Variation in the reporting of outcomes among pregnant women with epilepsy: a systematic review

Abstract

Studies on pregnant women with epilepsy should evaluate both neurological and pregnancy outcomes. We undertook a systematic review of the literature of studies on pregnant women with epilepsy to collate the outcomes reported, and the quality of outcomes report in these studies. We searched major electronic databases (from 1999 until January 2015). Two independent reviewers selected studies and extracted data on study design, the risk of bias of the studies, journal impact factor and the quality of reported outcomes. We assessed the quality outcomes report using a six items standardised tool (score range 0-6). There were 70 different outcomes reported in 232 studies (maternal neurological (13/70, 19%), fetal and neonatal (28/70, 40%), and obstetric outcomes (29/70, 41%)). Most studies reported on major congenital fetal abnormalities (103/232, 44%), followed by live birth (60/232, 26%). Quality of the reported outcomes was poor (mean 1.54, SD 1.36). It was associated with journal impact factor (p=0.007), but not with study design (p=0.60), or risk of bias (p=0.17). The outcomes reported in studies on pregnant women with epilepsy varied widely, and the quality of the outcomes report was poor. There is a need to identify a set of core outcome to harmonise reporting in future clinical studies.

Core outcome sets in women's and newborn health: a systematic review

“…All four had made a systematic review of the literature to identify potential core outcomes (Table S3). Three core outcome sets published their systematic reviews separately . Three reviews clearly described their search strategy, two searched multiple bibliographical databases (range 3–4) and a single review limited their search to the Cochrane Pregnancy and Childbirth Group's Specialised Register.…”

“…Twenty systematic reviews characterising the inconsistency in outcome reporting across a broad range of relevant healthcare conditions were identified (Table S2). 3,11,[19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35] Seventeen systematic reviews (85%) extracted and mapped outcome reporting across included studies, three (15%) mapped predefined outcomes across included studies, and two (10%) investigated outcome reporting bias. 36 The scope was clearly stated in a majority of systematic reviews (n = 12; 60%).…”

“…Three core outcome sets published their systematic reviews separately. 26,28,34,35,40 Three reviews clearly described their search strategy, two searched multiple bibliographical databases (range 3-4) and a single review limited their search to the Cochrane Pregnancy and Childbirth Group's Specialised Register. All reviews applied a date restriction and two described the rationale for this: Van 't Hooft et al 37 included randomised trials published after the first Consolidated Standards of Reporting Trials (CONSORT) statement and Potter et al 11 included studies from 1980 onwards following the development of modern reconstructive surgical techniques.…”

Weighing the risks of valproate in women who could become pregnant

“…Maternal mortality is almost 10 times greater for women with epilepsy than for those without epilepsy (100 versus 11/100 000 pregnancies) 57. Maternal mortality has fallen, but epilepsy related deaths increased over the past 30 years 58.…”

Reducing research waste through the standardisation of outcomes and definitions

“…repeating measures or training in the use of measurement tools. Researchers in women's health have applied this approach to the assessment of the quality of outcome reporting, and these assessments have presented a troubling picture . The situation is even more disconcerting when it comes to the reporting of methods to enhance the quality of outcome measures …”