Empirical validation of the reverse parametric waiting time distribution and standard methods to estimate prescription durations for warfarin

Julie Maria Thrane, Henrik Støvring, Maja Hellfritzsch, Jesper Hallas, Anton Pottegård*

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

5 Downloads (Pure)

Abstract

OBJECTIVES: In many prescription databases, the duration of treatment for the single prescription is not recorded. This study aimed to validate 2 different types of approaches for estimating prescription durations, using the oral anticoagulant warfarin as a case.

METHODS: The approaches undergoing empirical validation covered assumptions of a fixed daily intake of either 0.5 or 1.0 defined daily dose (DDD), as well as estimates based on the reverse parametric waiting time distribution (rWTD), with different sets of covariates. We converted estimates of prescription duration to daily dose and compared them to prescribed daily dose as recorded in a clinical registry (using Bland-Altman plots). Methods were compared based on their average prediction error (logarithmic scale) and their limit of agreement ratio (ratio of mean error ± 1.96 SD after transformation to original scale).

RESULTS: Estimates of daily doses were underestimated by 19% or overestimated by 62% when assumptions of 0.5 or 1.0 DDD were applied. The limit of agreement ratio was 6.721 for both assumptions. The rWTD-based approaches performed better when using the estimated mean value of the inter-arrival density, yielding on average negligible bias (relative difference of 0 to 2%) and with limit of agreement ratios decreasing upon additional covariate adjustment (from 6.857 with no adjustment to 4.036 with the fully adjusted model).

CONCLUSIONS: Comparing the different methods, the rWTD algorithm performed best and led to unbiased estimates of prescribed doses and thus prescription durations and reduced misclassification on the individual level upon inclusion of covariates.

Original languageEnglish
JournalPharmacoepidemiology and Drug Safety
Volume27
Issue number9
Pages (from-to)1011-1018
ISSN1053-8569
DOIs
Publication statusPublished - Sep 2018

Fingerprint

Warfarin
Prescriptions
Registries
Databases

Keywords

  • defined daily dose
  • Pharmacoepidemiology
  • prescription duration
  • validation
  • waiting time distribution
  • warfarin

Cite this

@article{6f1f2f33638f4b8e92bf860b9a75c832,
title = "Empirical validation of the reverse parametric waiting time distribution and standard methods to estimate prescription durations for warfarin",
abstract = "OBJECTIVES: In many prescription databases, the duration of treatment for the single prescription is not recorded. This study aimed to validate 2 different types of approaches for estimating prescription durations, using the oral anticoagulant warfarin as a case.METHODS: The approaches undergoing empirical validation covered assumptions of a fixed daily intake of either 0.5 or 1.0 defined daily dose (DDD), as well as estimates based on the reverse parametric waiting time distribution (rWTD), with different sets of covariates. We converted estimates of prescription duration to daily dose and compared them to prescribed daily dose as recorded in a clinical registry (using Bland-Altman plots). Methods were compared based on their average prediction error (logarithmic scale) and their limit of agreement ratio (ratio of mean error ± 1.96 SD after transformation to original scale).RESULTS: Estimates of daily doses were underestimated by 19{\%} or overestimated by 62{\%} when assumptions of 0.5 or 1.0 DDD were applied. The limit of agreement ratio was 6.721 for both assumptions. The rWTD-based approaches performed better when using the estimated mean value of the inter-arrival density, yielding on average negligible bias (relative difference of 0 to 2{\%}) and with limit of agreement ratios decreasing upon additional covariate adjustment (from 6.857 with no adjustment to 4.036 with the fully adjusted model).CONCLUSIONS: Comparing the different methods, the rWTD algorithm performed best and led to unbiased estimates of prescribed doses and thus prescription durations and reduced misclassification on the individual level upon inclusion of covariates.",
keywords = "Warfarin, defined daily dose, Pharmacoepidemiology, prescription duration, validation, waiting time distribution, warfarin",
author = "Thrane, {Julie Maria} and Henrik St{\o}vring and Maja Hellfritzsch and Jesper Hallas and Anton Potteg{\aa}rd",
note = "{\circledC} 2018 John Wiley & Sons, Ltd.",
year = "2018",
month = "9",
doi = "10.1002/pds.4581",
language = "English",
volume = "27",
pages = "1011--1018",
journal = "Pharmacoepidemiology and Drug Safety",
issn = "1053-8569",
publisher = "JohnWiley & Sons Ltd.",
number = "9",

}

Empirical validation of the reverse parametric waiting time distribution and standard methods to estimate prescription durations for warfarin. / Thrane, Julie Maria; Støvring, Henrik; Hellfritzsch, Maja; Hallas, Jesper; Pottegård, Anton.

In: Pharmacoepidemiology and Drug Safety, Vol. 27, No. 9, 09.2018, p. 1011-1018.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Empirical validation of the reverse parametric waiting time distribution and standard methods to estimate prescription durations for warfarin

AU - Thrane, Julie Maria

AU - Støvring, Henrik

AU - Hellfritzsch, Maja

AU - Hallas, Jesper

AU - Pottegård, Anton

N1 - © 2018 John Wiley & Sons, Ltd.

PY - 2018/9

Y1 - 2018/9

N2 - OBJECTIVES: In many prescription databases, the duration of treatment for the single prescription is not recorded. This study aimed to validate 2 different types of approaches for estimating prescription durations, using the oral anticoagulant warfarin as a case.METHODS: The approaches undergoing empirical validation covered assumptions of a fixed daily intake of either 0.5 or 1.0 defined daily dose (DDD), as well as estimates based on the reverse parametric waiting time distribution (rWTD), with different sets of covariates. We converted estimates of prescription duration to daily dose and compared them to prescribed daily dose as recorded in a clinical registry (using Bland-Altman plots). Methods were compared based on their average prediction error (logarithmic scale) and their limit of agreement ratio (ratio of mean error ± 1.96 SD after transformation to original scale).RESULTS: Estimates of daily doses were underestimated by 19% or overestimated by 62% when assumptions of 0.5 or 1.0 DDD were applied. The limit of agreement ratio was 6.721 for both assumptions. The rWTD-based approaches performed better when using the estimated mean value of the inter-arrival density, yielding on average negligible bias (relative difference of 0 to 2%) and with limit of agreement ratios decreasing upon additional covariate adjustment (from 6.857 with no adjustment to 4.036 with the fully adjusted model).CONCLUSIONS: Comparing the different methods, the rWTD algorithm performed best and led to unbiased estimates of prescribed doses and thus prescription durations and reduced misclassification on the individual level upon inclusion of covariates.

AB - OBJECTIVES: In many prescription databases, the duration of treatment for the single prescription is not recorded. This study aimed to validate 2 different types of approaches for estimating prescription durations, using the oral anticoagulant warfarin as a case.METHODS: The approaches undergoing empirical validation covered assumptions of a fixed daily intake of either 0.5 or 1.0 defined daily dose (DDD), as well as estimates based on the reverse parametric waiting time distribution (rWTD), with different sets of covariates. We converted estimates of prescription duration to daily dose and compared them to prescribed daily dose as recorded in a clinical registry (using Bland-Altman plots). Methods were compared based on their average prediction error (logarithmic scale) and their limit of agreement ratio (ratio of mean error ± 1.96 SD after transformation to original scale).RESULTS: Estimates of daily doses were underestimated by 19% or overestimated by 62% when assumptions of 0.5 or 1.0 DDD were applied. The limit of agreement ratio was 6.721 for both assumptions. The rWTD-based approaches performed better when using the estimated mean value of the inter-arrival density, yielding on average negligible bias (relative difference of 0 to 2%) and with limit of agreement ratios decreasing upon additional covariate adjustment (from 6.857 with no adjustment to 4.036 with the fully adjusted model).CONCLUSIONS: Comparing the different methods, the rWTD algorithm performed best and led to unbiased estimates of prescribed doses and thus prescription durations and reduced misclassification on the individual level upon inclusion of covariates.

KW - Warfarin

KW - defined daily dose

KW - Pharmacoepidemiology

KW - prescription duration

KW - validation

KW - waiting time distribution

KW - warfarin

U2 - 10.1002/pds.4581

DO - 10.1002/pds.4581

M3 - Journal article

VL - 27

SP - 1011

EP - 1018

JO - Pharmacoepidemiology and Drug Safety

JF - Pharmacoepidemiology and Drug Safety

SN - 1053-8569

IS - 9

ER -