The Danish National Patient Registry: a review of content, data quality, and research potential

Background

The Danish National Patient Registry (DNPR) is one of the world’s oldest nationwide hospital registries and is used extensively for research. Many studies have validated algorithms for identifying health events in the DNPR, but the reports are fragmented and no overview exists.

Objectives

To review the content, data quality, and research potential of the DNPR.

Methods

We examined the setting, history, aims, content, and classification systems of the DNPR. We searched PubMed and the Danish Medical Journal to create a bibliography of validation studies. We included also studies that were referenced in retrieved papers or known to us beforehand. Methodological considerations related to DNPR data were reviewed.

Results

During 1977–2012, the DNPR registered 8,085,603 persons, accounting for 7,268,857 inpatient, 5,953,405 outpatient, and 5,097,300 emergency department contacts. The DNPR provides nationwide longitudinal registration of detailed administrative and clinical data. It has recorded information on all patients discharged from Danish nonpsychiatric hospitals since 1977 and on psychiatric inpatients and emergency department and outpatient specialty clinic contacts since 1995. For each patient contact, one primary and optional secondary diagnoses are recorded according to the International Classification of Diseases. The DNPR provides a data source to identify diseases, examinations, certain in-hospital medical treatments, and surgical procedures. Long-term temporal trends in hospitalization and treatment rates can be studied. The positive predictive values of diseases and treatments vary widely (<15%–100%). The DNPR data are linkable at the patient level with data from other Danish administrative registries, clinical registries, randomized controlled trials, population surveys, and epidemiologic field studies – enabling researchers to reconstruct individual life and health trajectories for an entire population.

Conclusion

The DNPR is a valuable tool for epidemiological research. However, both its strengths and limitations must be considered when interpreting research results, and continuous validation of its clinical data is essential.

Incidence of adenocarcinoma among patients with Barrett's esophagus

BACKGROUND

Accurate population-based data are needed on the incidence of esophageal adenocarcinoma and high-grade dysplasia among patients with Barrett's esophagus.

METHODS

We conducted a nationwide, population-based, cohort study involving all patients with Barrett's esophagus in Denmark during the period from 1992 through 2009, using data from the Danish Pathology Registry and the Danish Cancer Registry. We determined the incidence rates (numbers of cases per 1000 person-years) of adenocarcinoma and high-grade dysplasia. As a measure of relative risk, standardized incidence ratios were calculated with the use of national cancer rates in Denmark during the study period.

RESULTS

We identified 11,028 patients with Barrett's esophagus and analyzed their data for a median of 5.2 years. Within the first year after the index endoscopy, 131 new cases of adenocarcinoma were diagnosed. During subsequent years, 66 new adenocarcinomas were detected, yielding an incidence rate for adenocarcinoma of 1.2 cases per 1000 person-years (95% confidence interval [CI], 0.9 to 1.5). As compared with the risk in the general population, the relative risk of adenocarcinoma among patients with Barrett's esophagus was 11.3 (95% CI, 8.8 to 14.4). The annual risk of esophageal adenocarcinoma was 0.12% (95% CI, 0.09 to 0.15). Detection of low-grade dysplasia on the index endoscopy was associated with an incidence rate for adenocarcinoma of 5.1 cases per 1000 person-years. In contrast, the incidence rate among patients without dysplasia was 1.0 case per 1000 person-years. Risk estimates for patients with high-grade dysplasia were slightly higher.

CONCLUSIONS

Barrett's esophagus is a strong risk factor for esophageal adenocarcinoma, but the absolute annual risk, 0.12%, is much lower than the assumed risk of 0.5%, which is the basis for current surveillance guidelines. Data from the current study call into question the rationale for ongoing surveillance in patients who have Barrett's esophagus without dysplasia. (Funded by the Clinical Institute, University of Aarhus, Aarhus, Denmark.).

Vagotomy and subsequent risk of Parkinson's disease

OBJECTIVE

Parkinson's disease (PD) may be caused by an enteric neurotropic pathogen entering the brain through the vagal nerve, a process that may take over 20 years. We investigated the risk of PD in patients who underwent vagotomy and hypothesized that truncal vagotomy is associated with a protective effect, whereas superselective vagotomy has a minor effect.

METHODS

We constructed cohorts of all patients in Denmark who underwent vagotomy during 1977-1995 and a matched general population cohort by linking Danish registries. We used Cox regression to compute hazard ratios (HRs) for PD and corresponding 95% confidence intervals (CIs), adjusting for potential confounders.

RESULTS

Risk of PD was decreased in patients who underwent truncal (HR = 0.85; 95% CI = 0.56-1.27; follow-up of >20 years: HR = 0.58; 95% CI: 0.28-1.20) compared to superselective vagotomy. Risk of PD was also decreased after truncal vagotomy when compared to the general population cohort (overall adjusted HR = 0.85; 95% CI: 0.63-1.14; follow-up >20 years, adjusted HR = 0.53; 95% CI: 0.28-0.99). In patients who underwent superselective vagotomy, risk of PD was similar to the general population (HR = 1.09; 95% CI: 0.84-1.43; follow-up of >20 years: HR = 1.16; 95% CI: 0.80-1.70). Statistical precision of risk estimates was limited. Results were consistent after external adjustment for unmeasured confounding by smoking.

INTERPRETATION

Full truncal vagotomy is associated with a decreased risk for subsequent PD, suggesting that the vagal nerve may be critically involved in the pathogenesis of PD.

Missing data and multiple imputation in clinical epidemiological research

Missing data are ubiquitous in clinical epidemiological research. Individuals with missing data may differ from those with no missing data in terms of the outcome of interest and prognosis in general. Missing data are often categorized into the following three types: missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR). In clinical epidemiological research, missing data are seldom MCAR. Missing data can constitute considerable challenges in the analyses and interpretation of results and can potentially weaken the validity of results and conclusions. A number of methods have been developed for dealing with missing data. These include complete-case analyses, missing indicator method, single value imputation, and sensitivity analyses incorporating worst-case and best-case scenarios. If applied under the MCAR assumption, some of these methods can provide unbiased but often less precise estimates. Multiple imputation is an alternative method to deal with missing data, which accounts for the uncertainty associated with missing data. Multiple imputation is implemented in most statistical software under the MAR assumption and provides unbiased and valid estimates of associations based on information from the available data. The method affects not only the coefficient estimates for variables with missing data but also the estimates for other variables with no missing data.

Use of glucocorticoids and risk of venous thromboembolism: a nationwide population-based case-control study

IMPORTANCE

Excess endogenous cortisol has been linked to venous thromboembolism (VTE) risk, but whether this relationship applies to exogenous glucocorticoids remains uncertain. Because the prevalence of glucocorticoid use and the incidence of VTE are high, an increased risk of VTE associated with glucocorticoid use would have important implications.

BACKGROUND

To examine the association between glucocorticoid use and VTE.

DESIGN

Population-based case-control study using nationwide databases.

SETTING

Denmark (population 5.6 million).

PARTICIPANTS

We identified 38,765 VTE cases diagnosed from January 1, 2005, through December 31, 2011, and 387,650 population controls included through risk-set sampling and matched by birth year and sex. The VTE diagnosis date for the case was the index date for cases and matched controls.

EXPOSURE

We classified individuals who filled their most recent glucocorticoid prescription 90 days or less, 91 to 365 days, and more than 365 days before the index date as present, recent, and former users, respectively. Present users were subdivided into new (first-ever prescription 90 days or less before the index date) and continuing users (others).

MAIN OUTCOMES AND MEASURES

We used conditional logistic regression adjusted for VTE risk factors to estimate incidence rate ratios (IRRs) and 95% CIs for glucocorticoid users vs nonusers.

RESULTS

Systemic glucocorticoids increased VTE risk among present (adjusted IRR, 2.31; 95% CI, 2.18-2.45), new (3.06; 2.77-3.38), continuing (2.02; 1.88-2.17), and recent (1.18; 1.10-1.26) users but not among former users (0.94; 0.90-0.99). The adjusted IRR increased from 1.00 (95% CI, 0.93-1.07) for a prednisolone-equivalent cumulative dose of 10 mg or less to 1.98 (1.78-2.20) for more than 1000 to 2000 mg, and to 1.60 (1.49-1.71) for doses higher than 2000 mg. New use of inhaled (adjusted IRR, 2.21; 95% CI, 1.72-2.86) and intestinal-acting (2.17; 1.27-3.71) glucocorticoids also increased VTE risk.

CONCLUSIONS AND RELEVANCE

The risk of VTE is increased among glucocorticoid users. Although residual confounding may partly explain this finding, we consider a biological mechanism likely because the association followed a clear temporal gradient, persisted after adjustment for indicators of severity of underlying disease, and existed also for noninflammatory conditions. Hence, our observations merit clinical attention.

Statin prescriptions and breast cancer recurrence risk: a Danish nationwide prospective cohort study

BACKGROUND

Accumulating evidence suggests that statins affect diseases other than cardiovascular disease, including cancer, and that these effects may depend on the lipid solubility of specific statins. Though many studies have reported an association between statin use and breast cancer incidence, the relationship between statin use and breast cancer recurrence has not been well studied.

METHODS

We conducted a nationwide, population-based prospective cohort study of all female residents in Denmark diagnosed with stage I-III invasive breast carcinoma who were reported to the Danish Breast Cancer Cooperative Group registry between 1996 and 2003 (n = 18,769). Women were followed for a median of 6.8 years after diagnosis. Prescriptions for lipophilic and hydrophilic statins were ascertained from the national electronic pharmacy database. Associations between statin prescriptions and breast cancer recurrence were estimated with generalized linear models and Cox proportional hazards regression with adjustment for age and menopausal status at diagnosis; histological grade; estrogen receptor status; receipt of adjuvant therapy; type of primary surgery received; pre-diagnosis hormone replacement therapy; and co-prescriptions of aspirin, angiotensin-converting enzyme inhibitors, nonsteroidal anti-inflammatory drugs, or anticoagulants. All statistical tests were two-sided.

RESULTS

Most prescriptions for lipophilic statins in the study population were for simvastatin. Exclusive simvastatin users experienced approximately 10 fewer breast cancer recurrences per 100 women after 10 years of follow-up (adjusted 10-year risk difference = -0.10, 95% confidence interval = -0.11 to -0.08), compared with women who were not prescribed a statin. Exclusive hydrophilic statin users had approximately the same risk of breast cancer recurrence as women not prescribed a statin over follow-up (adjusted 10-year risk difference = 0.05, 95% confidence interval = -0.01 to 0.11).

CONCLUSIONS

Simvastatin, a highly lipophilic statin, was associated with a reduced risk of breast cancer recurrence among Danish women diagnosed with stage I-III breast carcinoma, whereas no association between hydrophilic statin use and breast cancer recurrence was observed.

Existing data sources for clinical epidemiology: The Danish National Database of Reimbursed Prescriptions

The Danish health care system provides partial reimbursement of most prescription medications in Denmark. The dispensation of prescription medications is registered in administrative databases. Each time a prescription is redeemed at a pharmacy, an electronic record is generated with information related to the user, prescriber, the pharmacy, and the dispensed drug. The National Health Service gathers this information for administration of the drug reimbursement plan. Recently, this information became the basis for the establishment of a new research database, the Danish National Database of Reimbursed Prescriptions (DNDRP). In this paper, we review the content, coverage, quality, linkage, access, and research possibilities of this new database. The database encompasses the reimbursement records of all reimbursed drugs sold in community pharmacies and hospital-based outpatient pharmacies in Denmark since 2004. On average, approximately 3.5 million users are recorded in the database each year. During the coverage period, the number of annual prescription redemptions increased by 15%. Most dispensed prescriptions are in the categories "alimentary tract and metabolism", "cardiovascular system", "nervous system", and "respiratory system". Individuals are identified by the unique central personal registration (CPR) number assigned to all persons born in or immigrating to Denmark. The new database fully complies with Denmark's Act on Processing of Personal Data, while avoiding additional restrictions imposed on data use at the Danish National Prescription Registry, administered by Statistics Denmark. Most importantly, CPR numbers are reversibly encrypted, which allows re-identification of drug users; furthermore, the data access is possible outside the servers of Statistics Denmark. These features open additional opportunities for international collaboration, validation studies, studies on adverse drug effects requiring review of medical records, studies involving contact to general practitioners, and linkage of prescription data to other clinical and research databases. The DNDRP thus is a valuable data source for pharmacoepidemiological research.

Colorectal cancer in ulcerative colitis: a Scandinavian population-based cohort study

BACKGROUND

Ulcerative colitis (UC) is a risk factor for colorectal cancer (CRC). However, available studies reflect older treatment and surveillance paradigms, and most have assessed risks for incident CRC without taking surveillance and lead-time bias into account, such as by assessing CRC incidence by tumour stage, or stage-adjusted mortality from CRC. We aimed to compare both overall and country-specific risks of CRC mortality and incident CRC among patients with UC.

METHODS

In this population-based cohort study of 96 447 patients with UC in Denmark (n=32 919) and Sweden (n=63 528), patients were followed up for CRC incidence and CRC mortality between Jan 1, 1969, and Dec 31, 2017, and compared with matched reference individuals from the general population (n=949 207). Patients with UC were selected from national registers and included in the analysis if they had two or more records with a relevant International Classification of Disease in the patient register (in the country in question) or one such record plus a colorectal biopsy report with a morphology code suggestive of inflammatory bowel disease. For every patient with UC, we selected matched reference individuals from the total population registers of Denmark and Sweden, who were matched for sex, age, birth year, and place of residence. We used Cox regression to compute hazard ratios (HRs) for incident CRC, and for CRC mortality, taking tumour stage into account.

FINDINGS

During follow-up, we observed 1336 incident CRCs in the UC cohort (1·29 per 1000 person-years) and 9544 incident CRCs in reference individuals (0·82 per 1000 person-years; HR 1·66, 95% CI 1·57-1·76). In the UC cohort, 639 patients died from CRC (0·55 per 1000 person-years), compared with 4451 reference individuals (0·38 per 1000 person-years; HR 1·59, 95% CI 1·46-1·72) during the same time period. The CRC stage distribution in people with UC was less advanced (p<0·0001) than in matched reference individuals, but taking tumour stage into account, patients with UC and CRC remained at increased risk of CRC death (HR 1·54, 95% CI 1·33-1·78). The excess risks declined over calendar periods: during the last 5 years of follow-up (2013-17, Sweden only), the HR for incident CRC in people with UC was 1·38 (95% CI 1·20-1·60, or one additional case per 1058 patients with UC per 5 years) and the HR for death from CRC was 1·25 (95% CI 1·03-1·51, or one additional case per 3041 patients with UC per 5 years).

INTERPRETATION

Compared with those without UC, individuals with UC are at increased risk of developing CRC, are diagnosed with less advanced CRC, and are at increased risk of dying from CRC, although these excess risks have declined substantially over time. There still seems to be room for improvement in international surveillance guidelines.

FUNDING

The Swedish Medical Society, Karolinska Institutet, Stockholm County Council, Swedish Research Council, Swedish Foundation for Strategic Research, Independent Research Fund Denmark, Forte Foundation, Swedish Cancer Foundation.

Venous thromboembolism and subsequent hospitalisation due to acute arterial cardiovascular events: a 20-year cohort study

BACKGROUND

In some studies, venous thromboembolism has been associated with atherosclerosis and with the risk of arterial cardiovascular events such as myocardial infarction and stroke. Other studies, however, do not show this association. To help clarify these discrepant findings, we aimed to investigate the risk of arterial cardiovascular events in patients who were diagnosed with venous thromboembolism.

METHODS

We undertook a 20-year population-based cohort study using data from nationwide Danish medical databases. After excluding those with known cardiovascular disease, we assessed the risk of myocardial infarction and stroke in 25,199 patients with deep venous thrombosis, 16,925 patients with pulmonary embolism, and 163,566 population controls.

FINDINGS

For patients with deep venous thrombosis, the relative risks varied from 1.60 for myocardial infarction (95% CI 1.35-1.91) to 2.19 (1.85-2.60) for stroke in the first year after the thrombotic event. For patients with pulmonary embolism, the relative risks in that year were 2.60 (2.14-3.14) for myocardial infarction and 2.93 (2.34-3.66) for stroke. The relative risks were also raised, though less markedly, during the subsequent 20 years of follow-up, with 20-40% increases in risk for arterial cardiovascular events. Relative risks were similar for those with provoked and unprovoked deep venous thrombosis and pulmonary embolism.

INTERPRETATION

Patients with venous thromboembolism have a substantially increased long-term risk of subsequent arterial cardiovascular events.

Existing data sources for clinical epidemiology: the Danish National Pathology Registry and Data Bank

Diagnostic histological and cytological specimens are routinely stored in pathology department archives. These biobanks are a valuable research resource for many diseases, particularly if they can be linked to high quality population-based health registries, allowing large retrospective epidemiological studies to be carried out. Such studies are of significant importance, for example in the search for novel prognostic and predictive biomarkers in the era of personalized medicine. Denmark has a wealth of highly-regarded population-based registries that are ideally suited to conduct this type of epidemiological research. We describe two recent additions to these databases: the Danish National Pathology Registry (DNPR) and its underlying national online registration database, the Danish Pathology Data Bank (DPDB). The DNPR and the DPDB contain detailed nationwide records of all pathology specimens analyzed in Denmark since 1997, and an incomplete but nonetheless valuable record of specimens from some pathology departments dating back to the 1970s. The data are of high quality and completeness and are sufficient to allow precise and efficient localization of the specimens. We describe the relatively uncomplicated procedures required to use these pathology databases in clinical research and to gain access to the archived specimens.

Arterial events, venous thromboembolism, thrombocytopenia, and bleeding after vaccination with Oxford-AstraZeneca ChAdOx1-S in Denmark and Norway: population based cohort study

OBJECTIVE

To assess rates of cardiovascular and haemostatic events in the first 28 days after vaccination with the Oxford-AstraZeneca vaccine ChAdOx1-S in Denmark and Norway and to compare them with rates observed in the general populations.

DESIGN

Population based cohort study.

SETTING

Nationwide healthcare registers in Denmark and Norway.

PARTICIPANTS

All people aged 18-65 years who received a first vaccination with ChAdOx1-S from 9 February 2021 to 11 March 2021. The general populations of Denmark (2016-18) and Norway (2018-19) served as comparator cohorts.

MAIN OUTCOME MEASURES

Observed 28 day rates of hospital contacts for incident arterial events, venous thromboembolism, thrombocytopenia/coagulation disorders, and bleeding among vaccinated people compared with expected rates, based on national age and sex specific background rates from the general populations of the two countries.

RESULTS

The vaccinated cohorts comprised 148 792 people in Denmark (median age 45 years, 80% women) and 132 472 in Norway (median age 44 years, 78% women), who received their first dose of ChAdOx1-S. Among 281 264 people who received ChAdOx1-S, the standardised morbidity ratio for arterial events was 0.97 (95% confidence interval 0.77 to 1.20). 59 venous thromboembolic events were observed in the vaccinated cohort compared with 30 expected based on the incidence rates in the general population, corresponding to a standardised morbidity ratio of 1.97 (1.50 to 2.54) and 11 (5.6 to 17.0) excess events per 100 000 vaccinations. A higher than expected rate of cerebral venous thrombosis was observed: standardised morbidity ratio 20.25 (8.14 to 41.73); an excess of 2.5 (0.9 to 5.2) events per 100 000 vaccinations. The standardised morbidity ratio for any thrombocytopenia/coagulation disorders was 1.52 (0.97 to 2.25) and for any bleeding was 1.23 (0.97 to 1.55). 15 deaths were observed in the vaccine cohort compared with 44 expected.

CONCLUSIONS

Among recipients of ChAdOx1-S, increased rates of venous thromboembolic events, including cerebral venous thrombosis, were observed. For the remaining safety outcomes, results were largely reassuring, with slightly higher rates of thrombocytopenia/coagulation disorders and bleeding, which could be influenced by increased surveillance of vaccine recipients. The absolute risks of venous thromboembolic events were, however, small, and the findings should be interpreted in the light of the proven beneficial effects of the vaccine, the context of the given country, and the limitations to the generalisability of the study findings.

30-Year Mortality After Venous Thromboembolism

Background—

Studies on long-term mortality after venous thromboembolism (VTE) are sparse.

Methods and Results—

Using Danish medical databases, we conducted a 30-year nationwide population-based cohort study of 128 223 patients with first-time VTE (1980–2011) and a comparison cohort of 640 760 people from the general population (without VTE) randomly matched by sex, year of birth, and calendar period. The mortality risks for patients with deep venous thrombosis (DVT) and pulmonary embolism (PE) were markedly higher than for the comparison cohort during the first year, especially within the first 30 days (3.0% and 31% versus 0.4%). Using Cox regression, we assessed mortality rate ratios (MRRs) with 95% confidence intervals (CIs). The overall 30-year MRR was 1.55 (95% CI, 1.53–1.57) for DVT and 2.77 (95% CI, 2.74–2.81) for PE. The 30-day MRR was 5.38 (95% CI, 5.00–5.80) for DVT and 80.87 (95% CI, 76.02–86.02) for PE. Over time, the 30-day MRR was consistently 5- to 6-fold increased for DVT, whereas it improved for PE from 138 (95% CI, 125–153) in 1980 to 1989 to 36.08 (95% CI, 32.65–39.87) in 2000 to 2011. The 1- to 10-year and 11- to 30-year MRRs remained 25% to 40% increased after both DVT and PE but were 3- to 5-fold increased after DVT and 6- to 11-fold increased after PE when VTE was considered the immediate cause of death.

Conclusions—

Patients with VTE are at increased risk of dying, especially within the first year after diagnosis, but also during the entire 30 years of follow-up, with VTE as an important cause of death. Although 30-day mortality after DVT remained fairly constant over the last 3 decades, it improved markedly for PE.

Thromboembolic risk among Danish children and adults with inflammatory bowel diseases: a population-based nationwide study

BACKGROUND

Recommendations for venous thromboembolism (VTE) prophylaxis in patients with inflammatory bowel disease (IBD) can be refined by incorporating patient-specific risk factors.

OBJECTIVES

To determine the risk of deep venous thrombosis (DVT) and pulmonary embolism (PE) in children and adults with Crohn's disease and ulcerative colitis and evaluate whether this risk varies by age and/or presence of other risk factors.

METHODS

We performed a cohort study using Danish administrative data. Incidence rates of DVT and PE were calculated among patients with IBD and an age- and gender-matched comparison population and compared using Cox proportional hazards regression. We performed additional analyses stratifying by age, gender and disease type and restricting outcomes to unprovoked events (occurring without known malignancy, surgery, fracture/trauma or pregnancy). We next performed a nested case-control study to adjust for additional co-morbidities (congestive heart failure, diabetes, myocardial infarction and stroke) and the use of hormone replacement and antipsychotic medications.

RESULTS

The study included 49,799 patients with IBD (14,211 Crohn's disease, 35,229 ulcerative colitis) and 477,504 members of the general population. VTE risk was elevated in patients with IBD (HR=2.0 (95% CI 1.8 to 2.1) for total events, HR=1.6 (95% CI 1.5 to 1.8) for unprovoked events). Although the incidence of VTE increased with age, the RR was higher in younger patients. Among those ≤ 20 years old, HRs were 6.0 (95% CI 2.5 to 14.7) for DVT and 6.4 (95% CI 2.0 to 20.3) for PE. After further adjusting for co-morbidity and medication use in the case-control analysis, ORs for all events remained in the 1.5-1.8 range.

DISCUSSION

Patients with IBD have twice the incidence of PE or DVT as does the general population. This risk persisted after taking into account other VTE risk factors. Relative risks were particularly high at young ages, though actual incidence increased with age. These findings can further inform risk-benefit analysis of VTE prophylaxis.

Improved long-term clinical outcomes in patients with ST-elevation myocardial infarction undergoing remote ischaemic conditioning as an adjunct to primary percutaneous coronary intervention

AIMS

Remote ischaemic conditioning as an adjunct to primary percutaneous coronary intervention in patients with ST-elevation myocardial infarction increases myocardial salvage. We investigated the effect of remote ischaemic conditioning on long-term clinical outcome.

METHODS AND RESULTS

From February 2007 to November 2008, 333 patients with a suspected first acute ST-elevation myocardial infarction were randomized to receive primary percutaneous coronary intervention with (n = 166) or without (n = 167) remote ischaemic conditioning (intermittent arm ischaemia through four cycles of 5-min inflation followed by 5-min deflation of a blood-pressure cuff). Patient follow-up extended from the randomization date until an outcome, emigration or January 2012 (median follow-up = 3.8 years). The primary endpoint was major adverse cardiac and cerebrovascular events (MACCE)-a composite of all-cause mortality, myocardial infarction, readmission for heart failure, and ischaemic stroke/transient ischaemic attack. The individual components of the primary endpoint comprised the secondary endpoints. Outcomes were obtained from Danish nationwide medical registries and validated by medical record review and contact to patients' general practitioner. In the per-protocol analysis of 251 patient fulfilling trial criteria, MACCE occurred for 17 (13.5%) patients in the intervention group compared with 32 (25.6%) patients in the control group, yielding a hazard ratio (HR) of 0.49 (95% confidence interval: 0.27-0.89, P = 0.018). The HR for all-cause mortality was 0.32 (95% confidence interval: 0.12-0.88, P = 0.027). Although lower precision, the HRs were also directionally lower for all other secondary endpoints.

CONCLUSION

Remote ischaemic conditioning before primary percutaneous coronary intervention seemed to improve long-term clinical outcomes in patients with ST-elevation myocardial infarction.

Risk of adverse birth outcome and miscarriage in pregnant users of non-steroidal anti-inflammatory drugs: population based observational study and case-control study

OBJECTIVE

To estimate the risk of adverse birth outcome in women who take non-steroidal anti-inflammatory drugs during pregnancy.

DESIGN AND SETTING

Population based cohort study and a case-control study, both based on data from a prescription registry, the Danish birth registry, and one county's hospital discharge registry.

PARTICIPANTS

COHORT STUDY: 1462 pregnant women who had taken up prescriptions for non-steroidal anti-inflammatory drugs in the period from 30 days before conception to birth and 17 259 pregnant women who were not prescribed any drugs during pregnancy. CASE-CONTROL STUDY: 4268 women who had miscarriages, of whom 63 had taken non-steroidal anti-inflammatory drugs, and 29 750 primiparous controls who had live births.

MAIN OUTCOME MEASURES

Incidences of congenital abnormality, low birth weight, preterm birth, and miscarriage.

RESULTS

Odds ratios for congenital abnormality, low birth weight, and preterm birth among women who took up prescriptions for non-steroidal anti-inflammatory drugs were 1.27 (95% confidence interval 0.93 to 1.75), 0.79 (0.45 to 1.38), and 1.05 (0.80 to 1.39) respectively. Odds ratios for the taking up of prescriptions in the weeks before miscarriage ranged from 6.99 (2.75 to 17.74) when prescriptions were taken up during the last week before the miscarriage to 2.69 (1.81 to 4.00) when taken up between 7 and 9 weeks before. The risk estimates were no different when the analysis was restricted to missed abortions.

CONCLUSIONS

Use of non-steroidal anti-inflammatory drugs during pregnancy does not seem to increase the risk of adverse birth outcome but is associated with increased risk of miscarriage.

Stent Thrombosis, Myocardial Infarction, and Death After Drug-Eluting and Bare-Metal Stent Coronary Interventions

OBJECTIVES

The aim of the study was to examine outcomes subsequent to implantation of drug-eluting stents (DES) and bare-metal stents (BMS).

BACKGROUND

Use of DES might be associated with increased risk of stent thrombosis (ST), myocardial infarction (MI), and death.

METHODS

From January 2002 through June 2005, data from all percutaneous coronary interventions in western Denmark were prospectively recorded in the Western Denmark Heart Registry; 12,395 consecutive patients (17,152 lesions) treated with stent implantation were followed for 15 months. Data on death and MI were ascertained from the national databases. The Academic Research Consortium definition of ST was used.

RESULTS

The DES were implanted in 3,548 patients (5,422 lesions) and BMS were implanted in 8,847 patients (11,730 lesions). Definite, probable, or possible ST was found in 190 (2.15%) patients in the BMS group and in 64 (1.80%) patients in the DES. The risk of definite ST was similar in the 2 groups (DES: 0.65%; BMS: 0.61%). Very late definite ST (between 12 and 15 months after implantation) occurred more frequently in patients receiving DES (hazard ratio [HR] 10.93, 95% confidence interval [CI] 1.27 to 93.76). Also, the risk of MI between 12 and 15 months after implantation was higher in the DES group (HR 4.00, 95% CI 2.06 to 7.79). Mortality was similar in the 2 groups. Target lesion revascularization was reduced by 43% in patients treated with DES (HR 0.57, 95% CI 0.48 to 0.67).

CONCLUSIONS

The minor risk of ST and MI within 15 months after implantation of DES seems unlikely to outweigh the benefit of these stents.

Sustained Progression-Free Survival Benefit of Rituximab Maintenance in Patients With Follicular Lymphoma: Long-Term Results of the PRIMA Study

PURPOSE

The PRIMA study (ClinicalTrials.gov identifier: NCT00140582) established that 2 years of rituximab maintenance after first-line immunochemotherapy significantly improved progression-free survival (PFS) in patients with follicular lymphoma compared with observation. Here, we report the final PFS and overall survival (OS) results from the PRIMA study after 9 years of follow-up and provide a final overview of safety.

METHODS

Patients (> 18 years of age) with previously untreated high-tumor-burden follicular lymphoma were nonrandomly assigned to receive one of three immunochemotherapy induction regimens. Responding patients were randomly assigned (stratified by induction regimen, response to induction treatment, treatment center, and geographic region) 1:1 to receive 2 years of rituximab maintenance (375 mg/m², once every 8 weeks), starting 8 weeks after the last induction treatment, or observation (no additional treatment). All patients in the extended follow-up provided their written informed consent (data cutoff: December 31, 2016).

RESULTS

In total, 1,018 patients completed induction treatment and were randomly assigned to rituximab maintenance (n = 505) or observation (n = 513). Consent for the extended follow-up was provided by 607 patients (59.6%) of 1,018 (rituximab maintenance, n = 309; observation, n = 298). After data cutoff, median PFS was 10.5 years in the rituximab maintenance arm compared with 4.1 years in the observation arm (hazard ratio, 0.61; 95% CI, 0.52 to 0.73; P < .001). No OS difference was seen in patients randomly assigned to rituximab maintenance or observation (hazard ratio, 1.04; 95% CI, 0.77 to 1.40; P = .7948); 10-year OS estimates were approximately 80% in both study arms. No new safety signals were observed.

CONCLUSION

Rituximab maintenance after induction immunochemotherapy provides a significant long-term PFS, but not OS, benefit over observation.

A nationwide non-medical switch from originator infliximab to biosimilar CT-P13 in 802 patients with inflammatory arthritis: 1-year clinical outcomes from the DANBIO registry

OBJECTIVES

According to guidelines, a nationwide non-medical switch from originator (INX, Remicade) to biosimilar infliximab (Remsima, CT-P13) was conducted in Danish patients with rheumatoid arthritis (RA), psoriatic arthritis (PsA) and axial spondyloarthritis (AxSpA). We investigated disease activity before/after switching and retention rates in the DANBIO registry.

METHODS

Disease activities 3 months before and after switch and changes over time were calculated. Flare was defined as change in 28 Joint Disease Activity Score (∆DAS28) ≥1.2 (RA/PsA) or Ankylosing Spondylitis Disease Activity Score (∆ASDAS) ≥1.3 (AxSpA). Crude and adjusted retention rates were compared with a historic cohort of INX-treated patients.

RESULTS

Eight hundred and two patients switched (403 RA/120 PsA/279 AxSpA; 51% women, age (median (IQR): 55 (44-66)) years). Follow-up was 413 (339-442) days. Prior INX treatment duration was 6.8 (4.3-9.5) years. Disease activities were similar 3 months before/after switch. Crude 1-year CT-P13 retention rate (84.1 (95% CI 81.3 to 86.5)) was similar to the historic IFX cohort (86.2 (95% CI 84.0 to 88.0), p=0.22). The adjusted absolute retention rates were 83.4 (95% CI 80.8 to 86.2) and 86.8% (95% CI 84.8 to 88.8), respectively (p=0.03). In total 132 patients withdrew (lack of effect: 71/132=54%, adverse events: 37/132=28%). Patients with previous INX treatment duration >5 years had longer CT-P13 retention.

CONCLUSION

In 802 arthritis patients treated with INX for median >6 years, a nationwide non-medical switch to CT-P13 had no negative impact on disease activity. Adjusted 1-year CT-P13 retention rate was slightly lower than for INX in a historic cohort.

Combining electronic healthcare databases in Europe to allow for large-scale drug safety monitoring: the EU-ADR Project

PURPOSE

In this proof-of-concept paper we describe the framework, process, and preliminary results of combining data from European electronic healthcare record (EHR) databases for large-scale monitoring of drug safety.

METHODS

Aggregated demographic, clinical, and prescription data from eight databases in four countries (Denmark, Italy, Netherlands, the UK) were pooled using a distributed network approach by generation of common input data followed by local aggregation through custom-built software, Jerboa(©). Comparison of incidence rates of upper gastrointestinal bleeding (UGIB) and nonsteroidal anti-inflammatory drug (NSAID) utilization patterns were used to evaluate data harmonization and quality across databases. The known association of NSAIDs and UGIB was employed to demonstrate sensitivity of the system by comparing incidence rate ratios (IRRs) of UGIB during NSAID use to UGIB during all other person-time.

RESULTS

The study population for this analysis comprised 19,647,445 individuals corresponding to 59,929,690 person-years of follow-up. 39,967 incident cases of UGIB were identified during the study period. Crude incidence rates varied between 38.8 and 109.5/100,000 person-years, depending on country and type of database, while age-standardized rates ranged from 25.1 to 65.4/100,000 person-years. NSAID use patterns were similar for databases within the same country but heterogeneous among different countries. A statistically significant age- and gender-adjusted association between use of any NSAID and increased risk for UGIB was confirmed in all databases, IRR from 2.0 (95%CI:1.7-2.2) to 4.3 (95%CI: 4.1-4.5).

CONCLUSIONS

Combining data from EHR databases of different countries to identify drug-adverse event associations is feasible and can set the stage for changing and enlarging the scale for drug safety monitoring.

Prognostic significance of thrombocytosis in patients with primary lung cancer

In patients with malignancies, thrombocytosis has previously been related to disease stage, histological type, and survival. In the present study, the prevalence of thrombocytosis and the prognostic information provided by platelet counts were analysed in a large cohort of patients with primary lung cancer. At the time of diagnosis, pretreatment platelet counts were retrospectively recorded in 1,115 consecutive patients with histologically proven primary lung cancer. All patients were reviewed regarding histological type, tumour, node, metastasis (TNM) classification stage and survival. The prevalence of thrombocytosis in patients with lung cancer was compared with that in a series of 550 consecutive out-patients with benign lung disorders. In 269 surgically resected patients, postoperative platelet counts were recorded 1-3 months after resection of the tumour. In the follow-up period, thromboembolic episodes diagnosed either clinically or at autopsy were recorded. The overall prevalence of thrombocytosis (> 400 x 10(9) platelets.L-1) in the patients with lung cancer was 32%. The frequency of thrombocytosis was significantly higher compared with the control subjects (32 vs 6%; p < 0.0001). Platelet counts differed significantly among subgroups defined by the TNM classification, with the proportion of patients with > 400 x 10(9) platelets.L-1 greatest in the more advanced TNM stages (stage I and II 23% vs stage III and IV 37%; p < 0.0001). Patients with thrombocytosis had a significantly poorer survival than patients with normal platelet counts (p < 0.0001). In a multivariate survival analysis (Cox model), thrombocytosis continued to correlate strongly with poor survival even when adjusted for histological type, sex, age, and TNM stage (p < 0.001). In surgically resected patients, the frequency of preoperative and postoperative thrombocytosis differed significantly (23.0 vs 8.9%; p < 0.0001). Survival rate was significantly reduced in patients with preoperative thrombocytosis (p = 0.005). Thrombocytosis was not associated with an increased incidence of thromboembolism. In conclusion, thrombocytosis is an independent prognostic factor of survival in patients with primary lung cancer. We suggest that platelet counts should be included in future multivariate analyses of survival in patients with lung cancer.

Capitation, salary, fee-for-service and mixed systems of payment: effects on the behaviour of primary care physicians

BACKGROUND

It is widely believed that the method of payment of physicians may affect their clinical behaviour. Although payment systems may be used to achieve policy objectives (e.g. cost containment or improved quality of care), little is known about the effects of different payment systems in achieving these objectives.

OBJECTIVES

To evaluate the impact of different methods of payment (capitation, salary, fee for service and mixed systems of payment) on the clinical behaviour of primary care physicians (PCPs).

SEARCH STRATEGY

We searched the Cochrane Effective Practice and Organisation of Care Group specialised register; the Cochrane Controlled Trials Register; MEDLINE (1966 to October 1997); BIDS EMBASE (1980 to October 1997); BIDS ISI (1981 to October 1997); EconLit (1969 to October 1997); HealthStar (1975 to October 1997) Helmis (1984 to October 1997); health economics discussion paper series of the Universities of York, Aberdeen, Sheffield, Bristol, Brunel, and McMaster; Swedish Institute of Health Economics; RAND corporation; and reference lists of articles.

SELECTION CRITERIA

Randomised trials, controlled before and after studies and interrupted time series analyses of interventions comparing the impact of capitation, salary, fee for service (FFS) and mixed systems of payment on primary care physician satisfaction with working environment; cost and quantity of care; type and pattern of care; equity of care; and patient health status and satisfaction.

DATA COLLECTION AND ANALYSIS

Two reviewers independently extracted data and assessed study quality.

MAIN RESULTS

Four studies were included involving 640 primary care physicians and more than 6400 patients. There was considerable variation in study setting and the range of outcomes measured. FFS resulted in more primary care visits/contacts, visits to specialists and diagnostic and curative services but fewer hospital referrals and repeat prescriptions compared with capitation. Compliance with a recommended number of visits was higher under FFS compared with capitation payment. FFS resulted in more patient visits, greater continuity of care, higher compliance with a recommended number of visits, but patients were less satisfied with access to their physician compared with salaried payment.

REVIEWER'S CONCLUSIONS

It is noteworthy that so few studies met the inclusion criteria. There is some evidence to suggest that the method of payment of primary care physicians affects their behaviour, but the findings' generalisability is unknown. More evaluations of the effect of payment systems on PCP behaviour are needed, especially in terms of the relative impact of salary versus capitation payments.

Effect modification, interaction and mediation: an overview of theoretical insights for clinical investigators

We revisited the three interrelated epidemiological concepts of effect modification, interaction and mediation for clinical investigators and examined their applicability when using research databases. The standard methods that are available to assess interaction, effect modification and mediation are explained and exemplified. For each concept, we first give a simple “best-case” example from a randomized controlled trial, followed by a structurally similar example from an observational study using research databases. Our explanation of the examples is based on recent theoretical developments and insights in the context of large health care databases. Terminology is sometimes ambiguous for what constitutes effect modification and interaction. The strong assumptions underlying the assessment of interaction, and particularly mediation, require clinicians and epidemiologists to take extra care when conducting observational studies in the context of health care databases. These strong assumptions may limit the applicability of interaction and mediation assessments, at least until the biases and limitations of these assessments when using large research databases are clarified.

Increased Risk of Colorectal Cancer Development Among Patients With Serrated Polyps

BACKGROUND & AIMS

Sessile serrated adenomas/polyps (SSA/Ps) and traditional serrated adenomas (TSAs) are now distinguished from hyperplastic polyps and recognized as precursors to colorectal cancer (CRC). We studied CRC risks associated with serrated polyps.

METHODS

By using Danish databases (1977-2009), we conducted a nationwide population-based, case-control study nested within individuals who had received colonoscopies (n = 272,342), and identified 2045 CRC cases and 8105 CRC-free individuals (controls). For each case and control, we identified the first colorectal polyp(s) that underwent a biopsy or were excised during or after the initial colonoscopy, and obtained tissue blocks for hyperplastic lesions. Four expert pathologists reviewed these lesions using current terminology for serrated polyps. We used logistic regression to compute odds ratios (ORs) to associate the risk of CRC with polyp type and estimated the absolute risks by multiplying the risk in patients with no polyps by these ORs.

RESULTS

Seventy-nine cases and 142 controls had SSA/Ps (OR, 3.07; 95% confidence interval [CI], 2.30-4.10). SSA/Ps with cytology markers of dysplasia were associated with a particularly high OR (4.76; 95% CI, 2.59-8.73). Women with SSA/P had a higher risk for CRC than men with SSA/P (OR for women, 5.05; 95% CI, 3.05-8.37 vs OR for men, 2.18; 95% CI, 1.24-3.82); patients with SSA/P proximal to the splenic flexure had the highest risk for CRC (OR, 12.42; 95% CI, 4.88-31.58). The OR for CRC was 4.84 in the 14 cases vs 17 controls with TSAs (95% CI, 2.36-9.93), 2.51 in the 757 cases vs 1698 controls with conventional adenomas (95% CI, 2.25-2.80), and 1.30 in the 55 cases vs 235 controls with hyperplastic polyps (95% CI, 0.96-1.77). The 10-year risk for CRC was 4.4% for patients with SSA/P with dysplasia, 4.5% for patients with TSAs, and 2.3% for patients with conventional adenomas.

CONCLUSION

Patients with SSA/P or TSA are at increased risk for CRC; their level of risk is similar to or higher than that of patients with conventional adenomas.

Existing Data Sources in Clinical Epidemiology: Laboratory Information System Databases in Denmark

Routine biomarker results from hospital laboratory information systems, covering hospitals and general practitioners, in Denmark are available to researchers through access to the regional Clinical Laboratory Information System Research Database at Aarhus University and the nationwide Register of Laboratory Results for Research. This review describes these two data sources. The laboratory databases have different geographical and temporal coverage. They both include individual-level biomarker results that are electronically transferred from laboratory information systems. The biomarker results can be linked to all other Danish registries at the individual level, using the unique identifier, the CPR number. The databases include variables such as the CPR number, date and time (hour and minute) of sampling, NPU code, and name of the biomarker, identification code for the laboratory and the requisitioner, the test result with the corresponding unit, and the lower and upper reference limits. Access to the two databases differs since they are hosted by two different institutions. Data cannot be transferred outside Denmark, and direct access is provided only to Danish institutions. It is concluded that access to data on routine biomarkers expands the detailed biological and clinical information available on patients in the Danish healthcare system. The full potential is enabled through linkage to other Danish healthcare registries.