Assessing the Saskatchewan database for outcomes research studies of depression and its treatment

Abstract

This study was conducted to evaluate the validity of using the Saskatchewan Health administrative claims databases for conducting depression research. To develop a claims-based definition of depression, we identified a cohort of individuals who began a “new” period of antidepressant use (no use 180 days prior) from which we selected a stratified random sample (n = 600) for medical record abstraction. The medical record diagnosis was used as the gold standard for judging our database definitions of depression. After defining a primary database definition of depression, we tried to refine it using medically probable scenarios and assessed refinement by agreement statistics. Defining depression with ICD9 codes 296 (affective disorders), 309 (adjustment reaction), and 311 (depressive disorders), the sensitivity (Se), specificity (Sp), positive (PV+) and negative predictive (PV−) values were: 71%, 85%, 86%, and 70%, respectively. Algorithms that limited the number of false-negatives resulted in: Se = 84% and PV− = 77% whereas those that limited false-positives resulted in: Sp = 90% and PV+ = 86%. Although our depression definition requires treatment with antidepressants, this definition will allow us to conduct future studies of depression and its treatment using the Saskatchewan Health databases.

Introduction

The use of carefully conducted studies based on claims databases can provide insights about medical and pharmacy resource utilization in actual clinical practice. For example, much can be gained by further research on the rapidly expanding use of antidepressants. The major approved use of antidepressants, by the U.S. Food and Drug Administration (FDA), is for the treatment of depression. Other approved uses for specific antidepressants include anxiety disorder, obsessive compulsive disorder, bulimia nervosa, panic disorder, and most recently, social anxiety disorder [1]. Documented non-approved uses of antidepressants include phobic disorders, attention deficit disorder with hyperactivity, duodenal or peptic ulcer disease, premenstrual syndrome (PMS), allergic conditions, and post-traumatic stress disorder 1, 2, 3. Antidepressants are often prescribed for depression that co-exists with other disorders such as migraine 4, 5, 6, chronic fatigue syndrome 6, 7, fibromyalgia [6], obesity [1], irritable bowel syndrome [6], chronic pain syndrome 6, 8, and schizophrenia [1].

Because of the numerous conditions for which antidepressants are prescribed and the sensitive nature of mental illnesses, using administrative databases for conducting research on depression must be done with caution. Certain insurance plans may have reimbursement restrictions for specific International Classification of Disease, ninth revision, codes (ICD-9) [9] or may limit the number of ICD-9 codes that can be entered for each physician visit. In addition, due to the potential for stigma associated with mental illnesses in general, physicians and/or patients may prefer to keep a diagnosis of depression from appearing in the patient's medical records. Furthermore, depression is a difficult diagnosis to make due to its varied presentation and that there are no laboratory tests to determine a definitive diagnosis [10].

The purpose of this study was to determine whether the Saskatchewan Health (SH) databases would provide appropriate and adequate information for conducting research on depression and its treatment. In particular, our focus was on whether a depression diagnosis could be identified using SH administrative claims data because of the potential for incompleteness of the outpatient claims and the caveats mentioned above. For those individuals who were dispensed antidepressants, we examined a definition of depression using 3-digit ICD-9 codes based on outpatient diagnoses. We compared this definition with that derived from medical record abstraction, our “gold standard.” We considered several different algorithms for improving the definition using the patient's comorbidities, concurrent medications, and fee-for-service codes.