Assessing Heterogeneity in Meta-Analysis: Q Statistic or I

Social and Methodological Psychology, Universidad Autónoma de Madrid, Madrid, Madrid, Spain
Psychological Methods (Impact Factor: 4.45). 07/2006; 11(2):193-206. DOI: 10.1037/1082-989X.11.2.193
Source: PubMed


In meta-analysis, the usual way of assessing whether a set of single studies is homogeneous is by means of the Q test. However, the Q test only informs meta-analysts about the presence versus the absence of heterogeneity, but it does not report on the extent of such heterogeneity. Recently, the I(2) index has been proposed to quantify the degree of heterogeneity in a meta-analysis. In this article, the performances of the Q test and the confidence interval around the I(2) index are compared by means of a Monte Carlo simulation. The results show the utility of the I(2) index as a complement to the Q test, although it has the same problems of power with a small number of studies.

Download full-text


Available from: Fulgencio Marín-Martínez, Jun 06, 2014
  • Source
    • "REML was chosen a priori since the method is highly robust to violations of assumptions that cause other statistical pooling procedures to yield unstable val- ues[22], and heterogeneity was expected due to differences in trial characteristics. Dersimonian and Laird's Q test was used to assess the presence of heterogeneity of effects, and the I 2 index statistic was used to quantify the extent of heterogeneity[23,24]. Subgroup analyses were used to explore whether participant characteristics or trial features influenced effect estimates. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Importance: Acute ischemic stroke is a leading cause of death and disability worldwide. Several recent clinical trials have shown that endovascular treatment improves clinical outcomes among patients with acute ischemic stroke. Objective: To provide an overall and precise estimate of the efficacy of endovascular treatment predominantly using second-generation mechanical thrombectomy devices (stent-retriever devices) compared to medical management on clinical and functional outcomes among patients with acute ischemic stroke. Data sources: MEDLINE, EMBASE, Cochrane Collaboration Central Register of Controlled Clinical Trials, Web of Science, and NIH were searched through November 2015. Study selection: Searches returned 3,045 articles. After removal of duplicates, two authors independently screened titles and abstracts to assess eligibility of 2,495 potentially relevant publications. From these, 38 full-text publications were more closely assessed. Finally, 5 randomized controlled trials of endovascular treatment with predominant use of retrievable stents were selected. Data extraction and synthesis: Three authors independently extracted information on participant and trial characteristics and clinical events using a standardized protocol. Random effects models were used to pool endovascular treatment effects across outcomes. Main outcomes and measures: The primary outcome was better functional outcome as measured on the modified Rankin Scale at 90 days of follow-up. Secondary outcomes included all-cause mortality and symptomatic intra-cerebral hemorrhage. Results: Five trials representing 1,287 patients were included. Overall, patients randomized to endovascular therapy experienced 2.22 times greater odds of better functional outcome compared to those randomized to medical management (95% CI, 1.66 to 2.98; P < 0.0001). Endovascular therapy was not associated with mortality [OR (95% CI), 0.78 (0.54, 1.12); P = 0.1056] or symptomatic intracerebral hemorrhage [OR (95% CI), 1.19 (0.69, 2.05); P = 0.5348]. Meta-regression analysis suggested that shorter times from stroke onset to groin puncture and from stroke onset to reperfusion result in better functional outcomes in ischemic stroke patients (P = 0.0077 and P = 0.0089). There were no significant differences in the beneficial effects of endovascular treatment on functional outcomes across categories of gender, age, stroke severity, ischemic changes on computed tomography, or intravenous tissue plasminogen activator administration. Conclusions and relevance: This meta-analysis demonstrated superior functional outcomes in subjects receiving endovascular treatment compared to medical management. Further, this analysis showed that acute ischemic stroke patients may receive enhanced functional benefit from earlier endovascular treatment.
    Full-text · Article · Jan 2016 · PLoS ONE
  • Source
    • "The test for heterogeneity of the included studies indicated that there was no significant difference between the means of the lipid composition of lamb meat and the null hypothesis was rejected. Because of a relatively low power of the test with a small number of studies (Huedo-Medina et al., 2006), heterogeneity was additionally quantified with the I 2 statistics (Higgins et al., 2003). The I 2 statistics described the proportion of the total study variance that was due to between-study variation and was calculated as "
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this work was to study the intramuscular fatty acid profile of Mediterranean lambs using a meta-analysis. The database was created from five journal articles and one doctoral dissertation, and included lambs of the Talaverana, Comisana, Churra Tensina, Merino Branco and Pramenka breeds. All analyses were performed using Comprehensive Meta-Analysis and R statistical software. According to the overall pooled result, lambs of Mediterranean breeds included in our meta-analysis had 22.85 % of palmitic (C16:0), 14.78 % of stearic (C18:0) and 5.32 % of myristic acid (C14:0). The overall pooled result for oleic acid (C18:1) was 32.53 %, and it was 5.60 % for linoleic acid (C18:2 n-6) and 1.95 % for α-linolenic acid (C18:3 n-3). Regardless of the proportions of the main fatty acids, Mediterranean lambs had 47.85 % of SFA (saturated fatty acid), 37.87 % of MUFA (monounsaturated fatty acid) and 11.50 % of PUFA (polyunsaturated fatty acid). The overall pooled result for the PUFA / SFA ratio was 0.23, and it was 2.22 for the n-6 /n-3 ratio. The results of our meta-analysis enabled us to gain an important scientific insight into the lipid profile of intramuscular fat in Mediterranean lambs. A systematic combination of quantitative data from several carefully selected studies provided estimates of greater statistical power and enabled a more truthful comparison of results originating from different independent studies than any previously published research on this issue.
    Full-text · Article · Jan 2016 · Archiv fur Tierzucht
  • Source
    • "rejected the null hypothesis of homogeneity, and í µí°¼ 2 indicated considerable nonrandom variability (Higgins & Thompson, 2002; Huedo-Medina et al., 2006) of 62.45% (95% CI [59.67, 64.95]). Specifically, an í µí°¼ 2 of 62.45% implies moderate to large nonrandom heterogeneity (Borenstein et al., 2009; Higgins & Thompson, 2002; Huedo-Medina et al., 2006), justifying moderator analyses. The distribution of these effect sizes can be found in the histogram in Figure 1. "
    [Show abstract] [Hide abstract]
    ABSTRACT: A meta-analysis assessed the behavioral impact of and psychological processes associated with presenting words connected to an action or a goal representation. The average and distribution of 352 effect sizes (analyzed using fixed-effects and random-effects models) was obtained from 133 studies (84 reports) in which word primes were incidentally presented to participants, with a non-opposite control group, prior to measuring a behavioral dependent variable. Findings revealed a small behavioral priming effect (𝑑𝐹𝐸 = 0.332, 𝑑𝑅𝐸 = 0.352), which was robust across methodological procedures and only minimally biased by the publication of positive (vs. negative) results. Theory testing analyses indicated that more valued behavior or goal concepts (e.g., associated with important outcomes or values) were associated with stronger priming effects than were less valued behaviors. Furthermore, there was some evidence of persistence of goal effects over time. These results support the notion that goal activation contributes over and above perception-behavior in explaining priming effects. In sum, theorizing about the role of value and satisfaction in goal activation pointed to stronger effects of a behavior or goal concept on overt action. There was no evidence that expectancy (ease of achieving the goal) moderated priming effects.
    Full-text · Article · Aug 2015 · Psychological Bulletin
Show more