Original Article
Venous thromboembolism discharge diagnoses in the Danish National Patient Registry should be used with caution

https://doi.org/10.1016/j.jclinepi.2009.03.018Get rights and content

Abstract

Objective

We validated discharge diagnoses of venous thromboembolism (VTE) in the Danish National Patient Registry.

Study Design and Setting

We identified all first-time VTE discharge diagnoses in the Danish National Patient Registry among participants of the Danish cohort study “Diet, Cancer, and Health”, in the period from 1994 to 2006. Medical records were retrieved and VTE diagnoses were verified by one of the authors using a standard protocol. The positive predictive value (PPV) of a discharge diagnosis of VTE was calculated as percent of registry diagnoses with the corresponding true diagnosis in the chart among all registry diagnoses.

Results

We retrieved medical records from 1,100 of 1,135 participants (96.9%) registered with a discharge diagnosis of VTE; 626 diagnoses were confirmed and 17 were considered probable. The PPV of diagnoses coded at wards was 75.0% (95% confidence interval: 71.9, 77.9). Diagnoses from emergency departments were not valid. The PPV varied by type of VTE (deep venous thrombosis and pulmonary embolism), type of diagnosis (primary or secondary), and sex.

Conclusion

Data on VTE obtained from administrative registries are a valuable source of information but should be used with caution in medical research.

Introduction

Venous thromboembolism (VTE), that is, deep venous thrombosis (DVT) and pulmonary embolism (PE) is a common disease, with substantial clinical implications. Patients with VTE have a mortality of 16–21% the first year after diagnosis; the cumulative incidence of recurrent VTE is 30% and of postthrombotic syndrome is 29% after 8 years of follow-up [1], [2], [3], [4]. Possible risk factors for VTE, identified in recent cohort studies [2], [5], [6], [7], [8], [9], [10], [11], include both genetic predispositions and acquired factors. However, 25–50% of VTE events remain idiopathic, underscoring the need for well-designed studies elucidating the causes of this disease.

Hospital discharge registries are a potentially valuable source of data on VTE [12], with data readily available at low cost. Routine data collection and often universal registration of people in the target population help avoid biases of recall, diagnosis, or selection. However, lack of investigator's control over data collection and quality are important disadvantages of registry data [12], [13]. Quality of registry data varies considerably from disease to disease and poor data quality may invalidate results of epidemiologic studies. Thus, it is essential to validate routinely collected registry data before using them for research. Few studies have examined the data quality of VTE diagnoses in registries in selected patient groups [14], [15], [16], [17]. Validation was done only by secondary diagnoses to determine the frequencies of in hospital-acquired VTE by others [18], [19]. We aimed to determine the positive predictive value (PPV) of VTE diagnoses in the Danish National Patient Registry, a nationwide hospital discharge register.

Section snippets

The Diet, Cancer, and Health study

Diet, Cancer, and Health is a prospective cohort study, with the primary objective to investigate the etiologic role of diet in the development of cancer. The study has been described in detail elsewhere [20], [21]. Briefly, between December 1993 and May 1997, 80,996 men and 79,729 women, aged 50–64 years, were invited to participate in the study; 27,178 men and 29,876 women accepted the invitation. Eligible cohort members were born in Denmark, were living in the urban areas of Copenhagen and

Validation of VTE diagnoses in the Danish National Patient Registry

Figure 1 shows the results of the review of medical records. We identified 1,135 participants in the Diet, Cancer, and Health cohort study with a first-time VTE diagnosis recorded in the Danish National Patient Registry during the follow-up period. We were able to retrieve the medical records from 1,100 (96.9%) of these admissions. Of the 1,100 VTE diagnoses, 454 were given in emergency departments (Fig. 2). All but 36 patients were discharged into a ward unit within 1 week (n = 418). Of these,

Discussion

The PPV varied according to the subdiagnoses (DVT and PE), types of diagnoses (primary or secondary), type of hospital department, and gender. The highest (up to 87%) PPVs were found for PE diagnoses, in particular if it was a primary diagnosis; for diagnoses coded at wards (in contrast to emergency departments); and for diagnoses among men.

A number of other studies have examined the quality of VTE diagnoses in administrative databases [14], [15], [16], [17], [18], [19]. Two studies estimated

References (24)

  • F.A. Anderson et al.

    A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT Study

    Arch Intern Med

    (1991)
  • P.O. Hansson et al.

    Deep vein thrombosis and pulmonary embolism in the general population. ‘The Study of Men Born in 1913’

    Arch Intern Med

    (1997)
  • Cited by (0)

    View full text