Testing for coeliac disease rarely leads to a diagnosis: a population-based study.

BACKGROUND: Coeliac disease (CD) has an estimated prevalence of ∼1% in Europe with a significant gap between undiagnosed and diagnosed CD. Active case finding may help to bridge this gap yet the diagnostic yield of such active case finding in general practice by serological testing is unknown. OBJECTIVE: The aim of this study was to determine (1) the frequency of diagnosed CD in the general population, and (2) to investigate the yield of active case finding by general practitioners. METHODS: Electronic medical records of 207.200 patients registered in 49 general practices in The Netherlands in 2016 were analysed. An extensive search strategy, based on International Classification of Primary Care codes, free text and diagnostic test codes was performed to search CD- or gluten-related contacts. RESULTS: The incidence of CD diagnosis in general practice in 2016 was 0.01%. The prevalence of diagnosed CD reported in the general practice in the Netherlands was 0.19%, and considerably higher than previously reported in the general population. During the one year course of the study 0.95% of the population had a gluten-related contact with their GP; most of them (72%) were prompted by gastrointestinal complaints. Serological testing was performed in 66% (n = 1296) of these patients and positive in only 1.6% (n = 21). CONCLUSION: The number of diagnosed CD patients in the Netherlands is substantially higher than previously reported. This suggests that the gap between diagnosed and undiagnosed patients is lower than generally assumed. This may explain that despite a high frequency of gluten-related consultations in general practice the diagnostic yield of case finding by serological testing is low.Key pointsThe diagnostic approach of GPs regarding CD and the diagnostic yield is largely unknownCase finding in a primary health care practice has a low yield of 1.6%CD testing was mostly prompted by consultation for gastrointestinal symptomsThere is a heterogeneity in types of serological test performed in primary care.

Diagnostic behaviour of general practitioners when suspecting Lyme disease: a database study from 2010-2015.

BACKGROUND: Due to the raised public awareness of Lyme Borreliosis (LB), its increased incidence and the increased availability of serological tests, the demand for diagnostic testing on LB has increased. This may affect the diagnostic behaviour of general practitioners (GPs). Aim of our study was to describe GPs' diagnostic behaviour when suspecting LB. METHODS: In this descriptive study from January 2010 to June 2015, we used the anonymized electronic medical records of 56,996 patients registered in 12 general practices in Amsterdam, The Netherlands. The target population was identified by means of an extensive search strategy, based on International Classification of Primary Care (ICPC-1) codes, free text and diagnostic test codes. All contacts related to LB were included in the analysis. RESULTS: 2311 patients were included, accounting for 3861 LB contacts and 2619 LB episodes. The distribution of LB contacts showed annual peaks during spring and summer. Serological testing was performed in 36.4% of LB episodes and was mostly requested in patients presenting with general symptoms (71.4%). Unnecessary testing often occurred and only 5.9% of the tests turned out to be positive by immunoblot. From January 2010 to June 2015, no significant differences were found in the number of requested serological tests. The level of serological testing during LB episodes differed significantly between the general practices (19.2% to 75.8%). CONCLUSIONS: Contrary to clinical guidelines, GPs regularly requested serology even when there was a low suspicion of LB. The development of an easy-to-use diagnostic algorithm may decrease overuse of diagnostic tests and thereby reduce overtreatment of LB.

Usefulness of primary care electronic networks to assess the incidence of chlamydia, diagnosed by general practitioners.

BACKGROUND: Chlamydia is the most common curable sexually transmitted infection (STI) in the Netherlands. The majority of chlamydia diagnoses are made by general practitioners (GPs). Baseline data from primary care will facilitate the future evaluation of the ongoing large population-based screening in the Netherlands. The aim of this study was to assess the usefulness of electronic medical records for monitoring the incidence of chlamydia cases diagnosed in primary care in the Netherlands. METHODS: In the electronic records of two regional and two national networks, we identified chlamydia diagnoses by means of ICPC codes (International Classification of Primary Care), laboratory results in free text and the prescription of antibiotics. The year of study was 2007 for the two regional networks and one national network, for the other national network the year of study was 2005. We calculated the incidence of diagnosed chlamydia cases per sex, age group and degree of urbanization. RESULTS: A large diversity was observed in the way chlamydia episodes were coded in the four different GP networks and how easily information concerning chlamydia diagnoses could be extracted. The overall incidence ranged from 103.2/100,000 to 590.2/100,000. Differences were partly related to differences between patient populations. Nevertheless, we observed similar trends in the incidence of chlamydia diagnoses in all networks and findings were in line with earlier reports. CONCLUSIONS: Electronic patient records, originally intended for individual patient care in general practice, can be an additional source of data for monitoring chlamydia incidence in primary care and can be of use in assessing the future impact of population-based chlamydia screening programs. To increase the usefulness of data we recommend more efforts to standardize registration by (specific) ICPC code and laboratory results across the existing GP networks.