Record linkage for pharmacoepidemiological studies in cancer patients

Prescription ranitidine use and population exposure in 6 Canadian provinces, 1996 to 2019: a serial cross-sectional analysis

BACKGROUND

Ranitidine was the most prescribed histamine-2 receptor antagonist (H2RA) in Canada when recalled in 2019 because of potential carcinogenicity. We sought to compare geographic and temporal patterns in use of prescription ranitidine and 3 other HRAs and estimated population exposure to ranitidine in 6 provinces between 1996 and 2019.

METHODS

This population-based serial cross-sectional study used prescription claims for H2RAs dispensed from community pharmacies in Nova Scotia, Ontario, Manitoba, Saskatchewan, Alberta and British Columbia. We estimated the period prevalence of ranitidine use per 100 population by province, age category and sex. We estimated exposure to ranitidine between 2015 and 2019 using defined daily doses (DDDs).

RESULTS

Overall, 2.4 million ranitidine prescriptions were dispensed to patients aged 65 years and older, and 1.7 million were dispensed to younger adults. Among older adults, the median period prevalence of ranitidine use among females was 16% (interquartile range [IQR] 13%-27%) higher than among males. Among younger adults, the median prevalence was 50% (IQR 37%-70%) higher among females. Among older adults, between 1996 and 1999, use was highest in Nova Scotia (33%) and Ontario (30%), lower in the prairies (Manitoba [18%], Saskatchewan [26%], Alberta [17%]) and lowest in BC (11%). By 2015-2019, use of ranitidine among older adults dropped by at least 50% in all provinces except BC. We estimate that at least 142 million DDDs of prescribed ranitidine were consumed annually in 6 provinces (2015-2019).

INTERPRETATION

Over the 24-year period in 6 provinces, patients aged 65 years and older were dispensed 2.4 million prescriptions of ranitidine and younger adults were dispensed 1.7 million prescriptions of ranitidine. These estimates of ranitidine exposure can be used for planning studies of cancer risk and identifying target populations for cancer surveillance.

Cancer incidence and outcomes registries in an Australian context: a systematic review

BACKGROUND

Multiple cancer registries in Australia are used to track the incidence of cancer and the outcomes of their treatment. These registries can be broadly classed into a few types with an increasing number of registries comes a greater potential for collaboration and linkage. This article aims to critically review cancer registry types in Australia and evaluate the Australian Cancer registry landscape to identify these areas.

METHODS

A systematic review was performed through MEDLINE, EMBASE and Cochrane Library, updated to September 2022 using a predefined search strategy. Inclusion criteria were those that only analysed Australian and/or New Zealand based cancer registries, appraised the utility of cancer outcomes and/or incidence registries, and explored the utility of linked databases using cancer outcomes and/or incidence registries. The grey literature was searched for all operating cancer registries in Australia. Details of registry infrastructure was extracted for analysis and comparison.

RESULTS

Three thousand two hundred and sixteen articles identified from the three databases. Twelve met the inclusion criteria. Twenty-eight registries were identified using the grey literature. Strengths and weaknesses of Cancer Outcome Registries(COR) and Cancer Incidence Registries(CIR) were compared. Data linkage between registries or with other healthcare databases show great benefits in improving evidence for cancer research but are challenging to implement. Both registry types utilize differing modes of administration, influencing their accuracy and completeness.

CONCLUSION

Outcome registries provide detailed data but their weakness lies in incomplete data coverage. Incidence registries record a large dataset which contain inaccuracies. Improving coverage of quality outcome registries, and quality assurance of data in incidence registries is required to ensure collection of accurate, meaningful data. Areas for collaboration identified included establishment of defined definitions and outcomes, data linkage between registry types or with healthcare databases, and collaboration in logistical planning to improve clinical utility of cancer registries.

FAIRifizierung von Real World Data für die Gesundheitsforschung

Hintergrund

Die Bereitstellung von Real-World-Daten im Sinne der FAIR-Prinzipien ist die Voraussetzung einer effizienten Ausschöpfung des Potenzials von Gesundheitsdaten für Prävention und Versorgung.

Ziel der Arbeit

Möglichkeiten und Limitationen der Nachnutzung und Verknüpfung von Gesundheitsdaten in Deutschland werden dargestellt.

Material und Methoden

Es werden Initiativen zur Schaffung einer verbesserten Forschungsdateninfrastruktur vorgestellt und an einem Beispiel die Einschränkungen illustriert, die das Record Linkage personenbezogener Gesundheitsdaten behindern.

Ergebnisse

In der Regel erfüllen Gesundheitsdaten in Deutschland nicht die Anforderungen der FAIR-Prinzipien. Ihre Auffindbarkeit scheitert bereits daran, dass entweder keine Metadaten zur Verfügung stehen oder diese nicht standardisiert in suchbare Repositorien eingestellt werden. Die Verknüpfung von personenbezogenen Gesundheitsdaten ist durch restriktive Datenschutzbestimmungen und das Fehlen eines sog. Unique Identifiers extrem eingeschränkt. Datenschutzkonforme Lösungen für die Verknüpfung von Gesundheitsdaten, die in europäischen Nachbarländern erfolgreich praktiziert werden, könnten hier als Vorbild dienen.

Schlussfolgerung

Die Schaffung einer Nationalen Forschungsdateninfrastruktur (NFDI), insbesondere für personenbezogene Gesundheitsdaten (NFDI4Health), ist nur mit erheblichen Anstrengungen und Gesetzesänderungen realisierbar. Bereits vorliegende Strukturen und Standards, wie sie z. B. durch die Medizininformatik-Initiative und das Netzwerk Universitätsmedizin geschaffen wurden, sowie internationale Initiativen wie z. B. die European Open Science Cloud müssen dabei berücksichtigt werden.

Use of Linked Databases for Improved Confounding Control: Considerations for Potential Selection Bias

Pharmacoepidemiologic studies are increasingly conducted within linked databases, often to obtain richer confounder data. However, the potential for selection bias is frequently overlooked when linked data is available only for a subset of patients. We highlight the importance of accounting for potential selection bias by evaluating the association between antipsychotics and type 2 diabetes in youths within a claims database linked to a smaller laboratory database. We used inverse probability of treatment weights (IPTW) to control for confounding. In analyses restricted to the linked cohorts, we applied inverse probability of selection weights (IPSW) to create a population representative of the full cohort. We used pooled logistic regression weighted by IPTW only or IPTW and IPSW to estimate treatment effects. Metabolic conditions were more prevalent in linked cohorts compared with the full cohort. Within the full cohort, the confounding-adjusted hazard ratio was 2.26 (95% CI: 2.07, 2.49) comparing initiation of antipsychotics with initiation of control medications. Within the linked cohorts, a different magnitude of association was obtained without adjustment for selection, whereas applying IPSW resulted in point estimates similar to the full cohort's (e.g., an adjusted hazard ratio of 1.63 became 2.12). Linked database studies may generate biased estimates without proper adjustment for potential selection bias.

Temporal trends in healthcare resource use and associated costs of patients with cystic fibrosis

BACKGROUND

Better insights into the natural course of cystic fibrosis (CF) have led to treatment approaches that have improved pulmonary health and increased the life expectancy of affected individuals. This study evaluated how the combination of modified demographics and changes in CF management impacted resource consumption and the cost of care.

METHODS

Medical records of CF patients from 2006 to 2016 in the French CF Registry were linked to their corresponding claims data (SNDS). Medications, medical visits, procedures, hospitalisations, and indirect costs were annualized by calendar year from 2006 to 2017.

RESULTS

Of the 7,671 patients included in the French CF Registry, 6,187 patients (80.7%) were linked to the SNDS (51.9% male, mean age = 24.7 years). The average cost per patient was €14,174 in 2006, €21,920 in 2011 and €44,585 in 2017. Costs associated with hospital stays increased from €3,843 per patient in 2006 to €6,741 in 2017. In 2017, the mean cost per CF patient was allocated as follows: 72% for medications (of which 51% for modulator therapies), 15% for hospital stays, 7% for medical visits, 3% for indirect costs, 2% for medical devices, 1% for outpatient medical procedures.

CONCLUSION

There was a strong increase in the mean annual cost per CF patient between 2006 and 2017, mostly due to the cost of therapy after the introduction of cystic fibrosis transmembrane conductance regulator (CFTR) modulators. The combination of an increase in the number of CF patients - particularly adult patients - and an increase in the annual cost per patient led to a substantial increase in the total cost of CF disease care for the health systems.

The Value of Real-World Data in Understanding Prostate Cancer Risk and Improving Clinical Care: Examples from Swedish Registries

Simple Summary

Real-world data (RWD), i.e., data reflecting normal clinical practice collected outside the constraints of randomised controlled trials, provide important insights into our understanding of prostate cancer and its management. Clinical cancer registries are an important source of RWD. Depending on their scope and the potential linkage to other data sources, registry-based data can be utilised to address a variety of questions including risk factors, healthcare utilisation, treatment effectiveness, adverse effects, disparities in healthcare access, quality of care and healthcare economics. This review describes the various registry-based RWD sources for prostate cancer research in Sweden (namely the National Prostate Cancer Register, the Prostate Cancer data Base Sweden (PCBaSe) and the Patient-overview Prostate Cancer) and documents their utility for better understanding prostate cancer aetiology and improving clinical care.

Abstract

Real-world data (RWD), that is, data from sources other than controlled clinical trials, play an increasingly important role in medical research. The development of quality clinical registers, increasing access to administrative data sources, growing computing power and data linkage capacities have contributed to greater availability of RWD. Evidence derived from RWD increases our understanding of prostate cancer (PCa) aetiology, natural history and effective management. While randomised controlled trials offer the best level of evidence for establishing the efficacy of medical interventions and making causal inferences, studies using RWD offer complementary evidence about the effectiveness, long-term outcomes and safety of interventions in real-world settings. RWD provide the only means of addressing questions about risk factors and exposures that cannot be “controlled”, or when assessing rare outcomes. This review provides examples of the value of RWD for generating evidence about PCa, focusing on studies using data from a quality clinical register, namely the National Prostate Cancer Register (NPCR) Sweden, with longitudinal data on advanced PCa in Patient-overview Prostate Cancer (PPC) and data linkages to other sources in Prostate Cancer data Base Sweden (PCBaSe).

Use of Topical Tacrolimus and Topical Pimecrolimus in Four European Countries: A Multicentre Database Cohort Study

Background

Despite the concerns about a potential increased risk of skin cancer and lymphoma with the use of topical tacrolimus and pimecrolimus, no population-based studies have given an overview of the use of these drugs in Europe.

Objective

To assess the use of topical tacrolimus and pimecrolimus in children and adults in Europe.

Methods

Multicentre database cohort study comprising data from the Netherlands, Denmark, Sweden and the UK. We analysed users of topical tacrolimus and pimecrolimus starting from the date of first availability (between 2002 and 2003) or start establishment of the prescription database in Sweden (2006) through 2011. Use was assessed separately for children (≤ 18 years) and adults.

Results

32,052 children and 104,902 adults were treated with topical tacrolimus, and 32,125 children and 58,280 adults were treated with topical pimecrolimus. The number of users increased rapidly after first availability, especially for topical tacrolimus. Topical tacrolimus was more frequently used in all countries except Denmark. For both drugs, there was a decrease in users after 2004 in the Netherlands and Denmark and after 2005 in the UK, especially among children. This decrease was largest in Denmark. The decrease in the number of users was temporary for topical tacrolimus, while use remained relatively low for topical pimecrolimus.

Conclusions

The number of topical tacrolimus and pimecrolimus users increased rapidly after regulatory approval. A transient reduction in topical tacrolimus use and a persistent reduction in topical pimecrolimus use was seen after 2004 in the Netherlands and Denmark and after 2005 in the UK.

The Hoorn Diabetes Care System (DCS) cohort. A prospective cohort of persons with type 2 diabetes treated in primary care in the Netherlands

PURPOSE

People with type 2 diabetes (T2D) have a doubled morbidity and mortality risk compared with persons with normal glucose tolerance. Despite treatment, clinical targets for cardiovascular risk factors are not achieved. The Hoorn Diabetes Care System cohort (DCS) is a prospective cohort representing a comprehensive dataset on the natural course of T2D, with repeated clinical measures and outcomes. In this paper, we describe the design of the DCS cohort.

PARTICIPANTS

The DCS consists of persons with T2D in primary care from the West-Friesland region of the Netherlands. Enrolment in the cohort started in 1998 and this prospective dynamic cohort currently holds 12 673 persons with T2D.

FINDINGS TO DATE

Clinical measures are collected annually, with a high internal validity due to the centrally organised standardised examinations. Microvascular complications are assessed by measuring kidney function, and screening feet and eyes. Information on cardiovascular disease is obtained by 1) self-report, 2) electrocardiography and 3) electronic patient records. In subgroups of the cohort, biobanking and additional measurements were performed to obtain information on, for example, lifestyle, depression and genomics. Finally, the DCS cohort is linked to national cancer and all-cause mortality registers. A selection of published findings from the DCS includes identification of subgroups with distinct development of haemoglobin A1c, blood pressure and retinopathy, and their predictors; validation of a prediction model for personalised retinopathy screening; the assessment of the role of genetics in development and treatment of T2D, providing options for personalised medicine.

FUTURE PLANS

We will continue with the inclusion of persons with newly diagnosed T2D, follow-up of persons in the cohort and linkage to morbidity and mortality registries. Currently, we are involved in (inter)national projects on, among others, biomarkers and prediction models for T2D and complications and we are interested in collaborations with external researchers.

TRIAL REGISTRATION

ISRCTN26257579.

Chemotherapy for ovarian cancer in the Netherlands: a population-based study on treatment patterns and outcomes

Information on treatment patterns for ovarian cancer (OC) is limited. The aim of this study was to describe current patterns of chemotherapy and other systemic treatments for OC in the Netherlands and evaluate survival outcomes following subsequent lines of treatment. Data from the Eindhoven Cancer Registry, including on newly diagnosed cancer patients, were linked to the PHARMO Database Network, including information on in- and out-patient drug use. Patients diagnosed with OC between January 2000 and December 2010 were selected. An algorithm was used to identify separate lines of treatment. Data were studied descriptively. Detailed data on systemic drug use were available for 261 patients (17%) with OC. In first-line treatment, 87% of the patients (227/261) received platinum-based chemotherapy. Of the 161 patients receiving second-line treatment, 101 patients (63%) received platinum-based chemotherapy. In third line, this was 51% (53/103). The median number of treatment lines received by patients was two (interquartile range 1-3), and eight or more lines of chemotherapy were identified for 12 patients. Median survival from diagnosis onwards was 47 months from the end of first-line treatment, median survival was 32 months, and from the end of second-line treatment, it was 14 months. Predominantly beyond second-line treatment, there is much variety in treatment patterns with chemotherapy for OC. Although uncertainty remains regarding the desirability of this observed treatment variation, there seems a need for detailed clinical guidance, assuring that physicians can properly choose the most suitable treatment for each patient.

Asthma-Related Outcomes in Patients Initiating Extrafine Ciclesonide or Fine-Particle Inhaled Corticosteroids

Purpose

Extrafine-particle inhaled corticosteroids (ICS) have greater small airway deposition than standard fine-particle ICS. We sought to compare asthma-related outcomes after patients initiated extrafine-particle ciclesonide or fine-particle ICS (fluticasone propionate or non-extrafine beclomethasone).

Methods

This historical, matched cohort study included patients aged 12-60 years prescribed their first ICS as ciclesonide or fine-particle ICS. The 2 cohorts were matched 1:1 for key demographic and clinical characteristics over the baseline year. Co-primary endpoints were 1-year severe exacerbation rates, risk-domain asthma control, and overall asthma control; secondary endpoints included therapy change.

Results

Each cohort included 1,244 patients (median age 45 years; 65% women). Patients in the ciclesonide cohort were comparable to those in the fine-particle ICS cohort apart from higher baseline prevalence of hospitalization, gastroesophageal reflux disease, and rhinitis. Median (interquartile range) prescribed doses of ciclesonide and fine-particle ICS were 160 (160-160) µg/day and 500 (250-500) µg/day, respectively (P<0.001). During the outcome year, patients prescribed ciclesonide experienced lower severe exacerbation rates (adjusted rate ratio [95% CI], 0.69 [0.53-0.89]), and higher odds of risk-domain asthma control (adjusted odds ratio [95% CI], 1.62 [1.27-2.06]) and of overall asthma control (2.08 [1.68-2.57]) than those prescribed fine-particle ICS. The odds of therapy change were 0.70 (0.59-0.83) with ciclesonide.

Conclusions

In this matched cohort analysis, we observed that initiation of ICS with ciclesonide was associated with better 1-year asthma outcomes and fewer changes to therapy, despite data suggesting more difficult-to-control asthma. The median prescribed dose of ciclesonide was one-third that of fine-particle ICS.

What Can Big Data Offer the Pharmacovigilance of Orphan Drugs?

The pharmacovigilance of drugs for orphan diseases presents problems related to the small patient population. Obtaining high-quality information on individual reports of suspected adverse reactions is of particular importance for the pharmacovigilance of orphan drugs. The possibility of mining "big data" to detect suspected adverse reactions is being explored in pharmacovigilance generally but may have limited application to orphan drugs. Sources of big data such as social media may be infrequently used as communication channels by patients with rare disease or their caregivers or by health care providers; any adverse reactions identified are likely to reflect what is already known about the safety of the drug from the network of support that grows up around these patients. Opportunities related to potential future big data sources are discussed.

Identifying Cases of Type 2 Diabetes in Heterogeneous Data Sources: Strategy from the EMIF Project

Due to the heterogeneity of existing European sources of observational healthcare data, data source-tailored choices are needed to execute multi-data source, multi-national epidemiological studies. This makes transparent documentation paramount. In this proof-of-concept study, a novel standard data derivation procedure was tested in a set of heterogeneous data sources. Identification of subjects with type 2 diabetes (T2DM) was the test case. We included three primary care data sources (PCDs), three record linkage of administrative and/or registry data sources (RLDs), one hospital and one biobank. Overall, data from 12 million subjects from six European countries were extracted. Based on a shared event definition, sixteeen standard algorithms (components) useful to identify T2DM cases were generated through a top-down/bottom-up iterative approach. Each component was based on one single data domain among diagnoses, drugs, diagnostic test utilization and laboratory results. Diagnoses-based components were subclassified considering the healthcare setting (primary, secondary, inpatient care). The Unified Medical Language System was used for semantic harmonization within data domains. Individual components were extracted and proportion of population identified was compared across data sources. Drug-based components performed similarly in RLDs and PCDs, unlike diagnoses-based components. Using components as building blocks, logical combinations with AND, OR, AND NOT were tested and local experts recommended their preferred data source-tailored combination. The population identified per data sources by resulting algorithms varied from 3.5% to 15.7%, however, age-specific results were fairly comparable. The impact of individual components was assessed: diagnoses-based components identified the majority of cases in PCDs (93–100%), while drug-based components were the main contributors in RLDs (81–100%). The proposed data derivation procedure allowed the generation of data source-tailored case-finding algorithms in a standardized fashion, facilitated transparent documentation of the process and benchmarking of data sources, and provided bases for interpretation of possible inter-data source inconsistency of findings in future studies.

Post authorization safety study comparing quetiapine to risperidone and olanzapine

OBJECTIVES

To compare rates of specific adverse outcomes between patients starting quetiapine, olanzapine, or risperidone use in the Netherlands.

METHODS

Observational study using the PHARMO Database Network, including patients starting quetiapine (4658), olanzapine (5856), or risperidone (7229) in 2000-2009, comparing rates of all-cause mortality, failed suicide attempts, extrapyrimidal symptoms (EPS), diabetes mellitus (DM), hypothyroidism, and acute myocardial infarction (AMI).

KEY FINDINGS

Median follow-up until discontinuation/end of follow-up was 0.6 years. Prescribed doses were generally lower than the approved defined daily doses, especially for quetiapine. Quetiapine was significantly associated with lower EPS rates (HR 0.18; 95% CI 0.13-0.24), but higher failed suicide attempt rates (HR 2.07; 95% CI 1.35-3.16) compared to risperidone. Quetiapine was significantly associated with lower EPS rates (HR 0.59; 95% CI 0.42-0.84) and DM rates (HR 0.66; 95% CI 0.44-0.97) compared to olanzapine. Rates for all-cause mortality, hypothyroidism, and stroke were similar between groups. AMI events were too infrequent to draw conclusions.

CONCLUSIONS

Quetiapine was associated with lower EPS, but higher failed suicide attempt rates compared to risperidone. Quetiapine was associated with lower EPS and DM rates compared to olanzapine. The results should be interpreted with caution because of possible channelling and residual confounding. Copyright © 2016 John Wiley & Sons, Ltd.

Epidemiology, Risk Factors, and In-Hospital Mortality of Venous Thromboembolism in Liver Cirrhosis: A Single-Center Retrospective Observational Study

BACKGROUND

Risk of venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), may be increased in liver cirrhosis. We conducted a single-center study to explore the epidemiology, risk factors, and in-hospital mortality of VTE in Chinese patients with liver cirrhosis.

MATERIAL/METHODS

All patients with liver cirrhosis who were consecutively admitted to our hospital between January 2011 and December 2013 were retrospectively included.

RESULTS

Of 2006 patients with liver cirrhosis included, 9 patients were diagnosed with or developed VTE during hospitalization, including 5 patients with a previous history of DVT, 1 patient with either a previous history of DVT or new onset of PE, and 3 patients with new onset of VTE (PE, n=1; DVT, n=2). Risk factors for VTE included a significantly higher proportion of hypertension and significantly higher red blood cells, hemoglobin, alanine aminotransferase, aspartate aminotransferase, prothrombin time (PT), international normalized ratio (INR), D-dimer, and Child-Pugh scores. The in-hospital mortality was significantly higher in patients with VTE than those without VTE (33.3% [3/9] versus 3.4% [67/1997], P<0.001).

CONCLUSIONS

VTE was observed in 0.4% of patients with liver cirrhosis during hospitalization and it significantly increased the in-hospital mortality. Elevated PT/INR aggravated the risk of VTE.

The Epidemiology of Hip and Major Osteoporotic Fractures in a Dutch Population of Community-Dwelling Elderly: Implications for the Dutch FRAX® Algorithm

Background

Incidence rates of non-hip major osteoporotic fractures (MOF) remain poorly characterized in the Netherlands. The Dutch FRAX® algorithm, which predicts 10-year probabilities of hip fracture and MOF (first of hip, humerus, forearm, clinical vertebral), therefore incorporates imputed MOF rates. Swedish incidence rate ratios for hip fracture to MOF (Malmo 1987–1996) were used to perform this imputation. However, equality of these ratios between countries is uncertain and recent evidence is scarce. Aims were to estimate incidence rates of hip fracture and MOF and to compare observed MOF rates to those predicted by the imputation method for the Netherlands.

Methods

Using hospitalisation and general practitioner records from the Dutch PHARMO Database Network (2002–2011) we calculated age-and-sex-specific and age-standardized incidence rates (IRs) of hip and other MOFs (humerus, forearm, clinical vertebral) and as used in FRAX®. Observed MOF rates were compared to those predicted among community-dwelling individuals ≥50 years by the standardized incidence ratio (SIR; 95% CI).

Results

Age-standardized IRs (per 10,000 person-years) of MOF among men and women ≥50 years were 25.9 and 77.0, respectively. These numbers were 9.3 and 24.0 for hip fracture. Among women 55–84 years, observed MOF rates were significantly higher than predicted (SIR ranged between 1.12–1.50, depending on age). In men, the imputation method performed reasonable.

Conclusion

Observed MOF incidence was higher than predicted for community-dwelling women over a wide age-range, while it agreed reasonable for men. As miscalibration may influence treatment decisions, there is a need for confirmation of results in another data source. Until then, the Dutch FRAX® output should be interpreted with caution.

[Benefits of large healthcare databases for drug risk research]

Large electronic healthcare databases have become an important worldwide data resource for drug safety research after approval. Signal generation methods and drug safety studies based on these data facilitate the prospective monitoring of drug safety after approval, as has been recently required by EU law and the German Medicines Act. Despite its large size, a single healthcare database may include insufficient patients for the study of a very small number of drug-exposed patients or the investigation of very rare drug risks. For that reason, in the United States, efforts have been made to work on models that provide the linkage of data from different electronic healthcare databases for monitoring the safety of medicines after authorization in (i) the Sentinel Initiative and (ii) the Observational Medical Outcomes Partnership (OMOP). In July 2014, the pilot project Mini-Sentinel included a total of 178 million people from 18 different US databases. The merging of the data is based on a distributed data network with a common data model. In the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCEPP) there has been no comparable merging of data from different databases; however, first experiences have been gained in various EU drug safety projects. In Germany, the data of the statutory health insurance providers constitute the most important resource for establishing a large healthcare database. Their use for this purpose has so far been severely restricted by the Code of Social Law (Section 75, Book 10). Therefore, a reform of this section is absolutely necessary.

Epidemiology of venous thromboembolism in patients with liver diseases: a systematic review and meta-analysis

The risk of venous thromboembolism (VTE) may be increased in patients with liver diseases. A systematic review and meta-analysis were conducted to analyze the epidemiology of VTE in such patients. All relevant studies were searched via the PubMed, EMBASE, and Cochrane Library databases. The incidence and prevalence of VTE were pooled using random-effect models. Subgroup analyses were conducted according to the type of VTE [deep vein thrombosis (DVT), pulmonary embolism (PE)], type of liver disease (liver cirrhosis alone/unclassified liver diseases or non-cirrhotics), region in which the study was performed (USA/Europe/Asia), number of total observed patients with liver diseases (>1,000/<1,000 patients), study quality (high/low), and methods for identifying the cases (ICD codes/clinical charts). Of 4,843 papers initially identified, 20 were included. The incidence of VTE varied from 0.33 to 6.32 % in 14 studies with a pooled value of 1 % (95 % confidence interval (CI) 0.7-1.3 %). The pooled incidence of DVT and PE was 0.6 % (95 % CI 0.4-0.8 %) and 0.28 % (95 % CI 0.13-0.49 %), respectively. The prevalence of VTE varied from 0.6 to 4.69 % in six studies with a pooled value of 1.0 % (95 % CI 0.7-1.2 %). The pooled prevalence of DVT and PE was 0.7 % (95 % CI 0.6-0.9 %) and 0.36 % (95 % CI 0.13-0.7 %), respectively. The heterogeneity was statistically significant in the main and subgroup meta-analyses. In conclusion, about 1 % of patients with liver diseases develop or are diagnosed with VTE during their hospitalizations. However, the epidemiological data are very heterogeneous among studies.

Burden of Exacerbations in Patients with Moderate to Very Severe COPD in the Netherlands: A Real-life Study

OBJECTIVE

The objective of this study was to compare rates of different types of acute exacerbations of COPD (AECOPDs) and healthcare utilization among patients with different severities of COPD.

METHODS

Data for this study was obtained from the PHARMO Database Network, which includes drug dispensing records from pharmacies, hospitalization records and information from general practitioners. Patients with moderate to very severe COPD (GOLD II-III-IV) and a moderate or severe AECOPD between 2000 and 2010 were included in the study. Moderate and severe AECOPDs were defined by drug use and hospitalizations respectively. Study patients were followed from the first AECOPD to end of registration in PHARMO, death or end of study period, whichever occurred first. During follow-up, all recurrent AECOPDs were characterized and healthcare utilization was assessed.

RESULTS

Of 886 patients in the study, 52% had GOLD-II, 34% GOLD-III and 14% had GOLD-IV. The overall AECOPD recurrence rate per person year (PY) increased from 0.63 for patients with GOLD-II to 1.09 for patients with GOLD-III and 1.33 for patients with GOLD-IV. The rate of severe AECOPD was 0.06, 0.14 and 0.17 per PY, respectively.

CONCLUSION

AECOPD recurrence rates and healthcare utilization are significantly higher among patients with more severe COPD.

Linked health data for pharmacovigilance in children: perceived legal and ethical issues for stakeholders and data guardians

OBJECTIVE

The inclusion of the Community Health Index in the recording of National Health Service (NHS) contacts in Scotland facilitates national linkage of data such as prescribing and healthcare utilisation. This linkage could be the basis for identification of adverse drug reactions. The aim of this article is to report the views of healthcare professionals on data sharing, ownership and the legal and other applicable frameworks relevant to linkage of routinely collected paediatric healthcare data.

DESIGN

Qualitative study using semistructured face-to-face interviews addressing the study aims.

PARTICIPANTS

Purposive sample of professional stakeholders (n=25) including experts on ethics, data protection, pharmacovigilance, data linkage, legal issues and prescribing. Interviews were audio-recorded, transcribed and thematically analysed using a framework approach.

RESULTS

Participants identified existing data sharing systems in the UK. Access to healthcare data should be approved by the data owners. The definition of data ownership and associated legal responsibilities for linked healthcare data were seen as important factors to ensure accountability for the use of linked data. Yet data owners were seen as facilitators of the proposed data linkage. Twelve frameworks (legal, regulatory and governance) applicable to the linkage of healthcare data were identified.

CONCLUSIONS

A large number of potentially relevant legal and regulatory frameworks were identified. Ownership of the linked data was seen as an extension of responsibility for, or guardianship of, the source datasets. The consensus emerging from the present study was that clarity is required on the definition of data sharing, data ownership and responsibilities of data owners.

Validation of a proxy for estrogen receptor status in breast cancer patients using dispensing data

AIM

To assess the performance of a proxy for estrogen receptor (ER) status in breast cancer patients using dispensing data.

METHODS

We derived our proxy using 167 patients. ER+ patients had evidence of at least one dispensing record for hormone therapy during the lookback period, irrespective of diagnosis date and ER- had no dispensing records for hormone therapy during the period. We validated the proxy against our gold standard, ER status from pathology reports or medical records. We assessed the proxy's performance using three lookback periods: 4.5 years, 2 years, 1 year.

RESULTS

More than half of our cohort (62%) were >50 years, 54% had stage III/IV breast cancer at recruitment, (46%) were diagnosed with breast cancer in 2009 and 23% were diagnosed before 2006. Sensitivity and specificity were high for the 4.5 year lookback period (93%, 95% CI: 86-96%; and 95%: 83-99%), respectively) and remained high for the 2-year lookback period (91%: 84-95%; and 95%: 83-99%). Sensitivity decreased (83%: 75.2-89%) but specificity remained high (95%: 83-99%) using the 1-year lookback period and the period is long enough to allow sufficient time for hormone therapy to be dispensed.

CONCLUSION

Our proxy accurately infers ER status in studies of breast cancer treatment based on secondary health data. The proxy is most robust with a minimum lookback period of 2 years.