%; 1982: 71%; 1984: 86%). In 1976, for the first time more than 50% of the Swedish population were covered. Complete coverage (100%) was attained in 1987. County population data obtained from the government agency Statistics Sweden (Appendix).
Figure 3
Psychiatric care: coverage of the Swedish population. Blue = Proportion of the Swedish population living in counties where all psychiatric hospital discharges were reported to the Swedish Inpatient Register (1973: 86%; 1985: 94%; 1986: 98%). All counties in Sweden started to record psychiatric care in 1973. (Actually, psychiatric diagnoses were recorded before 1973 but then removed until 1973 - see text). County population data obtained from the government agency Statistics Sweden (Appendix).
Each year, there are about 1.5 million hospital discharges in the IPR (Figure 4), with the majority of these taking place in somatic care. From 1997 and onwards, surgical day care procedures are reported to the NBHW, and since 2001, counties are obliged to report hospital-based outpatient physician visits. However, primary health care data are still not reported on a national level to the NBHW. Whereas coverage of the IPR is currently almost 100%; coverage of hospital-based outpatient care is considerably lower (about 80%)[2]. In the outpatient register, data from private caregivers are missing (coverage of data from public caregivers in outpatient care is almost 100%). The number of hospitals reporting to the IPR increased rapidly in the 1970s. In the 1960s, 20 hospitals and roughly 80 nursing homes reported to the IPR [2]. In the 1980s, the number of units reporting to the IPR had increased to 580. Because of organizational changes, the number of reporting units has since declined.
Figure 4
Number of hospital discharges from 1964-2007[2]. Surgery = General surgery.
IPR variables
IPR variables can be divided into four categories: patient-related data, data about the caregiver, administrative data and medical data (Table 1). Figure 5 displays a typical dataset from the IPR as delivered to researchers.
Table 1
Variables in the Swedish IPR
Variable
Description
Patient-related data
Personal Identity Number (PIN)
Combination of date of birth, three-digit birth number and a check digit [3]. Personal identity number shall be reported for all admissions/discharges, except for induced abortions where it is not registered for legal reasons
Sex
1 = male; 2 = female
Age
Age in years at discharge. In individuals with missing birth dates, the difference between year of discharge and birth year is used to calculate age.
County
The county where the patient has his/her permanent residence (this is not necessarily the county where the patient is admitted).
Municipality and parish
Usually consists of six digits, where positions 1-2 indicate county, 3-4 municipality and 5-6 parish. Individuals living outside Sweden are assigned the value "99". Missing data have been replaced by data from Statistics Sweden.
Data about caregiver (hospital/department)
Hospital
Each hospital in Sweden has a unique 5-digit code assigned by the National Board of Health and Welfare (NBHW)(e.g., Lund University Hospital has code 41001).
Type of department
Each type of department or health centre has a unique code assigned by the NBHW (e.g. ophthalmology departments have code 511)
Administrative data
Admission date
Year-month-day
Discharge date
Year-month-day
Duration of admission
Number of days at hospital. Patients discharged on the day of admission are assigned the value "1".
Elective health care
1 = Yes, 2 = No
Mode of admission
1 = from other hospital/department, 2 = from special living (e.g., home for disabled people, or geriatric care), 3 = other (i.e. from home)
Mode of discharge
1 = to other hospital/department, 2 = to special living (e.g., home for disabled people or geriatric care), 3 = other (i.e. discharged to home), 4 = deceased.
Medical data
Diagnoses
In 1964-1996, the IPR permitted up to 6 diagnoses per discharge. Between 1997-2009 8 diagnoses could be recorded (one of them being the primary diagnosis).
Primary and
additional diagnoses
The primary diagnosis or "main condition" should be the condition diagnosed at the end of the episode of health care responsible for the patient's need for treatment or investigation.
The additional (secondary or contributory diagnoses/conditions) may or may not contribute to the primary diagnosis. They may be co-morbidities and/or complications. Since 2010 the number of possible additional diagnoses per case is unlimited (however, the NBHW will generally only deliver the first 7 additional diagnoses to researchers who request data from the IPR).
External cause of injury or poisoning (E-code) - or "Chapter XX codes".
Until 1997, only one E-code could be recorded per discharge; from 1998, numerous "E-codes" may be recorded. With the introduction of ICD-10 in 1997, E-codes should be referred to as "Chapter XX-codes". (In ICD-10, E00-E99 codes represent metabolic conditions).
Procedures
In 1964 the Swedish NBHW introduced a national classification of procedures based on an American classification of surgical procedures. It had four digit-codes (e.g. appendectomy 4510). Since 1997, a Swedish version of the NOMESCO Classification of Surgical Procedures is in use. This classification is based on five-character alpha-numeric codes (e.g. JEA01 for appendectomy). Current procedures are listed in the Swedish Classification of surgical and medical procedures (Swedish: "KVÅ" - klassifikation av vårdåtgärder)(issued by the NBHW).
Between 1964 and 1996, up to 6 operations/surgical procedures could be listed per discharge. From 1997, up to 12 operations/surgical procedures could be listed per discharge. In the future it will be possible to record more than 12 diagnoses per discharge. Since 2007, all performed procedures are mandatory to record, including medical procedures. The surgeon may also (voluntarily) report date of operation and type of anaesthesia and drugs used according to the ATC list.
Psychiatric care
0 = voluntary care, 1-4: compulsory psychiatric inpatient care (under different conditions or according to certain laws). If a patient has been treated according to categories 1, 2, 3 or 4, the condition prevailing most of the time shall be reported. Compulsory care can be further divided into "forensic" and "civil", depending on the reasons for compulsory care.
IPR = Inpatient Register. NBHW = National Board of Health and Welfare.
Since January 2009, the NBHW collects additional data on compulsory psychiatric care (psychiatric care under certain laws) in addition to the IPR. The data are collected three times per year.
In older versions of the IPR, the variable "Billing forms (between counties)" was also included.
Figure 5
A sample of variables from the Swedish Inpatient Register (as seen with the statistics programme SPSS). Each hospital discharge is listed on a row. This means that an individual may occupy several rows in the IPR (first, second, third hospital discharge, etc.). The variable lpnr (or lopnr) is constructed when the dataset is delivered to the researcher, and serves as unique serial number. In the original IPR dataset, each discharge is linked to a unique Personal Identity Number (PIN)[3]. Please note that the order of the variables above may differ from that in the original IPR dataset.
The basic unit of the IPR is not the patient but the admission/discharge. Individual patients can be identified by their unique PIN.
Personal identity number (PIN)
Each hospital discharge is keyed to an individual's PIN [3] (Table 1). Overall (1964-2008), the PIN is found missing in 2.9% of all hospital discharges.
Primary diagnosis
Overall, a primary diagnosis is listed in 99% of all hospital discharges. The highest rate of missing data occurred in 1968 (4.6%), which may be due to the change from ICD-7 to ICD-8 that occurred in that year. After 2000, missing primary diagnoses have been consistently more common in psychiatric care than in somatic care (5.7-9.4% in psychiatric care vs. 0.5-0.9% in somatic care). Since the start of the IPR, primary diagnoses are missing in 0.8% of somatic care, 2.4% of geriatric care, 3.1% of psychiatric care and 0.5% of general surgery.
The proportion of patients without a primary diagnosis does not differ by hospital type (university hospitals 1.4%, county hospitals 0.7%, small local hospitals 0.8%) but is slightly higher in nursing homes (3.1%).
Injuries and poisoning: external cause
All hospital admissions for injury or poisoning must be coded by an E code indicating the cause of the injury/poisoning (Figure 6).
Figure 6
Percentage of hospital discharges for injury and poisoning with reported external cause[2].
Mode of admission and discharge
The variables "mode of admission" and "mode of discharge" describe where the patient stays before and after admission, respectively (Table 1). These variables have generally been recorded in more than 95% of all hospital admissions (with the exception of the year 1979 and in single counties in 1997-2000).
Alternative registers
Even though the IPR contains important information on a wide spectrum of diagnoses, it is sometimes preferable to use other Swedish health registers, such as the Swedish Cancer Register)[4], the Cause of Death Register[5] and the Swedish Medical Birth Register[6]. There are also a large number of Swedish National Quality Registers (n = 89 in 2011)(http://www.kvalitetsregister.se, accessed April 19, 2011).
Earlier assessment of the IPR
The NBHW has previously examined the quality of the IPR on three separate occasions (one published study with data collection in 1986 (899 patients, patient chart validation)[7], one unpublished study with data collection in 1990 (n = 875, patient chart validation)[2, 8] and one comparison between the IPR and the National Quality Registers in 2009. The two patient chart studies focused on three types of diagnostic coding error detected in medical records.
1. Diagnostic errors, i.e. the patient received an incorrect diagnosis (the patient receives an ICD code that is not related to his or her actual main complaint). Diagnostic errors were more common in internal medicine records (especially in the 1986 study [7]) than in records from gynaecology departments, and slightly more common in older than in younger patients [2].
2. Translation errors, i.e. the ICD code in the IPR is different from the code actually listed in the patient chart. This type of error was detected in less than 1% of all medical records.
3. Coding errors, i.e. the faulty ICD code accompanies an otherwise correct diagnosis. Such coding errors occurred in 5.9% of hospital discharges in 1986 and in 8.3% in 1990.
In the 1990 validation, the risk of an incorrect primary diagnosis correlated with the number of secondary diagnoses [8]. The overall proportion of incorrect diagnoses at the ICD code 3-digit/character level (e.g., ICD-9: 571 "chronic liver disease and liver cirrhosis") was 13% in 1986 and 12% in 1990; at the four-digit level (e.g., ICD-9: 571E "chronic hepatitis"), it was 15% in 1986 and 14% in 1990 (B. Smedby, personal communication, Jan 30, 2010).
The comparison between the IPR and the National Quality Registers found that the IPR has high sensitivity for most surgical procedures (Table 2)[9], whereas sensitivity varied between 76.4% and 96.0% for three diseases not requiring surgery (multiple sclerosis, incident stroke and prostate cancer)(Table 2).
Table 2
Comparison between Swedish Quality Registers and the National Patient Register [9]
Disease/
Procedure
Quality Register
Total number of cases
Matching between the Quality Register and the Patient Register (%)
Proportion of all cases identified through the Patient Register (%)
* Quality register compared with the Patient Register (IPR and hospital-based outpatient data). In the other comparisons the Quality Register was compared only with the IPR.
# Currently part of the Register Swedeheart.
Use of the IPR
Systematic collection of medical data is essential for modern health care because such data are used to plan, evaluate and fund health care. Through the IPR, administrators, health care personnel and researchers are able to (a) evaluate the incidence and prevalence of diseases [10], (b) examine the effects and consequences of interventions (e.g., surgery [11]), including quality of care and (c) establish cohorts of patients with a certain disease [12] or condition.
The primary purpose of this paper was to review and validate the IPR. A second objective was to describe its potential use in population-based epidemiological research.
Methods
Sorensen et al suggest that administrative databases could be evaluated in three ways [13]:
(a)
Through comparison with other independent reference sources
(b)
Through patient chart reviews (medical records)
(c)
By comparing the total number of cases in different databases
The majority of the evaluations in this paper were based on (b), i.e. patient chart reviews.
Assessment by the current study
In January 2010, we began identifying papers that might concern the validity of the IPR (Figure 7) using database searches in PubMed and HighWire. We used the following search algorithm: "validat* (inpatient or hospital discharge) Sweden". We also contacted 218 members of the Swedish Society of Epidemiology and another 201 researchers with experience in register-based research. Altogether, we identified 132 papers, all of which were subsequently examined in detail. Tables 3 and 4 list papers that validated the IPR.
Figure 7
Collection of validation studies. In both the PubMed and HighWire Press search, we used the following search algorithm to identify relevant papers: validat* (inpatient or hospital discharge) Sweden. Databases were searched from the start of the databases until January 2010. *In the HighWire Press literature search, JFL manually screened all titles, authors, keywords and, when available, abstracts for the 840 hits. If a validation of the inpatient register could not be ruled out, the corresponding author was contacted. A number of publications could then be excluded; 14 "new papers" remained that had not previously been identified.
Table 3
Validation of diagnoses in the Swedish Inpatient Register by Positive Predictive Values (PPVs)
Diagnosis
Correct diagnosis in the IPR (%) = positive predictive value (PPV)
Atrial fibrillation (using ICD-codes for atrial fibrillation or atrial flutter)
97/100 (97)
95 individuals had ECGs consistent with atrial fibrillation. Two patients had no ECGs available but were regarded as having atrial fibrillation on the basis of medical records.
In 40 of the 386 cases with lymphoma, medical records did not include enough information to evaluate whether the RA criteria of the American College of Rheumatology were fulfilled but all available information supported the diagnosis of RA
235/236 (99.6) admissions were correct with regards to ICD-coding
117 patients with deep foot infections and type 1 or 2 diabetes, who had been referred to a multidisciplinary foot-care team at Lund University Hospital.
Due to the ICD-classification used at this time, ulcerative colitis could not be distinguished from Crohn's disease, and therefore "overall IBD" was evaluated
104 patients discharged in 1971 with a diagnosis of schizophrenia. DSMIII-criteria were used for this validation. From Stockholm County IPR. A strict review found a PPV of 75%. This figure increased to 85% when some lack of information was accepted (less strict criteria).
(Kappa = 0.37 correlating to an agreement rate of 68%)
All individuals with a diagnosis of schizophrenia in the IPR and who had an inpatient forensic psychiatric assessment using a national register of all such evaluations from 1988-2000 (n = 1638).
Four-week- inpatient assessment in forensic psychiatry department
Guillain-Barré Syndrome (GBS)
69/83 (83)
83% of patients fulfilled the National Institute of Neurological and Communicative Disorders and Stroke criteria for GBS [56]. PPV of first primary diagnosis with GBS was 84%; and for secondary diagnosis 75%.
Surgical procedures (and related medical conditions)
Major amputations (leg and arm)
610/624 (97.8)
Unique number of amputations was 610 in 624 patients.
Malmstedt
(PC, Feb 24)
MR
Appendicitis†
1661/1840 (90.3)
Performed in the Jönköping county. Incidental appendectomies were excluded. The IPR overestimated the prevalence of appendicitis with 6%. The negative predictive value of appendicitis was 94.0% and the accuracy 91.3%†
249/250 (99.6) admissions to hospital were correct with regards to ICD-coding
Based on 117 patients with deep foot infections and type 1 or 2 diabetes, referred to multidisciplinary foot-care team. Specificity for foot ulcers and concomitant diabetes was 98%.
Among 602 patients with a primary or secondary diagnosis of acute pancreatitis in the IPR. 84.0% had a definitive acute pancreatitis and another 14.8% a probable acute pancreatitis.
Lindblad
(PC, Feb 6)
MR + laboratory tests and radiological imaging
PC, Personal communication: All personal communications took place in 2010 (exact date is listed in the table).
MR, compared with Medical Records (patient charts).
PPV, Positive Predictive Value.
NPV, Negative Predictive Value.
ULSAM, Uppsala Longitudinal Study of Adult Men) cohort.
*From ICD-7 through ICD-9, no distinction was made between type 1 and type 2 diabetes. For practical reasons "diabetes" has been listed as an autoimmune disorder.
† These studies took place when the county in question did not yet report inpatient data to the IPR, but results are deemed valid for the IPR.
Table 4
Validation of diagnoses in the Swedish Inpatient Register by sensitivity
Diagnosis
Proportion identified through the IPR (%) (sensitivity)
Comment
Main Author
PubMed ID
Year
Gold Standard (Reference)
Cardiovascular disease
Myocardial Infarction (MI)
54/59 (91.5)
Cross-sectional, 1-year retrospective study. ≥20 year-olds from Degerfors area.
The researchers identified all MIs in Stockholm county in 1973 through the local IPR and the national Cause of Death Register (restricted to individuals living in Stockholm).
4746/5832 (81.4) of MI cases in community registers were found in IPR or the Cause of Death Register.
IPR and Cause of Death Register data from 1972-1981 for regions with IPR registers at the time. 81% of cases in community registers were found in IPR or Cause of Death Register. Meanwhile 85% of cases in IPR and the Cause of Death Register were found in Community registers (disregarding non-matching fatal cases).
MR. The study evaluates different methods to identify acute coronary syndrome in patients, using information not available in the IPR (e.g. ECG measurements).
Non-specified ischemic heart disease
44/206 (21.4)
Cross-sectional, 1-year retrospective study. ≥20 year-olds from Degerfors area.
Data were obtained from IPR, hospital-based outpatient care and primary health care
Type 1 and type 2 diabetes mellitus *
349/436 (80.0) patients with diabetes in "source 1" could be identified in the IPR
The researchers evaluated the presence of type 1 or type 2 diabetes among patients with atrial fibrillations at the Södersjukhuset Hospital. All but 8 patients in this validation had type 2 diabetes.
Friberg
-
(PC, Feb 8)
MR + biochemistry data + data on insulin or other medication against diabetes
County registers of primary health care, outpatient hospital care and inpatient care
Both type 1/2 diabetes and foot ulcer
235/280 (83.9) admissions could be identified through the IPR
Based on 117 patients with deep foot infections and type 1 or 2 diabetes, referred to multidisciplinary foot-care team. Specificity for foot ulcers and concomitant diabetes was 98%.
Carotid, infrainguinal bypass and aortic aneurysm (AAA) procedures
10,861/11,638 (93.3) of patients identified through Swedvasc (Svenska kärlregistret) or the IPR could be identified in the IPR
(Swedvasc) started in 1987 and became nationwide 1994. In the paper by Troëng et al data on infraingunal and carotid bypass originated from 5 years (2000-2004), while data on aortic aneurysms originated from 1 year (2006). Divided the sensitivity figures were: Carotid 96.8%; Infrainguinal 92.3% and aortic aneurysm 88.9%.
County registers of primary health care, outpatient hospital care and inpatient care
Dementia
23/87 (26)
Participants originated from the study "Aging in women and men: a longitudinal study of gender differences in health behaviour and health among the elderly (as part of the Swedish Twin Register). The specificity for dementia was 97% (399/411)
41.3% for prevalent cases and 42.3% for incident cases.
Participants originated from the study "Aging in women and men: a longitudinal study of gender differences in health behaviour and health among the elderly (as part of the Swedish Twin Register). The specificity for dementia was 97% (399/411)
Rizutto
(PC, May 14)
Comparison between cases detected in IPR with cases derived from the Kungsholmen (KP) and SNAC-K Projects [72]
Foot ulcer (only)
249/280 (88.9%) of admissions could be identified through the IPR
Based on 117 patients with deep foot infections and type 1 or 2 diabetes, referred to multidisciplinary foot-care team.
County registers of primary health care, outpatient hospital care and inpatient care
PC, Personal communication: All personal communications took place in 2010 (exact date is listed in the table).
ICU, Intensive Care Unit. MI, Myocardial Infarction. MR, compared with Medical Records (patient charts). RA, Rheumatoid arthritis.
ULSAM, Uppsala Longitudinal Study of Adult Men) cohort.
*Data from all departments at the Lund Hospital including the emergency unit, general practitioners, community nurses, physicians at the Neurology department, autopsy registers, Forensic department, computerized searches of outpatient medical records
Results
With few exceptions, validation of ICD codes from the IPR was made by comparing registered diagnoses in the IPR with information in medical records (Tables 3 and 4). The positive predictive values (PPVs) of IPR diagnoses were 85-95% for most diagnoses (3-digit level, see Table 3). In a review of patients dying in hospital 90-98% of patients with a primary discharge diagnosis of malignancy had the same malignancy as the underlying cause of death [5]. In addition, 90.3% of those with a primary discharge diagnosis of myocardial infarction (MI) had MI as the underlying cause of death and with a similar proportion of those with other vascular diseases (89.0%). Agreement between discharge diagnosis and death certificate was slightly lower for traffic accidents (87.8%), meningitis (74.3%) and ulcer of the stomach or duodenum (69.9%) to name a few [5].
Sensitivity of the IPR was high (above 90%) for MI [14] as well as for surgery for carotid stenosis, surgery on the carotid arteries, or surgery on the arteries in the leg (infrainguinal) and aorta [15](Table 4) but low for lipid disorders and hypertension [14]. Few studies have examined to what extent an individual without a specific disease is assigned an ICD code for that disease.
Some hospital admissions are due to trauma and not disease. In 2008, Backe et al [16] used ambulance records as gold standard to examine the proportion of injuries and suffocations that were then recorded in the IPR. Agreement between the two data sources varied, with high agreement for "falls" (W00-W19; 93.9%) but lower for "road traffic accidents" (ICD-10: V01-V99) and "suffocation, drowning/near drowning, etc." (ICD-10: W64-85), where the IPR recorded less than 50% of all injuries noted in the ambulance reports.
Several studies have examined date of hospital admission. For instance, Nordgren found that for 62% (257/413) of spinal cord injuries, the hospital admission date agreed with the injury date (≤2 days within the injury date [17]).
Discussion
This review found a high PPV for the majority of evaluated diagnoses but a lower sensitivity. The PPVs reported in this review are similar to those in the Danish IPR (febrile seizures in children: 93%[18], MIs: 92-94%[19], venous thromboembolism: 75%[20]). Furthermore, US hospital data suggest a PPV of about 90% for some diagnoses (e.g., acromegaly: 76% of the patients had a definite diagnosis and 14% a probable diagnosis [21]).
The proportion of valid diagnoses in the IPR is probably higher in patients with severe as opposed to mild disease and higher among patients with causally related complications in contrast to those without complications. Baecklund et al reported that the IPR diagnosis of rheumatoid arthritis was correct in 93.5% of individuals with later lymphoma but only in 87.1% in individuals who had not developed later lymphoma [22]. In this case the positive association between lymphoma and rheumatoid arthritis leads to higher specificity for rheumatoid arthritis in patients with lymphoma.
There are several ways to increase the specificity and the PPV of a diagnosis in the IPR. In a paper on sepsis in celiac disease by Ludvigsson et al [23] sensitivity analyses were performed among patients with (1) sepsis diagnosed in a department of infectious diseases (i.e. in a department where sepsis is likely to be correctly diagnosed), (2) sepsis listed as the primary diagnosis and (3) the risk of having at least two hospital admissions with sepsis. All these measures could increase the specificity of a diagnosis. For instance, there is a risk that individuals discharged from a dermatology department with a diagnosis of MI (ICD-10: I20.9) actually had an incorrectly recorded eczema (ICD-10: L20.9). When Parikh et al examined parity and risk of later cardiovascular disease, they restricted their discharges to patients with a primary diagnosis of cardiovascular disease (or death from cardiovascular disease)[24]. In their recent paper on schizophrenia, substance abuse and violent crime Fazel et al resolved to study patients with at least two hospital admissions with schizophrenia [25].
The extent to which a condition has been reported and recorded in the IPR depends on several factors [26], including care-seeking behaviour of an individual, access to health care and the propensity of a physician to admit a patient. Hospital fees, however, are no major obstacle to inpatient care access in that the (public) health system in Sweden is almost free of charge.
Over time, an increasing number of patients are treated as outpatients [27], a trend largely driven by economic restraints but also by data indicating that the prognosis of some diseases (e.g., stroke) has an improved prognosis in ambulatory care [28]. The trend towards outpatient care suggests that the sensitivity of the IPR may have decreased in recent years for some diseases. In fact, our validation showed that the IPR has low sensitivity for hypertension and lipid disorders. The introduction of day care anaesthesia has resulted in that certain procedures, such as small-intestinal biopsy preceding a diagnosis of celiac disease [29], which previously required inpatient care, are nowadays often performed on an outpatient basis.
When Elmberg et al estimated mortality in patients with hereditary haemochromatosis (HH)[30], they found a relative risk of death of 2.15 among HH patients identified through the IPR, but only 1.09 in patients identified through regional clinic registers and 1.15 in those identified through outpatient data sources [30]. Some evidence suggests that patients with a certain disorder identified through the IPR may suffer from more intense disease than the average patient and be at higher risk of complications than patients identified outside the IPR (a phenomenon sometimes called Berkson's bias [31]).
Another issue that deserves attention is that the first recorded admission with a disorder is not always equal to the incident admission. According to patient chart reviews, 1 in 3 patients with a hospital admission for stroke had had an earlier stroke (L. Olai, personal communication, Feb 4, 2010). In an effort to separate incident admissions from readmissions some authors have suggested using prediction models combining information from current and previous records in the IPR [32]. It should be noted that the Swedish ICD system does contain a number of codes representing late effects of disease, such as ICD code I69 ("late effects of cerebrovascular disease").
A number of non-medical factors influence the coding of hospital discharges. Although originally used to collect data on health care use, today the IPR coding is also used as the basis for management and financing. Some hospitals have introduced compulsory use of certain secondary codes (when such codes apply) because these codes generate extra funding (e.g., a secondary code of diabetes mellitus is "valuable"). Further, international research suggests that the coding pattern may differ between hospitals and general practice [33]. Financial incitements have therefore led to a "diagnostic drift" in which more secondary diagnoses are listed [27] and where it is financially more rewarding to assign a patient a severe primary diagnosis than a severe secondary diagnosis (e.g., type 1 diabetes is more "valuable" as a primary diagnosis than as a secondary diagnosis). The effects of financial incitements on ICD coding have probably been underestimated and are likely to have changed the epidemiological pattern. A standardized behaviour of assigning ICD codes is therefore of importance for all stakeholders, including the Swedish state [27].
Despite the extensive scope of the IPR, there is still a need for additional variables (Additional file 3), including laterality, index admission, earlier comorbidity and risk factors (e.g., smoking).
Conclusion
In conclusion, the Swedish IPR is a valuable resource for large-scale register-based research. A number of diagnoses have already been validated by the NBHW and by individual researchers. Current data suggest that the overall PPV of diagnoses in the register is about 85-95%.
List of abbreviations
IPR:
Swedish Inpatient Register (Slutenvårdsregistret)
MI:
Myocardial infarction
NBHW:
National Board of Health and Welfare (Socialstyrelsen)
PIN:
Personal Identity Number.
Declarations
Acknowledgements
We would like to thank all the researchers contributing to this paper. Special thanks go to Björn Smedby and Paul Blomqvist for their generous advice and to the University Library of the Karolinska Institutet for assistance.
The writing of this paper was made possible by a grant from the Swedish Society of Medicine, funding the salary of the main author, Jonas F Ludvigsson. The paper was written on behalf of SVEP - The Swedish Society of Epidemiology.
Authors’ Affiliations
(1)
Department of Paediatrics, Örebro University Hospital
(2)
Clinical Epidemiology Unit, Department of Medicine, Karolinska University Hospital, Karolinska Institutet
(3)
Section of Occupational and Environmental Medicine, University of Gothenburg
(4)
Institute of Environmental Medicine, Karolinska Institutet
(5)
Research Unit
(6)
Department of Public Health and Clinical Medicine, Umeå University
(7)
Department of Statistics and Analyses, National Board of Health and Welfare
References
Smedby B, Schiøler G: Health Classifications in the Nordic Countries. Historic development in a national and international perspective 2006. Book Health Classifications in the Nordic Countries. Historic development in a national and international perspective 2006 (Editor ed.^eds.). 2006, City: Nordisk Medicinalstatistisk KomiteGoogle Scholar
Forsberg L, Rydh H, Jacobsson A, Nyqvist K, Heurgren M: Kvalitet och innehåll i patientregistret. Utskrivningar från slutenvården 1964-2007 och besök i specialiserad öppenvård (exklusive primärvårdsbesök) 1997-2007. (Quality and content of the Patient Register)(2009-125-15). Book Kvalitet och innehåll i patientregistret. Utskrivningar från slutenvården 1964-2007 och besök i specialiserad öppenvård (exklusive primärvårdsbesök) 1997-2007. (Quality and content of the Patient Register)(2009-125-15). (Editor ed.^eds.). City. 2009Google Scholar
Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, Ekbom A: The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol. 2009, 24: 659-667. 10.1007/s10654-009-9350-y.View ArticlePubMedPubMed CentralGoogle Scholar
Mattsson B, Wallgren A: Completeness of the Swedish Cancer Register. Non-notified cancer cases recorded on death certificates in 1978. Acta Radiol Oncol. 1984, 23: 305-313. 10.3109/02841868409136026.View ArticlePubMedGoogle Scholar
Johansson LA, Westerling R: Comparing Swedish hospital discharge records with death certificates: implications for mortality statistics. Int J Epidemiol. 2000, 29: 495-502. 10.1093/ije/29.3.495.View ArticlePubMedGoogle Scholar
SNBHW: The Swedish Medical Birth Register: a summary of content and quality. Book The Swedish Medical Birth Register: a summary of content and quality. (Editor ed.^eds.). 2003, City: Swedish National Board of Health and Welfare, [http://www.sos.se/fulltext/112/2003-112-3/2003-112-3.pdf]Google Scholar
Nilsson AC, Spetz CL, Carsjo K, Nightingale R, Smedby B: Reliability of the hospital registry. The diagnostic data are better than their reputation. Lakartidningen. 1994, 91: 598-603-595PubMedGoogle Scholar
Carjsö K, Smedby B, Spetz CL: Evaluation of the Data quality in the Swedish National Hospital Discharge Register. Book Evaluation of the Data quality in the Swedish National Hospital Discharge Register (Editor ed.^eds.). pp. 1-10. City: WHO/HST/ICD/C/98.34. 1998, 1-10.Google Scholar
Kouchecki B, Köster M, Eckerström L, Wiberg-Hedman K: Bilaga 1-3 till Öppna jämförelser av hälso- och sjukvårdens kvalitet och effektivitet 2009. Book Bilaga 1-3 till Öppna jämförelser av hälso- och sjukvårdens kvalitet och effektivitet 2009. (Editor ed.^eds.). 2009, City: Socialstyrelsen (National Board of Health and Welfare)Google Scholar
Fang F, Valdimarsdottir U, Bellocco R, Ronnevi LO, Sparen P, Fall K, Ye W: Amyotrophic lateral sclerosis in Sweden, 1991-2005. Arch Neurol. 2009, 66: 515-519. 10.1001/archneurol.2009.13.View ArticlePubMedGoogle Scholar
Jonas E, Marsk R, Rasmussen F, Freedman J: Incidence of postoperative gallstone disease after antiobesity surgery: population-based study from Sweden. Surg Obes Relat Dis. 2010, 6: 54-58. 10.1016/j.soard.2009.03.221.View ArticlePubMedGoogle Scholar
Knight A, Askling J, Ekbom A: Cancer incidence in a population-based cohort of patients with Wegener's granulomatosis. Int J Cancer. 2002, 100: 82-85. 10.1002/ijc.10444.View ArticlePubMedGoogle Scholar
Sorensen HT, Sabroe S, Olsen J: A framework for evaluation of secondary data sources for epidemiological research. Int J Epidemiol. 1996, 25: 435-442. 10.1093/ije/25.2.435.View ArticlePubMedGoogle Scholar
Elo SL, Karlberg IH: Validity and utilization of epidemiological data: a study of ischaemic heart disease and coronary risk factors in a local population. Public Health. 2009, 123: 52-57. 10.1016/j.puhe.2008.07.010.View ArticlePubMedGoogle Scholar
Troeng T, Malmstedt J, Bjorck M: External validation of the Swedvasc registry: a first-time individual cross-matching with the unique personal identity number. Eur J Vasc Endovasc Surg. 2008, 36: 705-712. 10.1016/j.ejvs.2008.08.017.View ArticlePubMedGoogle Scholar
Backe SN, Andersson R: Monitoring the "tip of the iceberg'': ambulance records as a source of injury surveillance. Scand J Public Health. 2008, 36: 250-257. 10.1177/1403494807086973.View ArticlePubMedGoogle Scholar
Nordgren C: On the need of validating inpatient registers. Spinal Cord. 2008, 46: 748-752. 10.1038/sc.2008.42.View ArticlePubMedGoogle Scholar
Vestergaard M, Obel C, Henriksen TB, Christensen J, Madsen KM, Ostergaard JR, Olsen J: The Danish National Hospital Register is a valuable study base for epidemiologic research in febrile seizures. J Clin Epidemiol. 2006, 59: 61-66. 10.1016/j.jclinepi.2005.05.008.View ArticlePubMedGoogle Scholar
Madsen M, Davidsen M, Rasmussen S, Abildstrom SZ, Osler M: The validity of the diagnosis of acute myocardial infarction in routine statistics: a comparison of mortality and hospital discharge data with the Danish MONICA registry. J Clin Epidemiol. 2003, 56: 124-130. 10.1016/S0895-4356(02)00591-7.View ArticlePubMedGoogle Scholar
Severinsen MT, Kristensen SR, Overvad K, Dethlefsen C, Tjonneland A, Johnsen SP: Venous thromboembolism discharge diagnoses in the Danish National Patient Registry should be used with caution. J Clin Epidemiol. 63: 223-228.Google Scholar
Ron E, Gridley G, Hrubec Z, Page W, Arora S, Fraumeni JF: Acromegaly and gastrointestinal cancer. Cancer. 1991, 68: 1673-1677. 10.1002/1097-0142(19911015)68:8<1673::AID-CNCR2820680802>3.0.CO;2-0.View ArticlePubMedGoogle Scholar
Baecklund E, Iliadou A, Askling J, Ekbom A, Backlin C, Granath F, Catrina AI, Rosenquist R, Feltelius N, Sundstrom C, Klareskog L: Association of chronic inflammation, not its treatment, with increased lymphoma risk in rheumatoid arthritis. Arthritis Rheum. 2006, 54: 692-701. 10.1002/art.21675.View ArticlePubMedGoogle Scholar
Ludvigsson JF, Olen O, Bell M, Ekbom A, Montgomery SM: Coeliac disease and risk of sepsis. Gut. 2008, 57: 1074-1080. 10.1136/gut.2007.133868.View ArticlePubMedGoogle Scholar
Parikh NI, Cnattingius S, Dickman PW, Mittleman MA, Ludvigsson JF, Ingelsson E: Parity and risk of later-life maternal cardiovascular disease. Am Heart J. 159: 215-221. e216Google Scholar
Fazel S, Langstrom N, Hjern A, Grann M, Lichtenstein P: Schizophrenia, substance abuse, and violent crime. JAMA. 2009, 301: 2016-2023. 10.1001/jama.2009.675.View ArticlePubMedPubMed CentralGoogle Scholar
Blomqvist P: On the use of administrative databases in health care analyses (Dissertation). Karolinska Institutet. 1998Google Scholar
Heurgren M, Serden L: Patientregistret för 2008 ur ett DRG-perspektiv. (The Patient Register from a DRG perspective)(2010-4-14). Book Patientregistret för 2008 ur ett DRG-perspektiv. (The Patient Register from a DRG perspective)(2010-4-14). (Editor ed.^eds.). City. 2009Google Scholar
Collaborative systematic review of the randomised trials of organised inpatient (stroke unit) care after stroke. Stroke Unit Trialists' Collaboration. BMJ. 1997, 314: 1151-1159.Google Scholar
Ludvigsson JF, Montgomery SM, Ekbom A, Brandt L, Granath F: Small-intestinal histopathology and mortality risk in celiac disease. JAMA. 2009, 302: 1171-1178. 10.1001/jama.2009.1320.View ArticlePubMedGoogle Scholar
Elmberg M, Hultcrantz R, Ebrahim F, Olsson S, Lindgren S, Loof L, Stal P, Wallerstedt S, Almer S, Sandberg-Gertzen H, et al: Increased mortality risk in patients with phenotypic hereditary hemochromatosis but not in their first-degree relatives. Gastroenterology. 2009, 137: 1301-1309. 10.1053/j.gastro.2009.07.038.View ArticlePubMedGoogle Scholar
Berkson J: Limitations of the application of fourfold tables to hospital data. Biometrics Bulletin. 1946, 2: 47-53. 10.2307/3002000.View ArticlePubMedGoogle Scholar
Kljakovic M, Abernethy D, de Ruiter I: Quality of diagnostic coding and information flow from hospital to general practice. Inform Prim Care. 2004, 12: 227-234.PubMedGoogle Scholar
Linnersjo A, Hammar N, Gustavsson A, Reuterwall C: Recent time trends in acute myocardial infarction in Stockholm, Sweden. Int J Cardiol. 2000, 76: 17-21. 10.1016/S0167-5273(00)00366-1.View ArticlePubMedGoogle Scholar
Guidelines for the diagnosis of heart failure. The Task Force on Heart Failure of the European Society of Cardiology. Eur Heart J. 1995, 16: 741-751.Google Scholar
Ingelsson E, Arnlov J, Sundstrom J, Lind L: The validity of a diagnosis of heart failure in a hospital discharge register. Eur J Heart Fail. 2005, 7: 787-791. 10.1016/j.ejheart.2004.12.007.View ArticlePubMedGoogle Scholar
Smith JG, Platonov PG, Hedblad B, Engstrom G, Melander O: Atrial fibrillation in the Malmo diet and cancer study: a study of occurrence, risk factors and diagnostic validity. Eur J Epidemiol. 2009, 25: 95-102.View ArticleGoogle Scholar
Stegmayr B, Asplund K: Measuring stroke in the population: quality of routine statistics in comparison with a population-based stroke registry. Neuroepidemiology. 1992, 11: 204-213. 10.1159/000110933.View ArticlePubMedGoogle Scholar
Ros HS, Cnattingius S, Lipworth L: Comparison of risk factors for preeclampsia and gestational hypertension in a population-based cohort study. Am J Epidemiol. 1998, 147: 1062-1070.View ArticlePubMedGoogle Scholar
Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, Healey LA, Kaplan SR, Liang MH, Luthra HS, et al: The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988, 31: 315-324. 10.1002/art.1780310302.View ArticlePubMedGoogle Scholar
Leavitt RY, Fauci AS, Bloch DA, Michel BA, Hunder GG, Arend WP, Calabrese LH, Fries JF, Lie JT, Lightfoot RW, et al: The American College of Rheumatology 1990 criteria for the classification of Wegener's granulomatosis. Arthritis Rheum. 1990, 33: 1101-1107.View ArticlePubMedGoogle Scholar
Smedby KE, Akerman M, Hildebrand H, Glimelius B, Ekbom A, Askling J: Malignant lymphomas in coeliac disease: evidence of increased risks for lymphoma types other than enteropathy-type T cell lymphoma. Gut. 2005, 54: 54-59. 10.1136/gut.2003.032094.View ArticlePubMedPubMed CentralGoogle Scholar
Bensing S, Brandt L, Tabaroj F, Sjoberg O, Nilsson B, Ekbom A, Blomqvist P, Kampe O: Increased death risk and altered cancer incidence pattern in patients with isolated or combined autoimmune primary adrenocortical insufficiency. Clin Endocrinol (Oxf). 2008, 69: 697-704. 10.1111/j.1365-2265.2008.03340.x.View ArticleGoogle Scholar
Ragnarson Tennvall G, Apelqvist J, Eneroth M: The inpatient care of patients with diabetes mellitus and foot ulcers. A validation study of the correspondence between medical records and the Swedish Inpatient Registry with the consequences for cost estimations. J Intern Med. 2000, 248: 397-405. 10.1046/j.1365-2796.2000.00748.x.View ArticlePubMedGoogle Scholar
Ekbom A, Helmick C, Zack M, Adami HO: The epidemiology of inflammatory bowel disease: a large, population-based study in Sweden. Gastroenterology. 1991, 100: 350-358.View ArticlePubMedGoogle Scholar
Ekholm B, Ekholm A, Adolfsson R, Vares M, Osby U, Sedvall GC, Jonsson EG: Evaluation of diagnostic procedures in Swedish patients with schizophrenia and related psychoses. Nord J Psychiatry. 2005, 59: 457-464. 10.1080/08039480500360906.View ArticlePubMedGoogle Scholar
Dalman C, Broms J, Cullberg J, Allebeck P: Young cases of schizophrenia identified in a national inpatient register--are the diagnoses valid?. Soc Psychiatry Psychiatr Epidemiol. 2002, 37: 527-531. 10.1007/s00127-002-0582-3.View ArticlePubMedGoogle Scholar
Lichtenstein P, Bjork C, Hultman CM, Scolnick E, Sklar P, Sullivan PF: Recurrence risks for schizophrenia in a Swedish national cohort. Psychol Med. 2006, 36: 1417-1425. 10.1017/S0033291706008385.View ArticlePubMedGoogle Scholar
Reutfors J: Life time OPCRIT generated DSM-IV diagnoses among patients with an initial clinical ICD schizophrenia spectrum diagnosis. 29th Nordic Congress of Psychiatr; Stockholm. 2009Google Scholar
Kristjansson E, Allebeck P, Wistedt B: Validity of the diagnosis schizophrenia in a psychiatric inpatient register. 1987, 41: 229-234.Google Scholar
Bergman B, Belfrage H, Grann M: Mentally disorderd offenders in Sweden: Forensic and general psychiatric diagnoses. American Journal of Forensic Psychiatry. 1999, 20: 27-37.Google Scholar
Jin YP, Gatz M, Johansson B, Pedersen NL: Sensitivity and specificity of dementia coding in two Swedish disease registries. Neurology. 2004, 63: 739-741.View ArticlePubMedGoogle Scholar
Grann M, Haggård U, Tengström A, Woodhouse A, Långström N, Holmberg G, Kullgren G: Some experiences from registers of interest to forensic research in Sweden. Scand J Forensic Science. 1998, 78-80.Google Scholar
Criteria for diagnosis of Guillain-Barre syndrome. Ann Neurol. 1978, 3: 565-566.Google Scholar
Jiang GX, de Pedro-Cuesta J, Fredrikson S: Guillain-Barre syndrome in south-west Stockholm, 1973-1991, 1. Quality of registered hospital diagnoses and incidence. Acta Neurol Scand. 1995, 91: 109-117.View ArticlePubMedGoogle Scholar
Hjalmarsson A, Blomqvist P, Skoldenberg B: Herpes simplex encephalitis in Sweden, 1990-2001: incidence, morbidity, and mortality. Clin Infect Dis. 2007, 45: 875-880. 10.1086/521262.View ArticlePubMedGoogle Scholar
Michaelsson K, Baron JA, Farahmand BY, Johnell O, Magnusson C, Persson PG, Persson I, Ljunghall S: Hormone replacement therapy and risk of hip fracture: population based case-control study. The Swedish Hip Fracture Study Group. Bmj. 1998, 316: 1858-1863.View ArticlePubMedPubMed CentralGoogle Scholar
Andersson R, Hugander A, Thulin A, Nystrom PO, Olaison G: Indications for operation in suspected appendicitis and incidence of perforation. Bmj. 1994, 308: 107-110.View ArticlePubMedPubMed CentralGoogle Scholar
Zendehdel K, Nyren O, Edberg A, Ye W: Risk of Esophageal Adenocarcinoma in Achalasia Patients, a Retrospective Cohort Study in Sweden. Am J Gastroenterol. 2007Google Scholar
Nyren O, Yin L, Josefsson S, McLaughlin JK, Blot WJ, Engqvist M, Hakelius L, Boice JD, Adami HO: Risk of connective tissue disease and related disorders among women with breast implants: a nation-wide retrospective cohort study in Sweden. Bmj. 1998, 316: 417-422.View ArticlePubMedPubMed CentralGoogle Scholar
Merlo J, Lindblad U, Pessah-Rasmussen H, Hedblad B, Rastam J, Isacsson SO, Janzon L, Rastam L: Comparison of different procedures to identify probable cases of myocardial infarction and stroke in two Swedish prospective cohort studies using local and national routine registers. Eur J Epidemiol. 2000, 16: 235-243. 10.1023/A:1007634722658.View ArticlePubMedGoogle Scholar
Ahlbom A: Acute myocardial infarction in Stockholm--a medical information system as an epidemiological tool. Int J Epidemiol. 1978, 7: 271-276. 10.1093/ije/7.3.271.View ArticlePubMedGoogle Scholar
Hammar N, Nerbrand C, Ahlmark G, Tibblin G, Tsipogianni A, Johansson S, Wilhelmsen L, Jacobsson S, Hansen O: Identification of cases of myocardial infarction: hospital discharge data and mortality data compared to myocardial infarction community registers. Int J Epidemiol. 1991, 20: 114-120. 10.1093/ije/20.1.114.View ArticlePubMedGoogle Scholar
Forberg JL, Green M, Bjork J, Ohlsson M, Edenbrandt L, Ohlin H, Ekelund U: In search of the best method to predict acute coronary syndrome using only the electrocardiogram from the emergency department. J Electrocardiol. 2009, 42: 58-63. 10.1016/j.jelectrocard.2008.07.010.View ArticlePubMedGoogle Scholar
Hallstrom B, Jonsson AC, Nerbrand C, Petersen B, Norrving B, Lindgren A: Lund Stroke Register: hospitalization pattern and yield of different screening methods for first-ever stroke. Acta Neurol Scand. 2007, 115: 49-54. 10.1111/j.1600-0404.2006.00738.x.View ArticlePubMedGoogle Scholar
Wirehn AB, Karlsson HM, Carstensen JM: Estimating disease prevalence using a population-based administrative healthcare database. Scand J Public Health. 2007, 35: 424-431. 10.1080/14034940701195230.View ArticlePubMedGoogle Scholar
Gedeborg R, Furebring M, Michaelsson K: Diagnosis-dependent misclassification of infections using administrative data variably affected incidence and mortality estimates in ICU patients. J Clin Epidemiol. 2007, 60: 155-162.View ArticlePubMedGoogle Scholar
Askling J, Fored CM, Brandt L, Baecklund E, Bertilsson L, Coster L, Geborek P, Jacobsson LT, Lindblad S, Lysholm J, et al: Risk and case characteristics of tuberculosis in rheumatoid arthritis associated with tumor necrosis factor antagonists in Sweden. Arthritis Rheum. 2005, 52: 1986-1992. 10.1002/art.21137.View ArticlePubMedGoogle Scholar
Dahl A, Berg S, Nilsson SE: Identification of dementia in epidemiological research: a study on the usefulness of various data sources. Aging Clin Exp Res. 2007, 19: 381-389.View ArticlePubMedGoogle Scholar
Fratiglioni L, Viitanen M, Backman L, Sandman PO, Winblad B: Occurrence of dementia in advanced age: the study design of the Kungsholmen Project. Neuroepidemiology. 1992, 11 (Suppl 1): 29-36.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Comments
View archived comments (1)