Kirsten J Lum1,2, Craig W Newcomb1, Jason A Roy1,2, Dena M Carbonari1,2, M Elle Saine1,2, Serena Cardillo3, Harshvinder Bhullar4, Arlene M Gallagher5, Vincent Lo Re6,7,8. 1. Center for Clinical Epidemiology and Biostatistics, Department of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. 2. Center for Pharmacoepidemiology Research and Training, Department of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. 3. Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. 4. IMS Health, London, UK. 5. Clinical Practice Research Datalink, London, UK. 6. Center for Clinical Epidemiology and Biostatistics, Department of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. vincentl@mail.med.upenn.edu. 7. Center for Pharmacoepidemiology Research and Training, Department of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. vincentl@mail.med.upenn.edu. 8. Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. vincentl@mail.med.upenn.edu.
Abstract
PURPOSE: The extent to which days' supply data are missing in pharmacoepidemiologic databases and effective methods for estimation is unknown. We determined the percentage of missing days' supply on prescription and patient levels for oral anti-diabetic drugs (OADs) and evaluated three methods for estimating days' supply within the Clinical Practice Research Datalink (CPRD) and The Health Improvement Network (THIN). METHODS: We estimated the percentage of OAD prescriptions and patients with missing days' supply in each database from 2009 to 2013. Within a random sample of prescriptions with known days' supply, we measured the accuracy of three methods to estimate missing days' supply by imputing the following: (1) 28 days' supply, (2) mode number of tablets/day by drug strength and number of tablets/prescription, and (3) number of tablets/day via a machine learning algorithm. We determined incidence rates (IRs) of acute myocardial infarction (AMI) using each method to evaluate the impact on ascertainment of exposure time and outcomes. RESULTS: Days' supply was missing for 24 % of OAD prescriptions in CPRD and 33 % in THIN (affecting 48 and 57 % of patients, respectively). Methods 2 and 3 were very accurate in estimating days' supply for OADs prescribed at a consistent number of tablets/day. Method 3 was more accurate for OADs prescribed at varying number of tablets/day. IRs of AMI were similar across methods for most OADs. CONCLUSIONS: Missing days' supply is a substantial problem in both databases. Method 2 is easy and very accurate for most OADs and results in IRs comparable to those from method 3.
PURPOSE: The extent to which days' supply data are missing in pharmacoepidemiologic databases and effective methods for estimation is unknown. We determined the percentage of missing days' supply on prescription and patient levels for oral anti-diabetic drugs (OADs) and evaluated three methods for estimating days' supply within the Clinical Practice Research Datalink (CPRD) and The Health Improvement Network (THIN). METHODS: We estimated the percentage of OAD prescriptions and patients with missing days' supply in each database from 2009 to 2013. Within a random sample of prescriptions with known days' supply, we measured the accuracy of three methods to estimate missing days' supply by imputing the following: (1) 28 days' supply, (2) mode number of tablets/day by drug strength and number of tablets/prescription, and (3) number of tablets/day via a machine learning algorithm. We determined incidence rates (IRs) of acute myocardial infarction (AMI) using each method to evaluate the impact on ascertainment of exposure time and outcomes. RESULTS: Days' supply was missing for 24 % of OAD prescriptions in CPRD and 33 % in THIN (affecting 48 and 57 % of patients, respectively). Methods 2 and 3 were very accurate in estimating days' supply for OADs prescribed at a consistent number of tablets/day. Method 3 was more accurate for OADs prescribed at varying number of tablets/day. IRs of AMI were similar across methods for most OADs. CONCLUSIONS: Missing days' supply is a substantial problem in both databases. Method 2 is easy and very accurate for most OADs and results in IRs comparable to those from method 3.
Entities:
Keywords:
Clinical Practice Research Datalink (CPRD); Days’ supply; Electronic medical record databases; Missing data; Pharmacoepidemiology; The Health Improvement Network (THIN)
Authors: Ronac Mamtani; Kevin Haynes; Warren B Bilker; David J Vaughn; Brian L Strom; Karen Glanz; James D Lewis Journal: J Natl Cancer Inst Date: 2012-08-09 Impact factor: 13.506
Authors: Lucia Cea Soriano; Brian T Bateman; Luis A García Rodríguez; Sonia Hernández-Díaz Journal: Pharmacoepidemiol Drug Saf Date: 2014-05-02 Impact factor: 2.890
Authors: Tarek A Hammad; Mara A McAdams; Andrea Feight; Solomon Iyasu; Gerald J Dal Pan Journal: Pharmacoepidemiol Drug Saf Date: 2008-12 Impact factor: 2.890
Authors: Dena M Carbonari; M Elle Saine; Craig W Newcomb; Betina Blak; Jason A Roy; Kevin Haynes; Jennifer Wood; Arlene M Gallagher; Harshvinder Bhullar; Serena Cardillo; Sean Hennessy; Brian L Strom; Vincent Lo Re Journal: Pharmacoepidemiol Drug Saf Date: 2015-07-27 Impact factor: 2.890
Authors: Andrea V Margulis; Adel Abou-Ali; Marian M Strazzeri; Yulan Ding; Fatmatta Kuyateh; Eric Y Frimpong; Mark S Levenson; Tarek A Hammad Journal: Pharmacoepidemiol Drug Saf Date: 2013-06-03 Impact factor: 2.890
Authors: David Price; Mike Thomas; John Haughney; Richard A Lewis; Anne Burden; Julie von Ziegenweidt; Alison Chisholm; Elizabeth V Hillyer; Christopher J Corrigan Journal: Respir Med Date: 2013-05-03 Impact factor: 3.415
Authors: M Gore; N Brix Finnerup; A Sadosky; K-S Tai; J C Cappelleri; J Mardekian; C George Rice; E Nieshoff Journal: Spinal Cord Date: 2012-09-04 Impact factor: 2.772