Dataset Information

Data management and data analysis techniques in pharmacoepidemiological studies using a pre-planned multi-database approach: a systematic literature review.

ABSTRACT: To identify pharmacoepidemiological multi-database studies and to describe data management and data analysis techniques used for combining data.Systematic literature searches were conducted in PubMed and Embase complemented by a manual literature search. We included pharmacoepidemiological multi-database studies published from 2007 onwards that combined data for a pre-planned common analysis or quantitative synthesis. Information was retrieved about study characteristics, methods used for individual-level analyses and meta-analyses, data management and motivations for performing the study.We found 3083 articles by the systematic searches and an additional 176 by the manual search. After full-text screening of 75 articles, 22 were selected for final inclusion. The number of databases used per study ranged from 2 to 17 (median?=?4.0). Most studies used a cohort design (82%) instead of a case-control design (18%). Logistic regression was most often used for individual-level analyses (41%), followed by Cox regression (23%) and Poisson regression (14%). As meta-analysis method, a majority of the studies combined individual patient data (73%). Six studies performed an aggregate meta-analysis (27%), while a semi-aggregate approach was applied in three studies (14%). Information on central programming or heterogeneity assessment was missing in approximately half of the publications. Most studies were motivated by improving power (86%).Pharmacoepidemiological multi-database studies are a well-powered strategy to address safety issues and have increased in popularity. To be able to correctly interpret the results of these studies, it is important to systematically report on database management and analysis techniques, including central programming and heterogeneity testing.

SUBMITTER: Bazelier MT

PROVIDER: S-EPMC5034829 | biostudies-other | 2015 Sep

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Data management and data analysis techniques in pharmacoepidemiological studies using a pre-planned multi-database approach: a systematic literature review.

Bazelier Marloes T MT Eriksson Irene I de Vries Frank F Schmidt Marjanka K MK Raitanen Jani J Haukka Jari J Starup-Linde Jakob J De Bruin Marie L ML Andersen Morten M

Pharmacoepidemiology and drug safety 20150714 9

<h4>Purpose</h4>To identify pharmacoepidemiological multi-database studies and to describe data management and data analysis techniques used for combining data.<h4>Methods</h4>Systematic literature searches were conducted in PubMed and Embase complemented by a manual literature search. We included pharmacoepidemiological multi-database studies published from 2007 onwards that combined data for a pre-planned common analysis or quantitative synthesis. Information was retrieved about study characte ...[more]

PMID: 26175179

Similar Datasets

Project description:BackgroundSelecting the proper parameter settings for bioinformatic software tools is challenging. Not only will each parameter have an individual effect on the outcome, but there are also potential interaction effects between parameters. Both of these effects may be difficult to predict. To make the situation even more complex, multiple tools may be run in a sequential pipeline where the final output depends on the parameter configuration for each tool in the pipeline. Because of the complexity and difficulty of predicting outcomes, in practice parameters are often left at default settings or set based on personal or peer experience obtained in a trial and error fashion. To allow for the reliable and efficient selection of parameters for bioinformatic pipelines, a systematic approach is needed.ResultsWe present doepipeline, a novel approach to optimizing bioinformatic software parameters, based on core concepts of the Design of Experiments methodology and recent advances in subset designs. Optimal parameter settings are first approximated in a screening phase using a subset design that efficiently spans the entire search space, then optimized in the subsequent phase using response surface designs and OLS modeling. Doepipeline was used to optimize parameters in four use cases; 1) de-novo assembly, 2) scaffolding of a fragmented genome assembly, 3) k-mer taxonomic classification of Oxford Nanopore Technologies MinION reads, and 4) genetic variant calling. In all four cases, doepipeline found parameter settings that produced a better outcome with respect to the characteristic measured when compared to using default values. Our approach is implemented and available in the Python package doepipeline.ConclusionsOur proposed methodology provides a systematic and robust framework for optimizing software parameter settings, in contrast to labor- and time-intensive manual parameter tweaking. Implementation in doepipeline makes our methodology accessible and user-friendly, and allows for automatic optimization of tools in a wide range of cases. The source code of doepipeline is available at https://github.com/clicumu/doepipeline and it can be installed through conda-forge.

Project description:BackgroundIn women with late preterm pre-eclampsia, the optimal time to initiate delivery is unclear because limitation of maternal disease progression needs to be balanced against infant complications. The aim of this trial was to determine whether planned earlier initiation of delivery reduces maternal adverse outcomes without substantial worsening of neonatal or infant outcomes, compared with expectant management (usual care) in women with late preterm pre-eclampsia.MethodsIn this parallel-group, non-masked, multicentre, randomised controlled trial done in 46 maternity units across England and Wales, we compared planned delivery versus expectant management (usual care) with individual randomisation in women with late preterm pre-eclampsia from 34 to less than 37 weeks' gestation and a singleton or dichorionic diamniotic twin pregnancy. The co-primary maternal outcome was a composite of maternal morbidity or recorded systolic blood pressure of at least 160 mm Hg with a superiority hypothesis. The co-primary perinatal outcome was a composite of perinatal deaths or neonatal unit admission up to infant hospital discharge with a non-inferiority hypothesis (non-inferiority margin of 10% difference in incidence). Analyses were by intention to treat, together with a per-protocol analysis for the perinatal outcome. The trial was prospectively registered with the ISRCTN registry, ISRCTN01879376. The trial is closed to recruitment but follow-up is ongoing.FindingsBetween Sept 29, 2014, and Dec 10, 2018, 901 women were recruited. 450 women (448 women and 471 infants analysed) were allocated to planned delivery and 451 women (451 women and 475 infants analysed) to expectant management. The incidence of the co-primary maternal outcome was significantly lower in the planned delivery group (289 [65%] women) compared with the expectant management group (338 [75%] women; adjusted relative risk 0·86, 95% CI 0·79-0·94; p=0·0005). The incidence of the co-primary perinatal outcome by intention to treat was significantly higher in the planned delivery group (196 [42%] infants) compared with the expectant management group (159 [34%] infants; 1·26, 1·08-1·47; p=0·0034). The results from the per-protocol analysis were similar. There were nine serious adverse events in the planned delivery group and 12 in the expectant management group.InterpretationThere is strong evidence to suggest that planned delivery reduces maternal morbidity and severe hypertension compared with expectant management, with more neonatal unit admissions related to prematurity but no indicators of greater neonatal morbidity. This trade-off should be discussed with women with late preterm pre-eclampsia to allow shared decision making on timing of delivery.FundingNational Institute for Health Research Health Technology Assessment Programme.

Project description:BackgroundPoor diabetes management prior to conception, results in increased rates of fetal malformations and other adverse pregnancy outcomes. We describe the development of an integrated, pre-pregnancy management strategy to improve pregnancy outcomes among women of reproductive age with diabetes in a multi-ethnic district.MethodsThe strategy included (i) a narrative literature review of contraception and pre-pregnancy interventions for women with diabetes and development of a draft plan; (ii) a chart review of pregnancy outcomes (e.g. congenital malformations, neonatal hypoglycaemia and caesarean sections) among women with type 1 diabetes (T1D) (n?=?53) and type 2 diabetes (T2D) (n?=?46) between 2010 and 2015 (iii) interview surveys of women with T1D and T2D (n?=?15), and local health care professionals (n?=?13); (iv) two focus groups (n?=?4) and one-to-one interviews with women with T1D and T2D from an Australian background (n?=?5), women with T2D from cultural and linguistically diverse (CALD) (n?=?7) and indigenous backgrounds (n?=?1) and partners of CALD women (n?=?3); and (v) two group meetings, one comprising predominantly primary care, and another comprising district-wide multidisciplinary inter-sectoral professionals, where components of the intervention strategy were finalised using a Delphi approach for development of the final plan.ResultsOur literature review showed that a range of interventions, particularly multifaceted educational programs for women and healthcare professionals, significantly increased contraception uptake, and reduced adverse outcomes of pregnancy (e.g. malformations and stillbirth). Our chart-review showed that local rates of adverse pregnancy outcomes were similarly poor among women with both T1D and T2D (e.g. major congenital malformations [9.1% vs 8.9%] and macrosomia [34.7% vs 24.4%]). Challenges included lack of knowledge among women and healthcare professionals relating to diabetes management and limited access to specialist pre-pregnancy care. Group meetings led to a consensus to develop a district-wide approach including healthcare professional and patient education and a structured approach to identification and optimisation of self-management, including contraception, in women of reproductive age with diabetes.ConclusionsSufficient evidence exists for consensus on a district-wide strategy to improve pre-pregnancy management among women with pre-existing diabetes.

Project description:BackgroundPre-eclampsia is a pregnancy disorder, characterised by hypertension and multisystem complications in the mother. The adverse outcomes of pre-eclampsia include severe hypertension, stroke, renal and hepatic injury, haemorrhage, fetal growth restriction and even death. The optimal time to instigate delivery to prevent morbidity when pre-eclampsia occurs between 34 and 37?weeks' gestation, without increasing problems related to infant immaturity or complications, remains unclear.Methods/designThe PHOENIX trial is a non-masked, randomised controlled trial, comparing planned early delivery (with initiation of delivery within 48?h of randomisation) with usual care (expectant management) in women with pre-eclampsia between 34+?0 and 36+?6 weeks' gestation. The primary objectives of the trial are to determine if planned delivery reduces adverse maternal outcomes, without increasing the short-term harm to infants (composite of perinatal deaths or neonatal unit admissions up to infant hospital discharge) or impacting long-term infant neurodevelopmental status at 2 years corrected age (Parent Report of Cognitive Abilities-Revised).DiscussionCurrent practice in the UK at the time of trial commencement for management of pre-eclampsia varies by gestation. Previous trials have shown that in women with pre-eclampsia after 37?weeks of gestion, delivery is initiated, as maternal complications are reduced without increasing fetal risks. Prior to 34?weeks of gestation, usual management aims to prolong pregnancy for fetal benefit, unless severe complications occur, necessitating preterm delivery. This trial aims to address the uncertainty for women where the balance of benefits and risks of delivery compared to expectant management are uncertain. Previous trials in this area have been undertaken, but have not provided a definitive answer, and the research question remains active. The results of this trial are expected to influence clinical practice internationally, through direct adoption and by incorporation into guidelines in countries with similar settings.Trial registrationISRCTN01879376 . Registered on 25 November 2013.

Dataset Information

Data management and data analysis techniques in pharmacoepidemiological studies using a pre-planned multi-database approach: a systematic literature review.

Publications

Data management and data analysis techniques in pharmacoepidemiological studies using a pre-planned multi-database approach: a systematic literature review.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets