Declining Incidence of Systemic Lupus Erythematosus in Norway 1999-2017: Data From a Population Cohort Identified by International Classification of Diseases, 10th Revision Code and Verified by Classification

OBJECTIVE

The goal of this study was to provide complete, robust data on annual systemic lupus erythematosus (SLE) incidence rates over nearly two decades from the Southeast Norway area (2.9 million inhabitants) and assess accuracy of SLE-specific International Classification of Diseases (ICD) codes for SLE diagnosis.

METHODS

From administrative databases, we identified all cases International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) coded as SLE during 1999 through 2017 in Southeast Norway. We manually reviewed the chart of every case ICD-10 coded as SLE to either confirm or reject SLE diagnosis. Using SLE classification criteria, we classified all cases with confirmed SLE. We estimated annual incidence rates of classified SLE, and subsets, defined by age at diagnosis, sex, and parental country of birth. The chi-square test was applied for linear time-trend analyses of incidence.

RESULTS

Among the 3,488 cases ICD-10 coded as SLE, chart reviews confirmed SLE diagnosis in 1,558 (45%), of which 797 had new-onset disease during 1999 through 2017. Annual SLE incidence rates fell during 1999 to 2017. The fall was most pronounced in female persons 50 to 59 years old at diagnosis, in whom incidence fell from 3.4 to 1.1 per 100,000 persons (P trend < 0.001). Concurrent ecological data from the study area showed a 74% reduction in prescriptions of menopausal hormone treatment. Accuracy of ICD-10 codes for incident SLE diagnosis was acceptable in juveniles and young adults (up to 20 years) but otherwise low.

CONCLUSION

In a presumably complete population-based cohort, we identified decreasing incidence of SLE, especially among female persons 50 to 59 years old. Although reasons for declining incidence are not clear, ecological data indicate a possible role of environmental factors, for example, menopausal hormone treatments.

Prevalence and Incidence Trends of Attention Deficit/Hyperactivity Disorder in Children and Youth Aged 1-24 Years in Ontario, Canada: A Validation Study of Health Administrative Data Algorithms: Tendances de la prévalence et de l'incidence du trouble de déficit de l'attention/hyperactivité chez les enfants et les jeunes âgés de 1 à 24 ans, en Ontario, Canada: une étude de validation des algorithmes de données administratives de santé

OBJECTIVE

To estimate prevalence and incidence rates over time in children and youth with attention deficit/hyperactivity disorder from the validation of population-based administrative data algorithms using family physicians' electronic medical records as a reference standard.

METHODS

A retrospective cohort study was conducted in Ontario, Canada to identify attention deficit/hyperactivity disorder among children and youth aged 1-24 years in health administrative data derived from case-finding algorithms using family physicians' electronic medical records. Multiple administrative data algorithms identifying attention deficit/hyperactivity disorder cases were developed and tested from physician-diagnosis of attention deficit/hyperactivity disorder in the electronic medical record to determine their diagnostic accuracy. We calculated algorithm performance using sensitivity, specificity, and predictive values. The most optimal algorithm was used to estimate prevalence and incidence rates of attention deficit/hyperactivity disorder from 2014 to 2021 in Ontario.

RESULTS

The optimal performing algorithm was "2 physician visits for attention deficit/hyperactivity disorder in 1 year or 1 attention deficit/hyperactivity disorder-specific prescription" with sensitivity: 83.2% (95% confidence interval [CI], 81.8% to 84.5%), specificity: 98.6% (95% CI, 98.5% to 98.7%), positive predictive value: 78.6% (95% CI, 77.1% to 80.0%) and negative predictive value: 98.9% (95% CI, 98.8% to 99.0%). From 2014, prevalence rates for attention deficit/hyperactivity disorder increased from 5.29 to 7.48 per 100 population in 2021 (N = 281,785). Males had higher prevalence rates (7.49 to 9.59 per 100 population, 1.3-fold increase) than females (2.96-5.26 per 100 population, 1.8-fold increase) from 2014 to 2021. Incidence rates increased from 2014 (0.53 per 100 population) until 2018, decreased in 2020 then rose steeply in 2021 (0.89 per 100 population, N = 34,013). Males also had higher incidence rates than females from 2014 to 2020 with females surpassing males in 2021 (0.70-0.81 per 100 male population,1.2-fold increase versus 0.36-0.97 per 100 female population, 2.7-fold increase).

CONCLUSIONS

Attention deficit/hyperactivity disorder is increasing in prevalence. We developed an administrative data algorithm that can reliably identify children and youth with attention deficit/hyperactivity disorder with good diagnostic accuracy.

Scalable and interpretable alternative to chart review for phenotype evaluation using standardized structured data from electronic health records

OBJECTIVES

Chart review as the current gold standard for phenotype evaluation cannot support observational research on electronic health records and claims data sources at scale. We aimed to evaluate the ability of structured data to support efficient and interpretable phenotype evaluation as an alternative to chart review.

MATERIALS AND METHODS

We developed Knowledge-Enhanced Electronic Profile Review (KEEPER) as a phenotype evaluation tool that extracts patient's structured data elements relevant to a phenotype and presents them in a standardized fashion following clinical reasoning principles. We evaluated its performance (interrater agreement, intermethod agreement, accuracy, and review time) compared to manual chart review for 4 conditions using randomized 2-period, 2-sequence crossover design.

RESULTS

Case ascertainment with KEEPER was twice as fast compared to manual chart review. 88.1% of the patients were classified concordantly using charts and KEEPER, but agreement varied depending on the condition. Missing data and differences in interpretation accounted for most of the discrepancies. Pairs of clinicians agreed in case ascertainment in 91.2% of the cases when using KEEPER compared to 76.3% when using charts. Patient classification aligned with the gold standard in 88.1% and 86.9% of the cases respectively.

CONCLUSION

Structured data can be used for efficient and interpretable phenotype evaluation if they are limited to relevant subset and organized according to the clinical reasoning principles. A system that implements these principles can achieve noninferior performance compared to chart review at a fraction of time.

Postpartum Depression in Reproductive-Age Women With and Without Rheumatic Disease: A Population-Based Matched Cohort Study

OBJECTIVE

To examine postpartum depression (PPD) among women with axial spondyloarthritis (axSpA), psoriatic arthritis (PsA), or rheumatoid arthritis (RA) in comparison with a matched population without rheumatic disease (RD).

METHODS

A retrospective analysis using the 2013-2018 IBM MarketScan Commercial Claims and Encounters Database was conducted. Pregnant women with axSpA, PsA, or RA were identified, and the delivery date was used as the index date. We restricted the sample to women ≤ 55 years with continuous enrollment ≥ 6 months before date of last menstrual period and throughout pregnancy. Each patient was matched with 4 individuals without RD on: (1) maternal age at delivery, (2) prior history of depression, and (3) duration of depression before delivery. Cox frailty proportional hazards models estimated the crude and adjusted hazard ratios (aHR) and 95% CI of incident postpartum depression within 1 year among women with axSpA, PsA, or RA (axSpA/PsA/RA cohort) compared to the matched non-RD comparison group.

RESULTS

Overall, 2667 women with axSpA, PsA, or RA and 10,668 patients without any RD were included. The median follow-up time in days was 256 (IQR 93-366) and 265 (IQR 99-366) for the axSpA/PsA/RA cohort and matched non-RD comparison group, respectively. Development of PPD was more common in the axSpA/PsA/RA cohort relative to the matched non-RD comparison group (axSpA/PsA/RA cohort: 17.2%; matched non-RD comparison group: 12.8%; aHR 1.22, 95% CI 1.09-1.36).

CONCLUSION

Postpartum depression is significantly higher in women of reproductive age with axSpA/PsA/RA when compared to those without RD.

Systematic review of validated case definitions to identify hypertensive disorders of pregnancy in administrative healthcare databases

BACKGROUND

Administrative data are frequently used to study cardiovascular disease (CVD) risk in women with hypertensive disorders of pregnancy (HDP). Little is known about the validity of case-finding definitions (CFDs, eg, disease classification codes/algorithms) designed to identify HDP in administrative databases.

METHODS

A systematic review of the literature. We searched MEDLINE, Embase, CINAHL, Web of Science and grey literature sources for eligible studies. Two independent reviewers screened articles for eligibility and extracted data. Quality of reporting was assessed using checklists; risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool, adapted for administrative studies. Findings were summarised descriptively.

RESULTS

Twenty-six studies were included; most (62%) validated CFDs for a variety of maternal and/or neonatal outcomes. Six studies (24%) reported reference standard definitions for all HDP definitions validated; seven reported all 2×2 table values for ≥1 CFD or they were calculable. Most CFDs (n=83; 58%) identified HDP with high specificity (ie, ≥98%); however, sensitivity varied widely (3%-100%). CFDs validated for any maternal hypertensive disorder had the highest median sensitivity (91%, range: 15%-97%). Quality of reporting was generally poor, and all studies were at unclear or high risk of bias on ≥1 QUADAS-2 domain.

CONCLUSIONS

Even validated CFDs are subject to bias. Researchers should choose the CFD(s) that best align with their research objective, while considering the relative importance of high sensitivity, specificity, negative predictive value and/or positive predictive value, and important characteristics of the validation studies from which they were derived (eg, study prevalence of HDP, spectrum of disease studied, methodological rigour, quality of reporting and risk of bias). Higher quality validation studies on this topic are urgently needed.

PROSPERO REGISTRATION NUMBER

CRD42021239113.

Performance of French medico-administrative databases in epidemiology of infectious diseases: a scoping review

The development of medico-administrative databases over the last few decades has led to an evolution and to a significant production of epidemiological studies on infectious diseases based on retrospective medical data and consumption of care. This new form of epidemiological research faces numerous methodological challenges, among which the assessment of the validity of targeting algorithm. We conducted a scoping review of studies that undertook an estimation of the completeness and validity of French medico-administrative databases for infectious disease epidemiological research. Nineteen validation studies and nine capture-recapture studies were identified. These studies covered 20 infectious diseases and were mostly based on the evaluation of hospital claimed data. The evaluation of their methodological qualities highlighted the difficulties associated with these types of research, particularly those linked to the assessment of their underlying hypotheses. We recall several recommendations relating to the problems addressed, which should contribute to the quality of future evaluation studies based on medico-administrative data and consequently to the quality of the epidemiological indicators produced from these information systems.

Comanagement with rheumatology and prescription biologics filled during pregnancy in women with rheumatic diseases: a retrospective analysis of US administrative claims data

OBJECTIVES

To evaluate comanagement with rheumatology and biological prescriptions filled during pregnancy among women with axial spondyloarthritis (axSpA), rheumatoid arthritis (RA) or psoriatic arthritis (PsA) and to examine factors associated with receiving comanagement with rheumatology during pregnancy.

DESIGN

A retrospective analysis of US claims data.

SETTING

Commercially insured enrollees using data from the 2013-2018 IBM MarketScan Commercial Claims and Encounters Database.

PARTICIPANTS

We identified 4131 pregnant women aged ≤55 years from the 2013-2018 IBM MarketScan Commercial Claims and Encounters Database with an International Classification of Disease, 9th Revision/10th Revision codes for RA, axSpA or PsA, with continuous enrolment at ≥3 months before the date of the last menstrual period (LMP) (index date) and throughout pregnancy.

PRIMARY OUTCOMES

Filled biologics (prescriptions and infusions) claims were categorised by 90 days before the LMP and trimester, as were primary care, obstetrician and rheumatological claims.

RESULTS

The prevalence of axSpA, RA and PsA was 0.7%, 0.2% and 0.04% among reproductive age women. The average maternal age was 32.7 years (SD 5.7). During pregnancy, 9.1% of those with axSpA (n=2,410) and 56.4% of those with RA/PsA (n=1,721) had a rheumatological claim. Biologics claims were less common among those with axSpA (90 days before LMP: 1.6%, during pregnancy: 1.1%) than those with RA/PsA (90 days before LMP: 11.9%, during pregnancy: 6.9%). Medications during pregnancy included corticosteroids (axSpA: 0.3%, RA/PsA: 2.2%), non-biological disease-modifying antirheumatic drugs (axSpA: 0.2%, RA/PsA: 1.7%), non-steroidal anti-inflammatory drugs (axSpA: 0.2%, RA/PsA: 1.3%) and opioids (axSpA: 0.2%, RA/PsA: 0.6%). Established rheumatological care and biologics claims during the 90 days before LMP showed good prediction accuracy for receiving comanagement with rheumatology during pregnancy (axSpA: area under the receiver operator curve (AUC) 0.73, RA/PsA: AUC 0.70).

CONCLUSION

Comanagement with rheumatology during pregnancy occurs infrequently, especially for women with axSpA. Biologics claims during pregnancy may not align with published guidelines. Future research is warranted to improve comanagement with rheumatology during pregnancy.

Validity of algorithms for identifying mild traumatic brain injury in the French national emergency department database OSCOUR: a retrospective multicentre validation study protocol

INTRODUCTION

The French emergency department (ED) surveillance network OSCOUR transmits data on ED visits to Santé publique France (the national public health agency). As these data are collected daily and are almost exhaustive at a national level, it would seem relevant to use them for national epidemiological surveillance of mild traumatic brain injury (mTBI). This article presents the protocol of a planned study to validate algorithms for identifying mTBI in the OSCOUR database. Algorithms to be tested will be based on International Classification of Diseases (ICD)-10 codes.

METHODS AND ANALYSIS

We will perform a multicentre validation study of algorithms for identifying mTBI in OSCOUR. Different combinations of ICD-10 codes will be used to identify cases of mTBI in the OSCOUR database. A random sample of mTBI cases and non-cases will be selected from four EDs. Medical charts will serve as the reference standard to validate the algorithms. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the different algorithms, as well as their 95% CIs, will be calculated and compared.

ETHICS AND DISSEMINATION

The ethics committee of the French National Data Protection Authority (CNIL) approved this study (n° 921152, 1 August 2021). Results will be submitted to national and international peer-reviewed journals and presented at conferences dedicated to trauma and to methodologies for the construction and validation of algorithms.

Distinct care trajectories among persons living with arthritic conditions: A two-year state sequence analysis

OBJECTIVES

Developing solutions to optimize care trajectories (CareTs) requires examining patient journeys through the health care system. This study aimed to describe CareTs among people living with arthritis and evaluate their association with self-reported health outcomes.

METHODS

Analyses were conducted using the TorSaDE Cohort (n = 102,148), which connects the 2007 to 2016 Canadian Community Health Surveys (CCHS) with Quebec administrative databases (longitudinal claims). CareTs of participants living with arthritis according to CCHS (n = 16,631), over the two years before CCHS completion, were clustered using state sequence analysis (months as a time unit). CareT group membership was then put in association with self-reported outcomes (pain intensity and interference, self-perceived general and mental health).

RESULTS

The analysis revealed five CareT groups characterized predominantly by: (1) arthritis-related visits to a specialist (n = 2,756; 16.6%), (2) arthritis-related emergency department visits (n = 2,928; 17.6%), (3) very high all-cause health care utilization and arthritis-related hospitalizations (n = 1,570; 9.4%), (4) arthritis-related medical visits to general practitioners and specialists (n = 2,708; 16.3%), (5) low all-cause health care utilization (n = 6,669; 40.1%). Multivariable results revealed that CareT group membership was associated with higher levels of pain interference (CareT group #3 vs. #5: OR: 1.4, 95%CI: 1.1-1.8) and fair/poor self-perceived general health (CareT group #1 vs. #5: OR: 1.551, 95%CI: 1.319-1.824; #2 vs. #5: OR: 1.244, 95%CI: 1.062-1.457; #3 vs. #5: OR: 1.771, 95%CI: 1.451-2.162; #4 vs. #5: OR: 1.481, 95%CI: 1.265-1.735).

DISCUSSION

Sate sequence analysis is an innovative method of studying CareTs and valuable for making evidence-based decisions taking into account inter- and intra-individual variability.

Long-term Outcomes Following Multiply Recurrent Clostridioides difficile Infection and Fecal Microbiota Transplantation

BACKGROUND AND AIMS

Fecal microbiota transplantation (FMT) is a commonly used therapy for multiply recurrent Clostridioides difficile (mrCDI). By altering the gut microbiota, there is the potential for FMT to impact the risk for cardiometabolic, intestinal or immune-mediated conditions. Likewise, the microbiota disturbance associated with mrCDI could potentially lead to these conditions. We aimed to assess the associations of mrCDI and FMT with cardiometabolic, immune-mediated diseases, and irritable bowel syndrome.

METHODS

This retrospective cohort study using a United States commercial claims database included persons diagnosed with CDI or undergoing FMT. We created 2 pairwise comparisons: mrCDI vs non-mrCDI, and non-mrCDI or mrCDI treated with FMT vs mrCDI without FMT.

RESULTS

We found no significant association between mrCDI (vs non-mrCDI) and inflammatory bowel disease (adjusted hazard ratio (aHR) = 1.65; 95% confidence interval, 0.67-4.04), rheumatoid arthritis (HR = 0.86; 0.47-1.56), psoriasis (HR = 0.72; 0.23-2.27), diabetes (aHR = 0.97; 0.67-1.40), hypertension (aHR = 1.05; 0.76-1.44), myocardial infarction (aHR = 0.82; 0.63-1.06), stroke (aHR = 0.83; 0.62-1.12), or irritable bowel syndrome (HR = 0.94; 0.61-1.45). Similarly, we found no association of CDI with FMT (vs mrCDI without FMT) and diabetes (aHR = 0.92; 0.27-3.11), hypertension (aHR = 1.41; 0.64-3.15), stroke (aHR = 1.27; 0.69-2.34) or inflammatory bowel syndrome (aHR = 0.80; 0.26-2.46). However, the incidence of myocardial infarction was increased following FMT (aHR = 1.68; 1.01-2.81).

CONCLUSION

Relative to those with CDI, persons with mrCDI do not appear to be intrinsically at higher risk of cardiometabolic, immune-mediated diseases, or irritable bowel syndrome. However, those who underwent FMT for CDI had a higher incidence of myocardial infarction. Future studies should assess this association to assess reproducibility.

Use of diagnostic likelihood ratio of outcome to evaluate misclassification bias in the planning of database studies

Background

The diagnostic likelihood ratio (DLR) and its utility are well-known in the field of medical diagnostic testing. However, its use has been limited in the context of an outcome validation study. We considered that wider recognition of the utility of DLR would enhance the practices surrounding database studies. This is particularly timely and important since the use of healthcare-related databases for pharmacoepidemiology research has greatly expanded in recent years. In this paper, we aimed to advance the use of DLR, focusing on the planning of a new database study.

Methods

Theoretical frameworks were developed for an outcome validation study and a comparative cohort database study; these two were combined to form the overall relationship. Graphical presentations based on these relationships were used to examine the implications of validation study results on the planning of a database study. Additionally, novel uses of graphical presentations were explored using some examples.

Results

Positive DLR was identified as a pivotal parameter that connects the expected positive-predictive value (PPV) with the disease prevalence in the planned database study, where the positive DLR is equal to sensitivity/(1-specificity). Moreover, positive DLR emerged as a pivotal parameter that links the expected risk ratio with the disease risk of the control group in the planned database study. In one example, graphical presentations based on these relationships provided a transparent and informative summary of multiple validation study results. In another example, the potential use of a graphical presentation was demonstrated in selecting a range of positive DLR values that best represented the relevant validation studies.

Conclusions

Inclusion of the DLR in the results section of a validation study would benefit potential users of the study results. Furthermore, investigators planning a database study can utilize the DLR to their benefit. Wider recognition of the full utility of the DLR in the context of a validation study would contribute meaningfully to the promotion of good practice in planning, conducting, analyzing, and interpreting database studies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-01757-1.

Validation of algorithms for selecting rheumatoid arthritis patients in the Tuscan healthcare administrative databases

Validation of algorithms for selecting patients from healthcare administrative databases (HAD) is recommended. This PATHFINDER study section is aimed at testing algorithms to select rheumatoid arthritis (RA) patients from Tuscan HAD (THAD) and assessing RA diagnosis time interval between the medical chart date and that of THAD. A population was extracted from THAD. The information of the medical charts at the Rheumatology Unit of Pisa University Hospital represented the reference. We included first ever users of biologic disease modifying anti-rheumatic drugs (bDMARDs) between 2014 and 2016 (index date) with at least a specialist visit at the Rheumatology Unit of the Pisa University Hospital recorded from 2013 to the index date. Out of these, we tested four index tests (algorithms): (1) RA according to hospital discharge records or emergency department admissions (ICD-9 code, 714*); (2) RA according to exemption code from co-payment (006); (3) RA according to hospital discharge records or emergency department admissions AND RA according to exemption code from co-payment; (4) RA according to hospital discharge records or emergency department admissions OR RA according to exemption code from co-payment. We estimated sensitivity, specificity, positive and negative predicted values (PPV and NPV) with 95% confidence interval (95% CI) and the RA diagnosis median time interval (interquartile range, IQR). Two sensitivity analyses were performed. Among 277 reference patients, 103 had RA. The fourth algorithm identified 96 true RA patients, PPV 0.78 (95% CI 0.70-0.85), sensitivity 0.93 (95% CI 0.86-0.97), specificity 0.84 (95% CI 0.78-0.90), and NPV 0.95 (95% CI 0.91-0.98). The sensitivity analyses confirmed performance. The time measured between the actual RA diagnosis date recorded in medical charts and that assumed in THAD was 2.2 years (IQR 0.5-8.4). In conclusion, this validation showed the fourth algorithm as the best. The time interval elapsed between the actual RA diagnosis date in medical charts and that extrapolated from THAD has to be considered in the design of future studies.

A strategy for validation of variables derived from large-scale electronic health record data

Purpose:

Standardized approaches for rigorous validation of phenotyping from large-scale electronic health record (EHR) data have not been widely reported. We proposed a methodologically rigorous and efficient approach to guide such validation, including strategies for sampling cases and controls, determining sample sizes, estimating algorithm performance, and terminating the validation process, hereafter referred to as the San Diego Approach to Variable Validation (SDAVV).

Methods:

We propose sample size formulae which should be used prior to chart review, based on pre-specified critical lower bounds for positive predictive value (PPV) and negative predictive value (NPV). We also propose a stepwise strategy for iterative algorithm development/validation cycles, updating sample sizes for data abstraction until both PPV and NPV achieve target performance.

Results:

We applied the SDAVV to a Department of Veterans Affairs study in which we created two phenotyping algorithms, one for distinguishing normal colonoscopy cases from abnormal colonoscopy controls and one for identifying aspirin exposure. Estimated PPV and NPV both reached 0.970 with a 95% confidence lower bound of 0.915, estimated sensitivity was 0.963 and specificity was 0.975 for identifying normal colonoscopy cases. The phenotyping algorithm for identifying aspirin exposure reached a PPV of 0.990 (a 95% lower bound of 0.950), an NPV of 0.980 (a 95% lower bound of 0.930), and sensitivity and specificity were 0.960 and 1.000.

Conclusions:

A structured approach for prospectively developing and validating phenotyping algorithms from large-scale EHR data can be successfully implemented, and should be considered to improve the quality of “big data” research.

The validity of the claims-based definition of rheumatoid arthritis evaluated in 64 hospitals in Japan

BACKGROUND

An administrative database covering a whole population such as the national database in Japan may be used to estimate the nationwide prevalence of diseases including rheumatoid arthritis (RA) when a well-validated definition of the disease is available. In Japan, the record linkage between the administrative database and medical charts in hospitals is strictly prohibited. A "hospital-based" validation study is one of few possible validation studies where claims kept inside the study hospital are rearranged into the database structure.

METHODS

We selected random samples of 19,734 patients from approximately 1.6 million patients who received medical care between February 2018 and January 2019 in one of the 64 hospitals of the Tokushukai Medical Group. We excluded patients whose observation period was less than 365 days and identified 334 patients who met the definition of "possible cases of RA" whose medical charts were then independently evaluated by two rheumatologists. In a sensitivity analysis, we assessed bias due to misclassifying some patients with RA who did not meet the definition of "possible cases of RA" as a patient with no RA.

RESULTS

The kappa coefficient between the two rheumatologists was 0.80. The prevalence of RA in the study population was estimated to be 0.56%. We found that [condition code of RA] and ([any disease-modifying antirheumatic drug] or [oral corticosteroid with no systemic autoimmune diseases (other than RA) and no polymyalgia rheumatica]) had a relatively high sensitivity (approximately 73%) and a high positive predictive value (approximately 80%). In a sensitivity analysis, we found that when some patients with RA who did not meet the definition of "possible cases of RA" were misclassified as a patient with no RA, then this would lead to underestimation of the prevalence of the definition-positive patients and the adjusted prevalence.

CONCLUSIONS

We recommend using the claims-based definition of RA (found in the current validation study) to estimate the prevalence of RA in Japan. We also suggest estimating the adjusted prevalence using the quantitative bias analysis method, since the prevalence of the disease in the "hospital-based" validation study is different from that in the administrative database.

TRIAL REGISTRATION

The current study is not a clinical trial and hence not subject to trial registration.

Accuracy of an algorithm to identify rheumatoid arthritis in the Longitudinal Ageing Study Amsterdam population: a validation study

Objective: In view of global ageing and the scarcity of knowledge about disease determinants in older individuals with rheumatoid arthritis (RA), an algorithm with optimal diagnostic accuracy was developed to identify RA patients in the Longitudinal Ageing Study Amsterdam (LASA).Method: Four case ascertainment algorithms were constructed and assessed for validity in LASA, an ongoing cohort study (≥ 55 years) representing the general older population of the Netherlands. Data sources used to identify the diagnosis RA were: self-reported morbidity, specialist diagnosis, and medication. A validation subsample of LASA participants was taken to verify RA diagnosis by a standard procedure using a checklist.Results: Data from 272/300 (91%) participants were verified. Four algorithms were developed: 'treatment', 'diagnosis', 'treatment or diagnosis', and 'treatment and diagnosis'. The algorithm 'treatment and diagnosis' showed the best measurement properties: specificity 100%, positive predictive value 100%, and area under the receiver operating characteristics curve 0.72. Applying this algorithm in the LASA sample (mean age 71 years) revealed a prevalence of RA of 1.0% (19/1908 participants).Conclusion: An algorithm for RA identification in the LASA population was developed, with high diagnostic accuracy. It provides an accurate tool to identify older adults with RA in LASA and, after validation, may be applicable in other large population-based studies.

Development and validation of lupus nephritis case definitions using United States veterans affairs electronic health records

OBJECTIVE

International Classification of Diseases (ICD) codes are commonly used to identify patients with rare diseases in electronic health records (EHRs). However, misclassification is common, impacting the validity of study results. In this study, we compared the accuracies of several ICD-based case definitions of lupus nephritis (LN) in identifying United States veterans with LN.

METHODS

Using the Department of Veterans Affairs (VA) EHR, we identified all veterans with ≥1 ICD-9 or 10 diagnostic codes for systemic lupus erythematosus (SLE) between October 1, 1999 and September 30, 2017. A cohort was randomly selected for diagnostic validation and 9 ICD-based LN case definitions were applied to this cohort. The diagnostic accuracy of each definition was assessed against gold standard criterion of biopsy-proven LN.

RESULTS

18,420 veterans had ≥1 ICD-9 or 10 diagnostic codes for SLE; 981 were randomly selected for diagnostic validation. 95 veterans (9.7%) had biopsy-proven LN. The case definitions had high specificity and NPV but variable sensitivity and PPV. The definition containing ≥2 ICD -9 codes for SLE and ≥2 nephritis indicators had the highest combination of sensitivity and specificity (87.4% and 94.6% respectively). ICD-10 code for LN had high specificity (99.8%) and PPV (93.9%).

CONCLUSION

ICD-based case definitions of LN in the VA population have high specificity and NPV but variable sensitivity and PPV. Our results may help guide the design of future LN studies in VA cohorts. The choice of specific case definitions depends on the relative importance of different accuracy measures to individual studies.

The association between waiting time and multidisciplinary pain treatment outcomes in patients with rheumatic conditions

BACKGROUND

Access to multidisciplinary pain treatment facilities (MPTF) is limited by extensive waiting time in many countries. However, there is a lack of knowledge about the impact of waiting time on clinical outcomes, particularly for patients with rheumatic conditions. This study examined the association between waiting time for MPTF and clinical outcomes in patients with rheumatic conditions.

METHODS

Data were extracted from the Quebec Pain Registry, a large database of patients who received services in MPTF. The associations between waiting time (classified as < 2 months, 2-6 months and >  6 months) and change in pain interference, pain intensity and health-related quality of life, from the initial visit at the MPTF to the 6-month follow-up, were tested using generalized estimating equations.

RESULTS

A total of 3230 patients with rheumatic conditions (mean age: 55.8 ± 14.0 years; 66% were women) were included in the analysis. Small significant differences in improvement between waiting time groups were revealed, with patients waiting less than 2 months having a larger improvement in all clinical outcomes compared to patients who waited 2-6 months or over 6 months before their initial visit (adjusted time X group effect p ≤ 0.001). Only patients waiting less than 2 months reached a clinically important improvement in pain interference (1.12/10), pain intensity (1.3/10) and physical and mental quality of life (3.9 and 3.7/100).

CONCLUSIONS

Longer delays experienced by patients before receiving services in MPTF were associated with statistically significant smaller improvements in pain interference, pain intensity and health-related quality of life; these differences were, however, not clinically significant. Based on these results, we advise that strategies are developed not only to reduce waiting times and mitigate their impacts on patients with rheumatic conditions, but also to improve treatment effectiveness in MPTF.

The validity of diagnostic algorithms to identify asthma patients in healthcare administrative databases: a systematic literature review

OBJECTIVES

To review the available evidence supporting the validity of algorithms to identify asthma patients in healthcare administrative databases.

METHODS

A systematic literature search was conducted on multiple databases from inception to March 2020 to identify studies that reported the validity of case-finding asthma algorithms applied to healthcare administrative data. Following an initial screening of abstracts, two investigators independently assessed the full text of studies which met the pre-determined eligibility criteria. Data on study population and algorithm characteristics were extracted. A revised version of the Quality Assessment of Diagnostic Accuracy Studies tool was used to evaluate the risk of bias and generalizability of studies.

RESULTS

A total of 20 studies met the eligibility criteria. Algorithms which incorporated ≥1 diagnostic code for asthma over a 1-year period appeared to be valid in both adult and pediatric populations (sensitivity ≥ 85%; specificity ≥ 89%; PPV ≥ 70%). The validity was enhanced when: (1) the time frame to capture asthma cases was increased to two years; (2) ≥2 asthma diagnostic codes were considered; and (3) when diagnoses were recorded by a pulmonologist. Algorithms which integrated pharmacy claims data appeared to correctly identify asthma patients; however, the extent to which asthma medications can improve the validity remains unclear. The quality of several studies was high, although disease progression bias and biases related to self-reported data was observed in some studies.

CONCLUSIONS

Healthcare administrative databases are adequate sources to identify asthma patients. More restrictive definitions based on both asthma diagnoses and asthma medications may enhance validity, although further research is required to confirm this hypothesis.

Validity of Algorithms for Identification of Individuals Suffering from Chronic Noncancer Pain in Administrative Databases: A Systematic Review

BACKGROUND

Secondary analysis of health administrative databases is indispensable to enriching our understanding of health trajectories, health care utilization, and real-world risks and benefits of drugs among large populations.

OBJECTIVES

This systematic review aimed at assessing evidence about the validity of algorithms for the identification of individuals suffering from nonarthritic chronic noncancer pain (CNCP) in administrative databases.

METHODS

Studies reporting measures of diagnostic accuracy of such algorithms and published in English or French were searched in the Medline, Embase, CINAHL, AgeLine, PsycINFO, and Abstracts in Social Gerontology electronic databases without any dates of coverage restrictions up to March 1, 2018. Reference lists of included studies were also screened for additional publications.

RESULTS

Only six studies focused on commonly studied CNCP conditions and were included in the review. Some algorithms showed a ≥60% combination of sensitivity and specificity values (back pain disorders in general, fibromyalgia, low back pain, migraine, neck/back problems studied together). Only algorithms designed to identify fibromyalgia cases reached a ≥80% combination (without replication of findings in other studies/databases).

CONCLUSIONS

In summary, the present investigation informs us about the limited amount of literature available to guide and support the use of administrative databases as valid sources of data for research on CNCP. Considering the added value of such data sources, the important research gaps identified in this innovative review provide important directions for future research. The review protocol was registered with PROSPERO (CRD42018086402).

Administrative database as a source for assessment of systemic lupus erythematosus prevalence: Estonian experience

Background

Administrative database research is widely applied in the field of epidemiology. However, the results of the studies depend on the type of database used and the algorithms applied for case ascertainment. The optimal methodology for identifying patients with rheumatic diseases from administrative databases is yet not known. Our aim was to describe an administrative database as a source for estimation of epidemiological characteristics on an example of systemic lupus erythematosus (SLE, ICD-10 code M32) prevalence assessment in the database of the Estonian Health Insurance Fund (EHIF).

Methods

Code M32 billing episodes were extracted from the EHIF database 2006–2010. For all cases where M32 was assigned by a rheumatologist less than four times during the study period, diagnosis verification process using health care providers’ (HCP) databases was applied. For M32 cases assigned by a rheumatologist four times or more, diagnoses were verified for a randomly selected sample.

Results

From 677 persons with code M32 assigned in EHIF database, 404 were demonstrated having “true SLE”. The code M32 positive predictive value (PPV) for the whole EHIF database was 60%; PPV varies remarkably by specialty of a physician and repetition of the code assignment. The false positive M32 codes were predominantly initial diagnoses which were not confirmed afterwards; in many cases, a rheumatic condition other than SLE was later diagnosed.

Conclusions

False positive codes due to tentative diagnoses may be characteristic for conditions with a complicated diagnosis process like SLE and need to be taken into account when performing administrative database research.

PheValuator: Development and evaluation of a phenotype algorithm evaluator

BACKGROUND

The primary approach for defining disease in observational healthcare databases is to construct phenotype algorithms (PAs), rule-based heuristics predicated on the presence, absence, and temporal logic of clinical observations. However, a complete evaluation of PAs, i.e., determining sensitivity, specificity, and positive predictive value (PPV), is rarely performed. In this study, we propose a tool (PheValuator) to efficiently estimate a complete PA evaluation.

METHODS

We used 4 administrative claims datasets: OptumInsight's de-identified Clinformatics™ Datamart (Eden Prairie,MN); IBM MarketScan Multi-State Medicaid); IBM MarketScan Medicare Supplemental Beneficiaries; and IBM MarketScan Commercial Claims and Encounters from 2000 to 2017. Using PheValuator involves (1) creating a diagnostic predictive model for the phenotype, (2) applying the model to a large set of randomly selected subjects, and (3) comparing each subject's predicted probability for the phenotype to inclusion/exclusion in PAs. We used the predictions as a 'probabilistic gold standard' measure to classify positive/negative cases. We examined 4 phenotypes: myocardial infarction, cerebral infarction, chronic kidney disease, and atrial fibrillation. We examined several PAs for each phenotype including 1-time (1X) occurrence of the diagnosis code in the subject's record and 1-time occurrence of the diagnosis in an inpatient setting with the diagnosis code as the primary reason for admission (1X-IP-1stPos).

RESULTS

Across phenotypes, the 1X PA showed the highest sensitivity/lowest PPV among all PAs. 1X-IP-1stPos yielded the highest PPV/lowest sensitivity. Specificity was very high across algorithms. We found similar results between algorithms across datasets.

CONCLUSION

PheValuator appears to show promise as a tool to estimate PA performance characteristics.

A systematic review of database validation studies among fertility populations

STUDY QUESTION

Are routinely collected data from fertility populations adequately validated?

SUMMARY ANSWER

Of the 19 studies included, only one validated a national fertility registry and none reported their results in accordance with recommended reporting guidelines for validation studies.

WHAT IS KNOWN ALREADY

Routinely collected data, including administrative databases and registries, are excellent sources of data, particularly for reporting, quality assurance, and research. However, these data are subject to misclassification bias due to misdiagnosis or errors in data entry and therefore need to be validated prior to using for clinical or research purposes.

STUDY DESIGN SIZE DURATION

We conducted a systematic review by searching Medline, Embase, and CINAHL from inception to 6 October 2016 to identify validation studies of databases that contain routinely collected data in an ART setting. Webpages of international ART centers were also searched.

PARTICIPANTS/MATERIALS SETTING METHODS

We included studies that compared at least two data sources to validate ART population data. Key words and MeSH terms were adapted from previous systematic reviews investigating routinely collected data (e.g. administrative databases and registries), measures of validity (including sensitivity, specificity, and predictive value), and ART (including infertility, IVF, advanced reproductive age, and diminished ovarian reserve). Only full-text studies in English were considered. Results were synthesized qualitatively. The electronic search yielded 1074 citations, of which 19 met the inclusion criteria.

MAIN RESULTS AND THE ROLE OF CHANCE

Two studies validated a fertility database using medical records; seven studies used an IVF registry to validate vital records or maternal questionnaires, and two studies failed to adequately describe their reference standard. Four studies investigated the validity of mode of conception from birth registries; two studies validated diagnoses or treatments in a fertility database; four studies validated a linkage algorithm between a fertility registry and another administrative database; one study created an algorithm in a single database to identify a patient population. Sensitivity was the most commonly reported measure of validity (12 studies), followed by specificity (9 studies). Only three studies reported four or more measures of validation, and five studies presented CIs for their estimates. The prevalence of the variable in the target population (pre-test prevalence) was reported in seven studies; however, only four of the studies had prevalence estimates from the study population (post-test prevalence) within a 2% range of the pre-test estimate. The post-test estimate was largely discrepant from the pre-test value in two studies.

LIMITATIONS REASONS FOR CAUTION

The search strategy was limited to the studies and reports published in English, which may not capture validation studies from countries that do not speak English. Furthermore, only three specific fertility-based diagnostic variables (advanced reproductive age, diminished ovarian reserve, and chorionicity) were searched in Medline, Embase, and CINAHL. Consequently, published studies with other diagnoses or conditions relevant to infertility may not have been captured in our review.

WIDER IMPLICATIONS OF THE FINDINGS

There is a paucity of literature on validation of routinely collected data from a fertility population. Furthermore, the prevalence of the markers that have been validated are not being presented, which can lead to biased estimates. Stakeholders rely on these data for monitoring outcomes of treatments and adverse events; therefore, it is essential to ascertain the accuracy of these databases and make the reports publicly available.

STUDY FUNDING/COMPETING INTERESTS

This study was supported by Canadian Institutes of Health Research (CIHR) (FDN-148438). There are no competing interests for any of the authors.

REGISTRATION NUMBER

International Prospective Register of Systematic Reviews ID: CRD42016048466.

A Population-level Analysis of the Differing Effects of Rheumatoid Arthritis and Spondyloarthritis on Peripartum Outcomes

OBJECTIVE

The effects of rheumatoid arthritis (RA) and spondyloarthritis (SpA) on maternal and neonatal outcomes at a population level have not previously been well compared.

METHODS

A contemporary pregnancy cohort of 312,081 women and corresponding birth events was assembled for the province of Alberta from the random selection of 1 live birth event per woman. We identified 3 groups: (1) no inflammatory arthritis (no IA, n = 308,989), (2) RA (n = 631), and (3) SpA (n = 2461). We compared maternal and neonatal outcomes, comorbid conditions, and medication use among the 3 groups. Multivariable logistic regression models evaluated the independent association between RA and SpA, relative to no IA, and the outcomes of small for gestation age (SGA) and hypertensive disorders during pregnancy.

RESULTS

Pregnant women with RA were significantly more likely to have preterm delivery (13.5%), cesarean delivery (33.9%), hypertensive disorders in pregnancy (10.5%), and SGA babies (15.6%), compared to pregnant women with either SpA or no IA. Nonsteroidal antiinflammatory drugs and corticosteroid use were significantly higher in pregnant women with RA compared to the other groups. Women with RA were significantly more likely to have an SGA baby (OR 1.51, 95% CI 1.21-1.88; p < 0.01), and hypertensive disorder in pregnancy (OR 1.51, 95% CI 1.16-1.97; p < 0.01), compared to women with no IA, while no difference was found between women with SpA and those with no IA.

CONCLUSION

Women with RA have a higher risk of worse maternal and neonatal outcomes, whereas the risk of these events is similar between women with and without SpA.

Identifying Rheumatoid Arthritis Cases within the Quebec Health Administrative Database

Objective.

Our objective was to calculate rheumatoid arthritis (RA) point prevalence estimates in the CARTaGENE cohort, as well as to estimate the sensitivity and specificity of our ascertainment approach, using physician billing data. We investigated the effects of using varying observation windows in the Régie de l’assurance maladie du Québec (RAMQ) health services administrative databases, alone or in combination with self-reported diagnoses and drugs.

Methods.

We studied subjects enrolled in the CARTaGENE cohort, which recruited 19,995 participants from 4 metropolitan regions in Québec from August 2009 to October 2010. A series of Bayesian latent class models were developed to assess the effects of 3 factors: the number of years of billing data, the addition of self-reported information on RA diagnoses and drugs, and the adjustment for misclassification error.

Results.

The 3-year 2010 point prevalence estimate among cohort members aged 40–69 years, using physician billing plus self-report, adjusting for misclassification error in each source, was 0.9% [95% credible interval (CrI) 0.7–1.2] with RAMQ sensitivity of 84.0% (95% CrI 74.0–93.7) and a specificity of 99.8% (95% CrI 99.6–100.0). Our results show variations in the prevalence point estimates related to all 3 factors investigated.

Conclusion.

Our study illustrates that multiple data sources identify more RA cases and thus a higher prevalence estimate. RA point prevalence estimates using billing data are lower if fewer years of data are used.

Validation of algorithms to identify knee osteoarthritis patients in the claims database

BACKGROUND

To identify knee osteoarthritis (OA) patients among OA patients in the claims database.

METHODS

All patients with OA diagnostic codes for any sites (M15 to M19) in 2014 were recruited from a single academic referral hospital. After excluding patients who had inflammatory arthritis or were less than 50 years of age, we identified data for the overall OA population. Radiographic knee OA of Kellgren and Lawrence grades ≥2 is considered the gold standard for knee OA, and we evaluated the sensitivity, specificity, and positive predictive value (PPV) of three operational definitions using the diagnostic codes in the claims database. The operational definitions were: (1) gonarthrosis (M17); (2) any site of OA (M15 to M19) with knee X-ray; and (3) (1) or (2).

RESULTS

A total of 7959 OA patients were included in this study of whom 74.5% were women. The PPV of gonarthrosis (M17) was 0.67 (95% CI 0.65-0.69), and sensitivity was 0.44 (95% CI 0.42-0.46). The PPV and sensitivity of any OA site (M15 to M19) with knee X-ray were 0.65 (95% CI 0.62-0.67), and 0.37 (95% CI 0.35-0.39), respectively. When knee OA was defined as satisfying either of the two above definitions, PPV was 0.63 (95% CI 0.62-0.65) and sensitivity 0.55 (95% CI 0.53-0.57).

CONCLUSIONS

Knee OA patients can be identified in a claims database using the algorithms of gonarthrosis (M17) or any site of OA (M15 to M19) with a performed knee X-ray.

Validation of Algorithm to Identify Persons with Non-traumatic Spinal Cord Dysfunction in Canada Using Administrative Health Data

Background: Administrative health data, such as the hospital Discharge Abstract Database (DAD), can potentially be used to identify patients with non-traumatic spinal cord dysfunction (NTSCD). Algorithms utilizing administrative health data for this purpose should be validated before clinical use. Objective: To validate an algorithm designed to identify patients with NTSCD through DAD. Method: DAD between 2006 and 2016 for Southern Alberta in Canada were obtained through Alberta Health Services. Cases of NTSCD were identified using the algorithm designed by the research team. These were then validated by chart review using electronic medical records where possible and paper records where electronic records were unavailable. Measures of diagnostic accuracy including sensitivity, specificity, and positive and negative predictive values and 95% confidence intervals (CI) were computed. Results: Two hundred and eighty cases were identified to have both the administrative codes for neurological impairments and NTSCD etiology. Twenty-eight cases were excluded from analysis as 5 had inadequate medical record information, 17 had traumatic spinal cord injury, and 6 were considered "other" non-spinal cord conditions. Measures of diagnostic accuracy that were computed were sensitivity 97% (95% CI, 94%-98%), specificity 60% (95% CI, 47%-73%), positive predictive value (PPV) 92% (95% CI, 88%-95%), and negative predictive value (NPV) 80% (95% CI, 65%-90%). The most prevalent etiologies were degenerative (36.9%), infection (19.0%), oncology malignant (15.1%), and vascular (10.3%). Conclusion: Our algorithm has high sensitivity and PPV and satisfactory specificity and NPV for the identification of persons with NTSCD using DAD, though the limitations for using this method should be recognized.

Identifying Psoriasis and Psoriatic Arthritis Patients in Retrospective Databases When Diagnosis Codes Are Not Available: A Validation Study Comparing Medication/Prescriber Visit-Based Algorithms with Diagnosis Codes

BACKGROUND

Using diagnosis code-based algorithms is the primary method of identifying patient cohorts for retrospective studies; nevertheless, many databases lack reliable diagnosis code information.

OBJECTIVES

To develop precise algorithms based on medication claims/prescriber visits (MCs/PVs) to identify psoriasis (PsO) patients and psoriatic patients with arthritic conditions (PsO-AC), a proxy for psoriatic arthritis, in Canadian databases lacking diagnosis codes.

METHODS

Algorithms were developed using medications with narrow indication profiles in combination with prescriber specialty to define PsO and PsO-AC. For a 3-year study period from July 1, 2009, algorithms were validated using the PharMetrics Plus database, which contains both adjudicated medication claims and diagnosis codes. Positive predictive value (PPV), negative predictive value (NPV), sensitivity, and specificity of the developed algorithms were assessed using diagnosis code as the reference standard. Chosen algorithms were then applied to Canadian drug databases to profile the algorithm-identified PsO and PsO-AC cohorts.

RESULTS

In the selected database, 183,328 patients were identified for validation. The highest PPVs for PsO (85%) and PsO-AC (65%) occurred when a predictive algorithm of two or more MCs/PVs was compared with the reference standard of one or more diagnosis codes. NPV and specificity were high (99%-100%), whereas sensitivity was low (≤30%). Reducing the number of MCs/PVs or increasing diagnosis claims decreased the algorithms' PPVs.

CONCLUSIONS

We have developed an MC/PV-based algorithm to identify PsO patients with a high degree of accuracy, but accuracy for PsO-AC requires further investigation. Such methods allow researchers to conduct retrospective studies in databases in which diagnosis codes are absent.

Population trends in the incidence and initial management of osteoarthritis: age-period-cohort analysis of the Clinical Practice Research Datalink, 1992-2013

Objective

To determine recent trends in the rate and management of new cases of OA presenting to primary healthcare using UK nationally representative data.

Methods

Using the Clinical Practice Research Datalink we identified new cases of diagnosed OA and clinical OA (including OA-relevant peripheral joint pain in those aged over 45 years) using established code lists. For both definitions we estimated annual incidence density using exact person-time, and undertook descriptive analysis and age-period-cohort modelling. Demographic characteristics and management were described for incident cases in each calendar year. Sensitivity analyses explored the robustness of the findings to key assumptions.

Results

Between 1992 and 2013 the annual age-sex standardized incidence rate for clinical OA increased from 29.2 to 40.5/1000 person-years. After controlling for period effects, the consultation incidence of clinical OA was higher for successive cohorts born after the mid-1950s, particularly women. In contrast, with the exception of hand OA, we observed no increase in the incidence of diagnosed OA: 8.6/1000 person-years in 2004 down to 6.3 in 2013. In 2013, 16.4% of clinical OA cases had an X-ray referral. While NSAID prescriptions fell from 2004, the proportion prescribed opioid analgesia rose markedly (0.1% of diagnosed OA in 1992 to 1.9% in 2013).

Conclusion

Rising rates of clinical OA, continued use of plain radiography and a shift towards opioid analgesic prescription are concerning. Our findings support the search for policies to tackle this common problem that promote joint pain prevention while avoiding excessive and inappropriate health care.

A Longitudinal Study of Ambulatory Physician Encounters, Emergency Room Visits, and Hospitalizations by Patients with Rheumatoid Arthritis: A 13-year Population Health Study

Objective.

To determine total physician encounters, emergency room (ER) visits, and hospitalizations in an incident cohort of rheumatoid arthritis (RA) cases and matched control patients over 13 years.

Methods.

A retrospective cohort study was performed using administrative healthcare data from about 1 million people with access to universal healthcare. Using the International Classification of Diseases, 9th ed (ICD-9) and ICD-10 diagnostic codes, 7 RA case definitions were used. Each case was matched by age and sex to 4 randomly selected controls. Data included physician billings, ER visits, and hospital discharges over 13 years.

Results.

The number of incident RA cases varied from 3497 to 27,694, depending on the case definition. The mean age varied from 54.3 to 65.0 years, and the proportion of women from 67.8% to 71.3%. The number of physician encounters by patients with RA was significantly higher than by controls. It was highest in the index year and declined promptly thereafter for all case definitions and by 12.2%–46.8% after 10 years. Encounters with subspecialty physicians fell by 61% (rheumatologists) and 34% (internal medicine). In contrast, clinical encounters with family physicians and other physicians fell by only 9%. Visits to the ER and hospital admissions were also significantly higher in RA cases, particularly early in the disease, and fell significantly over the followup.

Conclusion.

In patients with RA, healthcare use is highest in the first year following the diagnosis, which is also the time of maximal involvement by rheumatologists. Use declines over time, and encounters with patients’ family physicians predominate over other physician groups.

[Prevalence of rheumatoid arthritis in Germany based on health insurance data : Regional differences and first results of the PROCLAIR study]

BACKGROUND

Rheumatoid arthritis (RA) is the most common chronic inflammatory joint disease with a prevalence of up to 1 % in the adult population.

OBJECTIVE

This study describes the prevalence of RA diagnoses in outpatient health insurance claims data, based on different case definitions and stratified by age, sex and region of residence.

METHODS

Based on data from a nationwide statutory health insurance fund (BARMER GEK) from the year 2013, a cross-sectional study of insurants aged 18 years or older was conducted. The following case definitions were applied: A) a diagnosis of seropositive rheumatoid arthritis (M05) or other rheumatoid arthritis (M06) according to the international classification of diseases 10 German modification (ICD-10-GM) in at least two quarterly periods of the year 2013, B) case definition A plus determination of C‑reactive protein (CRP) or erythrocyte sedimentation rate (ESR) at least once, C) case definition B plus specific drug therapy and D) case definition A plus treatment by a rheumatologist. Raw as well as age and sex-standardized prevalences were calculated and stratified according to the federal state.

RESULTS

The study population consisted of 7,155,315 insurants of whom 60.2 % were female. Overall, RA prevalences for the respective case definitions were 1.62 % (A), 1.11 % (B), 0.94 % (C) and 0.64 % (D). When standardized to the German population the prevalences were 1.38 % (A), 0.95 % (B), 0.81 % (C) and 0.55 % (D). The proportion of women was approximately 80 % for all case definitions. Prevalences increased with age, peaking in the age group 70-79 years old and showing the highest values in eastern and the lowest in southern Germany for raw as well as standardized measures.

CONCLUSION

Regional differences in the prevalence of RA diagnoses in health insurance claims data were observed independent of age, sex and case definition. The expected prevalence according to the results of international studies was best achieved when case definitions with CRP or ESR were considered.

Validation of administrative case ascertainment algorithms for chronic childhood arthritis in Manitoba, Canada

We validated case ascertainment algorithms for juvenile idiopathic arthritis (JIA) in the provincial health administrative databases of Manitoba, Canada. A population-based pediatric rheumatology clinical database from April 1st 1980 to March 31st 2012 was used to test case definitions in individuals diagnosed at ≤15 years of age. The case definitions varied the number of diagnosis codes (1, 2, or 3), time frame (1, 2 or 3 years), time between diagnoses (ever, >1 day, or ≥8 weeks), and physician specialty. Positive predictive value (PPV), sensitivity, and specificity with 95% confidence intervals (CIs) are reported. A case definition of 1 hospitalization or ≥2 diagnoses in 2 years by any provider ≥8 weeks apart using diagnosis codes for rheumatoid arthritis and ankylosing spondylitis produced a sensitivity of 89.2% (95% CI 86.8, 91.6), specificity of 86.3% (95% CI 83.0, 89.6), and PPV of 90.6% (95% CI 88.3, 92.9) when seronegative enthesopathy and arthropathy (SEA) was excluded as JIA; and sensitivity of 88.2% (95% CI 85.7, 90.7), specificity of 90.4% (95% CI 87.5, 93.3), and PPV of 93.9% (95% CI 92.0, 95.8) when SEA was included as JIA. This study validates case ascertainment algorithms for JIA in Canadian administrative health data using diagnosis codes for both rheumatoid arthritis (RA) and ankylosing spondylitis, to better reflect current JIA classification than codes for RA alone. Researchers will be able to use these results to define cohorts for population-based studies.

Prevalence of rheumatoid arthritis in the United States adult population in healthcare claims databases, 2004-2014

This study aimed to determine the prevalence of rheumatoid arthritis in the United States (US) adult insured population from 2004 to 2014. This was an observational, retrospective, cross-sectional study based on US administrative health insurance claims databases (Truven Health MarketScan^® Research database and IMS PharMetrics Plus database). Trends in RA prevalence focusing on the 10-year period covering January 1, 2004-December 31, 2014 were analyzed using a validated algorithm for the identification of RA. Prevalence rates in the databases were determined and age- and gender-adjusted rates were projected to the US population in 2014. Analysis of data from the two databases indicated that the RA prevalence rate in commercially insured adult US population ranged from 0.41 to 0.54% from 2004 to 2014. The prevalence varied substantially by gender and age in each year and increased gradually across the years for most subgroups. In 2014, out of 31,316,902 adult patients with continuous enrollment in the Truven Health MarketScan^® Research database, 157,634 (0.50%) patients met our criteria for RA. Similarly, out of 35,083,356 adult patients in the IMS PharMetrics Plus database, 139,300 (0.50%) patients met our criteria for RA. In 2014, the overall age-adjusted prevalence of RA ranged from 0.53 to 0.55% (0.29-0.31% for males and 0.73-0.78% for females). The prevalence of RA in the US appeared to increase during the period from 2004 to 2014, affecting a conservative estimate of 1.28-1.36 million adults in 2014.

Health-selective migration among patients with rheumatoid arthritis in Québec: a cohort study using administrative data

Little is known about how rheumatoid arthritis (RA) affects an individual's ability to relocate. The current literature suggests the relationship between health and migration is often disease-specific. We sought to estimate the impact of RA diagnosis on migration within a Canadian province, comparing migration rates in residents before and after RA diagnosis. We identified a cohort of 81,181 individuals diagnosed with RA between 1998 and 2009 using Québec administrative databases. A migration was defined as a change in the first three characters of the postal code. We categorized migrations as urban or rural depending upon an individual's origin and destination. We estimated the association between RA diagnosis and migration by fitting marginal models using a generalized estimating equations approach, adjusting for age, sex, and population level socioeconomic status indicators. The vast majority of moves after RA diagnosis were within urban areas. RA diagnosis was associated with increased migration except for people around age 50 moving within urban areas. Although RA was associated with increased inter-urban migration in many demographic groups, the net result did not translate to higher rates of rural-to-urban migration after RA diagnosis. Our results suggest fairly complex associations between RA diagnosis and migration. Both age and location (urban or rural) modify this effect. Overall, we did not see a greater movement from rural-to-urban areas after RA diagnosis. This is of interest for studies of regional environmental effects on chronic disease patterns.

Diagnostic accuracy of administrative data algorithms in the diagnosis of osteoarthritis: a systematic review

Background

Administrative health care data are frequently used to study disease burden and treatment outcomes in many conditions including osteoarthritis (OA). OA is a chronic condition with significant disease burden affecting over 27 million adults in the US. There are few studies examining the performance of administrative data algorithms to diagnose OA. The purpose of this study is to perform a systematic review of administrative data algorithms for OA diagnosis; and, to evaluate the diagnostic characteristics of algorithms based on restrictiveness and reference standards.

Methods

Two reviewers independently screened English-language articles published in Medline, Embase, PubMed, and Cochrane databases that used administrative data to identify OA cases. Each algorithm was classified as restrictive or less restrictive based on number and type of administrative codes required to satisfy the case definition. We recorded sensitivity and specificity of algorithms and calculated positive likelihood ratio (LR+) and positive predictive value (PPV) based on assumed OA prevalence of 0.1, 0.25, and 0.50.

Results

The search identified 7 studies that used 13 algorithms. Of these 13 algorithms, 5 were classified as restrictive and 8 as less restrictive. Restrictive algorithms had lower median sensitivity and higher median specificity compared to less restrictive algorithms when reference standards were self-report and American college of Rheumatology (ACR) criteria. The algorithms compared to reference standard of physician diagnosis had higher sensitivity and specificity than those compared to self-reported diagnosis or ACR criteria.

Conclusions

Restrictive algorithms are more specific for OA diagnosis and can be used to identify cases when false positives have higher costs e.g. interventional studies. Less restrictive algorithms are more sensitive and suited for studies that attempt to identify all cases e.g. screening programs.

Electronic supplementary material

The online version of this article (doi:10.1186/s12911-016-0319-y) contains supplementary material, which is available to authorized users.

Influenza and pneumococcal vaccination in patients with rheumatoid arthritis in comparison with age- and sex-matched controls: results of a claims data analysis

The aim of this study was to assess the vaccination status for influenza and pneumonia and the prevalence of hospitalised pneumonia in rheumatoid arthritis (RA) patients and population controls in Germany. Members of a large statutory health insurance fund in Germany who were continuously insured between 2009 and 2013 and had a diagnosis of RA in 2013 were age and sex matched 1:5 to members without RA. Pneumococcal and influenza vaccinations were evaluated with regard to age, sex and region of residence. Logistic regression models were used to determine predictors for influenza vaccination in RA patients. Prevalences of pneumonia that required hospitalisation were compared to regional vaccination rates. The data of 111,482 RA patients and 557,410 matched controls were available for analysis. Compared to controls, RA patients were vaccinated more frequently against influenza (40.8 vs. 32.2 %) and pneumonia (15.0 vs. 10.0 %). Vaccination rates increased with older age and differed between the federal states (highest in East Germany, lowest in South Germany). The region of residence, comorbidities, rheumatologic care and biologic treatment was associated with a higher probability of an influenza vaccination. Prevalences of pneumonia that required hospitalisation were 2-3 times higher in patients compared to controls and tended to be higher in regions with low vaccination rates. The increased pneumonia prevalence in RA patients confirms their status as a risk group. RA patients are vaccinated more frequently than controls, but vaccination rates are still low. The lower pneumonia prevalence in East Germany indicates that vaccination may help to reduce pneumonia in RA.

Epidemiology and Treatment of New-Onset and Established Rheumatoid Arthritis in an Insured US Population

OBJECTIVE

To investigate the epidemiology and treatment of rheumatoid arthritis (RA) in a population broadly representative of employed adults in the US, using a retrospective cohort design.

METHODS

Incident and prevalent RA cohorts were defined from a sample of 4.66 million adults with complete followup data from the period of January 2005 through September 2008 in the Pharmetrics medical claims database. Demographics, comorbidity, and medical therapies were summarized using descriptive statistics.

RESULTS

Median duration in the database was 5.7 years. Age- and sex-adjusted incidence in 2006 was 0.71 per 1,000 persons at risk (n = 3,992) and prevalence in 2005 was 0.63% (n = 30,530). Within 12 months after diagnosis, 65%, 64%, and 20% of the incident cohort had been prescribed corticosteroids, nonbiologic disease-modifying antirheumatic drugs (DMARDs), and tumor necrosis factor (TNF) inhibitors, respectively. Median time to first anti-TNF prescription was 6 months; 31% switched to a second drug and 15% to a third. An aggressive subcohort (11% of incident patients) received more DMARDs (83%) and TNF inhibitors (43%), and was more likely to switch. Twenty-eight percent of incident patients received only symptomatic therapy over a minimum of 1.75 years of followup; these patients were older with more comorbidities and contraindications to methotrexate.

CONCLUSION

In this insured population-based cohort, only two-thirds of newly diagnosed RA patients were prescribed a DMARD in year 1 and 28% received no antirheumatic therapy. Although limited by lack of clinical information and by left-censoring, administrative databases capture clinical practice and suggest that gaps exist in treatment options available to a significant number of patients.

Case definitions in Swedish register data to identify systemic lupus erythematosus

OBJECTIVE

To develop and investigate the utility of several different case definitions for systemic lupus erythematosus (SLE) using national register data in Sweden.

METHODS

The reference standard consisted of clinically confirmed SLE cases pooled from four major clinical centres in Sweden (n=929), and a sample of non-SLE comparators randomly selected from the National Population Register (n=24,267). Demographics, comorbidities, prescriptions and autoimmune disease family history were obtained from multiple registers and linked to the reference standard. We first used previously published SLE definitions to create algorithms for SLE. We also used modern data mining techniques (penalised least absolute shrinkage and selection operator logistic regression, elastic net regression and classification trees) to objectively create data-driven case definitions. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated for the case definitions identified.

RESULTS

Defining SLE by using only hospitalisation data resulted in the lowest sensitivity (0.79). When SLE codes from the outpatient register were included, sensitivity and PPV increased (PPV between 0.97 and 0.98, sensitivity between 0.97 and 0.99). Addition of medication information did not greatly improve the algorithm's performance. The application of data mining methods did not yield different case definitions.

CONCLUSIONS

The use of SLE International Classification of Diseases (ICD) codes in outpatient clinics increased the accuracy for identifying individuals with SLE using Swedish registry data. This study implies that it is possible to use ICD codes from national registers to create a cohort of individuals with SLE.

The use of administrative health care databases to identify patients with rheumatoid arthritis

OBJECTIVE

To validate and compare the decision rules to identify rheumatoid arthritis (RA) in administrative databases.

METHODS

A study was performed using administrative health care data from a population of 1 million people who had access to universal health care. Information was available on hospital discharge abstracts and physician billings. RA cases in health administrative databases were matched 1:4 by age and sex to randomly selected controls without inflammatory arthritis. Seven case definitions were applied to identify RA cases in the health administrative data, and their performance was compared with the diagnosis by a rheumatologist. The validation study was conducted on a sample of individuals with administrative data who received a rheumatologist consultation at the Arthritis Center of Nova Scotia.

RESULTS

We identified 535 RA cases and 2,140 non-RA, noninflammatory arthritis controls. Using the rheumatologist's diagnosis as the gold standard, the overall accuracy of the case definitions for RA cases varied between 68.9% and 82.9% with a kappa statistic between 0.26 and 0.53. The sensitivity and specificity varied from 20.7% to 94.8% and 62.5% to 98.5%, respectively. In a reference population of 1 million, the estimated annual number of incident cases of RA was between 176 and 1,610 and the annual number of prevalent cases was between 1,384 and 5,722.

CONCLUSION

The accuracy of case definitions for the identification of RA cases from rheumatology clinics using administrative health care databases is variable when compared to a rheumatologist's assessment. This should be considered when comparing results across studies. This variability may also be used as an advantage in different study designs, depending on the relative importance of sensitivity and specificity for identifying the population of interest to the research question.

Validity of Diagnostic Codes for Acute Stroke in Administrative Databases: A Systematic Review

OBJECTIVE

To conduct a systematic review of studies reporting on the validity of International Classification of Diseases (ICD) codes for identifying stroke in administrative data.

METHODS

MEDLINE and EMBASE were searched (inception to February 2015) for studies: (a) Using administrative data to identify stroke; or (b) Evaluating the validity of stroke codes in administrative data; and (c) Reporting validation statistics (sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), or Kappa scores) for stroke, or data sufficient for their calculation. Additional articles were located by hand search (up to February 2015) of original papers. Studies solely evaluating codes for transient ischaemic attack were excluded. Data were extracted by two independent reviewers; article quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool.

RESULTS

Seventy-seven studies published from 1976-2015 were included. The sensitivity of ICD-9 430-438/ICD-10 I60-I69 for any cerebrovascular disease was ≥ 82% in most [≥ 50%] studies, and specificity and NPV were both ≥ 95%. The PPV of these codes for any cerebrovascular disease was ≥ 81% in most studies, while the PPV specifically for acute stroke was ≤ 68%. In at least 50% of studies, PPVs were ≥ 93% for subarachnoid haemorrhage (ICD-9 430/ICD-10 I60), 89% for intracerebral haemorrhage (ICD-9 431/ICD-10 I61), and 82% for ischaemic stroke (ICD-9 434/ICD-10 I63 or ICD-9 434&436). For in-hospital deaths, sensitivity was 55%. For cerebrovascular disease or acute stroke as a cause-of-death on death certificates, sensitivity was ≤ 71% in most studies while PPV was ≥ 87%.

CONCLUSIONS

While most cases of prevalent cerebrovascular disease can be detected using 430-438/I60-I69 collectively, acute stroke must be defined using more specific codes. Most in-hospital deaths and death certificates with stroke as a cause-of-death correspond to true stroke deaths. Linking vital statistics and hospitalization data may improve the ascertainment of fatal stroke.

A validation study of a new classification algorithm to identify rheumatoid arthritis using administrative health databases: case-control and cohort diagnostic accuracy studies. Results from the RECord linkage On Rheumatic Diseases study of the Italian Society for Rheumatology

OBJECTIVES

To develop and validate a new algorithm to identify patients with rheumatoid arthritis (RA) and estimate disease prevalence using administrative health databases (AHDs) of the Italian Lombardy region.

DESIGN

Case-control and cohort diagnostic accuracy study.

METHODS

In a randomly selected sample of 827 patients drawn from a tertiary rheumatology centre (training set), clinically validated diagnoses were linked to administrative data including diagnostic codes and drug prescriptions. An algorithm in steps of decreasing specificity was developed and its accuracy assessed calculating sensitivity/specificity, positive predictive value (PPV)/negative predictive value, with corresponding CIs. The algorithm was applied to two validating sets: 106 patients from a secondary rheumatology centre and 6087 participants from the primary care. Alternative algorithms were developed to increase PPV at population level. Crude and adjusted prevalence estimates taking into account algorithm misclassification rates were obtained for the Lombardy region.

RESULTS

The algorithms included: RA certification by a rheumatologist, certification for other autoimmune diseases by specialists, RA code in the hospital discharge form, prescription of disease-modifying antirheumatic drugs and oral glucocorticoids. In the training set, a four-step algorithm identified clinically diagnosed RA cases with a sensitivity of 96.3 (95% CI 93.6 to 98.2) and a specificity of 90.3 (87.4 to 92.7). Both external validations showed highly consistent results. More specific algorithms achieved >80% PPV at the population level. The crude RA prevalence in Lombardy was 0.52%, and estimates adjusted for misclassification ranged from 0.31% (95% CI 0.14% to 0.42%) to 0.37% (0.25% to 0.47%).

CONCLUSIONS

AHDs are valuable tools for the identification of RA cases at the population level, and allow estimation of disease prevalence and to select retrospective cohorts.

Clinic variation in recruitment metrics, patient characteristics and treatment use in a randomized clinical trial of osteoarthritis management

Background

The Patient and PRovider Interventions for Managing Osteoarthritis (OA) in Primary Care (PRIMO) study is one of the first health services trials targeting OA in a multi-site, primary care network. This multi-site approach is important for assessing generalizability of the interventions. These analyses describe heterogeneity in clinic and patient characteristics, as well as recruitment metrics, across PRIMO study clinics.

Methods

Baseline data were obtained from the PRIMO study, which enrolled n = 537 patients from ten Duke Primary Care practices. The following items were examined across clinics with descriptive statistics: (1) Practice Characteristics, including primary care specialty, numbers and specialties of providers, numbers of patients age 55+, urban/rural location and county poverty level; (2) Recruitment Metrics, including rates of eligibility, refusal and randomization; (3) Participants’ Characteristics, including demographic and clinical data (general and OA-related); and (4) Participants’ Self-Reported OA Treatment Use, including pharmacological and non-pharmacological therapies. Intraclass correlation coefficients (ICCs) were computed for participant characteristics and OA treatment use to describe between-clinic variation.

Results

Study clinics varied considerably across all measures, with notable differences in numbers of patients age 55+ (1,507-5,400), urban/rural location (ranging from “rural” to “small city”), and proportion of county households below poverty level (12%-26%). Among all medical records reviewed, 19% of patients were initially eligible (10%-31% across clinics), and among these, 17% were randomized into the study (13%-21% across clinics). There was considerable between-clinic variation, as measured by the ICC (>0.01), for the following patient characteristics and OA treatment use variables: age (means: 60.4-66.1 years), gender (66%-88% female), race (16%-61% non-white), low income status (5%-27%), presence of hip OA (26%-68%), presence both knee and hip OA (23%-61%), physical therapy for knee OA (24%-61%) and hip OA (0%-71%), and use of knee brace with metal supports (0%-18%).

Conclusions

Although PRIMO study sites were part of one primary care practice network in one health care system, clinic and patient characteristics varied considerably, as did OA treatment use. This heterogeneity illustrates the importance of including multiple, diverse sites in trials for knee and hip OA, to enhance the generalizability and evaluate potential for real-world implementation.

Trial registration

Clinical Trial Registration Number: NCT 01435109

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2474-15-413) contains supplementary material, which is available to authorized users.

Mining rich health data from Canadian physician claims: features and face validity

Background

Physician claims data are one of the largest sources of coded health information unique to Canada. There is skepticism from data users about the quality of this data. This study investigated features of diagnostic codes used in the Alberta physician claims database.

Methods

Alberta physician claims from January 1 to March 31, 2011 are analyzed. Claims contain coded diagnoses using the International Classification of Diseases, 9th revision (ICD-9), procedures, physician specialty and service-fee type. Descriptive statistics examined the diversity and frequency of unique ICD-9 diagnostic codes used and the level of code extension (e.g. 3- or 4-digit coding).

Results

A total of 7,441,005 claims by 6,601 physicians were analyzed. The average number of claims per physician was 1,079, with ranges between 1,330 for family medicine, 690 for internal medicine, 722 for surgery, 516 for pediatrics and 409 for neurology. Family physicians used an average of 121 diagnostic codes, internal medicine physicians 32, surgery 36, pediatrics 46 and neurology 27. Overall, 43.5% of claims had a more detailed diagnosis (ICD code with >3 digits). Physicians on a fee-for-service plan submitted 1,184 claims and used 88 unique diagnosis codes on average compared to 438 claims and 44 unique diagnosis codes from physicians on an alternative payment plan (APP).

Conclusions

Face validity of diagnosis coded in physician claims is substantially high and the features of diagnosis codes seem to reasonably reflect the clinical specialty. Physicians submit a diverse array of ICD 9 diagnostic codes and nearly half of the ICD-9 diagnostic codes examined were more detailed than required (i.e. ICD code with >3 digits). Finally, guidelines and policies should be explored to assess the submission of shadow billings for physicians on APPs.

Validity of heart failure diagnoses in administrative databases: a systematic review and meta-analysis

Objective

Heart failure (HF) is an important covariate and outcome in studies of elderly populations and cardiovascular disease cohorts, among others. Administrative data is increasingly being used for long-term clinical research in these populations. We aimed to conduct the first systematic review and meta-analysis of studies reporting on the validity of diagnostic codes for identifying HF in administrative data.

Methods

MEDLINE and EMBASE were searched (inception to November 2010) for studies: (a) Using administrative data to identify HF; or (b) Evaluating the validity of HF codes in administrative data; and (c) Reporting validation statistics (sensitivity, specificity, positive predictive value [PPV], negative predictive value, or Kappa scores) for HF, or data sufficient for their calculation. Additional articles were located by hand search (up to February 2011) of original papers. Data were extracted by two independent reviewers; article quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool. Using a random-effects model, pooled sensitivity and specificity values were produced, along with estimates of the positive (LR+) and negative (LR−) likelihood ratios, and diagnostic odds ratios (DOR = LR+/LR−) of HF codes.

Results

Nineteen studies published from1999–2009 were included in the qualitative review. Specificity was ≥95% in all studies and PPV was ≥87% in the majority, but sensitivity was lower (≥69% in ≥50% of studies). In a meta-analysis of the 11 studies reporting sensitivity and specificity values, the pooled sensitivity was 75.3% (95% CI: 74.7–75.9) and specificity was 96.8% (95% CI: 96.8–96.9). The pooled LR+ was 51.9 (20.5–131.6), the LR− was 0.27 (0.20–0.37), and the DOR was 186.5 (96.8–359.2).

Conclusions

While most HF diagnoses in administrative databases do correspond to true HF cases, about one-quarter of HF cases are not captured. The use of broader search parameters, along with laboratory and prescription medication data, may help identify more cases.

Identification of patients with systemic lupus erythematosus in administrative healthcare databases

Objective

Our aim was to validate and compare decision rules for the identification of patients with systemic lupus erythematosus (SLE) in administrative healthcare databases.

Methods

A retrospective cohort study was performed using administrative health care data from a population of 1 million people with access to universal healthcare. Information was available on hospital discharges and physician billings over a 10-year period. Each SLE case was matched 4:1 by age and gender to randomly selected controls. Seven case definitions were applied to identify SLE cases and their performance compared with the diagnosis by a rheumatologist.

Results

We identified 373 SLE cases and 1492 non-SLE controls, all of whom had been reviewed by a rheumatologist. The overall accuracy of the case definitions for SLE cases varied between 88.2–95.6% with a kappa statistic between 0.53–0.86. The sensitivity varied from 41.0–86.6% and the specificity between 92.4–99.9%. In a total reference population of 1 million the mean estimated annual incidence of SLE was between 29–255 and the mean estimated annual prevalence was between 172–920.

Conclusion

The accuracy of case definitions for the identification of SLE patients in administrative healthcare databases is variable and this should be considered when comparing results across studies. This variability may also be used to advantage in different study designs depending on the relative importance of sensitivity and specificity for identifying the population of interest to the research question.