Concept: Multiple comparisons
- Proceedings of the National Academy of Sciences of the United States of America
- Published about 2 years ago
The most widely used task functional magnetic resonance imaging (fMRI) analyses use parametric statistical methods that depend on a variety of assumptions. In this work, we use real resting-state data and a total of 3 million random task group analyses to compute empirical familywise error rates for the fMRI software packages SPM, FSL, and AFNI, as well as a nonparametric permutation method. For a nominal familywise error rate of 5%, the parametric statistical methods are shown to be conservative for voxelwise inference and invalid for clusterwise inference. Our results suggest that the principal cause of the invalid cluster inferences is spatial autocorrelation functions that do not follow the assumed Gaussian shape. By comparison, the nonparametric permutation test is found to produce nominal results for voxelwise as well as clusterwise inference. These findings speak to the need of validating the statistical methods being used in the field of neuroimaging.
Over the past ten years, unconventional gas and oil drilling (UGOD) has markedly expanded in the United States. Despite substantial increases in well drilling, the health consequences of UGOD toxicant exposure remain unclear. This study examines an association between wells and healthcare use by zip code from 2007 to 2011 in Pennsylvania. Inpatient discharge databases from the Pennsylvania Healthcare Cost Containment Council were correlated with active wells by zip code in three counties in Pennsylvania. For overall inpatient prevalence rates and 25 specific medical categories, the association of inpatient prevalence rates with number of wells per zip code and, separately, with wells per km2 (separated into quantiles and defined as well density) were estimated using fixed-effects Poisson models. To account for multiple comparisons, a Bonferroni correction with associations of p<0.00096 was considered statistically significant. Cardiology inpatient prevalence rates were significantly associated with number of wells per zip code (p<0.00096) and wells per km2 (p<0.00096) while neurology inpatient prevalence rates were significantly associated with wells per km2 (p<0.00096). Furthermore, evidence also supported an association between well density and inpatient prevalence rates for the medical categories of dermatology, neurology, oncology, and urology. These data suggest that UGOD wells, which dramatically increased in the past decade, were associated with increased inpatient prevalence rates within specific medical categories in Pennsylvania. Further studies are necessary to address healthcare costs of UGOD and determine whether specific toxicants or combinations are associated with organ-specific responses.
The objective of this study was to identify blood-based protein biomarkers of early stage Mycobacterium tuberculosis (Mtb) infection. We utilized plasma and serum specimens from TB patients and their contacts (age≥12) enrolled in a household contact study in Uganda. In the discovery phase cross-sectional samples from 104 HIV-uninfected persons classified as either active TB, latent Mtb infection (LTBI), tuberculin skin test (TST) converters, or persistent TST-negative were analyzed. Two hundred eighty-nine statistically significant (false discovery rate corrected p<0.05) differentially expressed proteins were identified across all comparisons. Proteins associated with cellular immunity and lipid metabolism were induced early after Mtb infection. One hundred and fifty-nine proteins were selected for a targeted mass spectrometry assay. A set of longitudinal samples from 52 TST-negative subjects who converted to TST-positive or remained TST-negative were analyzed, and multivariate logistic regression was used to identify unique protein panels able to predict TST conversion with cross-validated AUC>0.85. Panel performance was confirmed with an independent validation set of longitudinal samples from 16 subjects. These candidate protein biomarkers may allow for the identification of recently Mtb infected individuals at highest risk for developing active TB and most likely to benefit from preventive therapy.
In a consortium including 23 637 breast cancer patients and 25 579 controls of East Asian ancestry, we investigated 70 single-nucleotide polymorphisms (SNPs) in 67 independent breast cancer susceptibility loci recently identified by genome-wide association studies (GWASs) conducted primarily in European-ancestry populations. SNPs in 31 loci showed an association with breast cancer risk at P < 0.05 in a direction consistent with that reported previously. Twenty-one of them remained statistically significant after adjusting for multiple comparisons with the Bonferroni-corrected significance level of <0.0015. Eight of the 70 SNPs showed a significantly different association with breast cancer risk by estrogen receptor (ER) status at P < 0.05. With the exception of rs2046210 at 6q25.1, the seven other SNPs showed a stronger association with ER-positive than ER-negative cancer. This study replicated all five genetic risk variants initially identified in Asians and provided evidence for associations of breast cancer risk in the East Asian population with nearly half of the genetic risk variants initially reported in GWASs conducted in European descendants. Taken together, these common genetic risk variants explain ∼10% of excess familial risk of breast cancer in Asian populations.
Various dietary supplements are claimed to have cutaneous anti-aging properties; however, there are a limited number of research studies supporting these claims. The objective of this research was to study the effectiveness of collagen hydrolysate (CH) composed of specific collagen peptides on skin biophysical parameters related to cutaneous aging. In this double-blind, placebo-controlled trial, 69 women aged 35-55 years were randomized to receive 2.5 g or 5.0 g of CH or placebo once daily for 8 weeks, with 23 subjects being allocated to each treatment group. Skin elasticity, skin moisture, transepidermal water loss and skin roughness were objectively measured before the first oral product application (t0) and after 4 (t1) and 8 weeks (t2) of regular intake. Skin elasticity (primary interest) was also assessed at follow-up 4 weeks after the last intake of CH (t3, 4-week regression phase). At the end of the study, skin elasticity in both CH dosage groups showed a statistically significant improvement in comparison to placebo. After 4 weeks of follow-up treatment, a statistically significantly higher skin elasticity level was determined in elderly women. With regard to skin moisture and skin evaporation, a positive influence of CH treatment could be observed in a subgroup analysis, but data failed to reach a level of statistical significance. No side effects were noted throughout the study. © 2013 S. Karger AG, Basel.
OBJECTIVE: The aim of this study was to systematically review the current literature on the clinical effects of sugar-free chewing gum on plaque indices and parameters of gingival inflammation. MATERIAL AND METHODS: The MEDLINE-PubMed, Cochrane-CENTRAL and EMBASE databases were searched up to 20 April 2012 to identify any appropriate studies. Plaque indices and parameters of gingival inflammation were selected as outcome variables. RESULTS: An independent screening of the 594 unique titles and abstracts identified six non-brushing and four brushing studies that met the eligibility criteria. In the non-brushing studies, the use of chewing gum did not significantly affect the parameters of interest. In the descriptive analysis of the brushing studies, four of five comparisons showed a statistically significant effect in favour of the sugar-free chewing gum with respect to plaque scores. The meta-analysis for the Quigley & Hein (J Am Dent Assoc 1962; 65: 26) plaque index scores in the brushing studies also showed a significant difference (DiffM -0.24, 95% CI [-0.41; -0.08]). For bleeding tendency, the descriptive analysis showed that one of the two comparisons identified a significant difference in favour of chewing gum. The meta-analysis, however, did not substantiate this difference. CONCLUSION: Within the limitations of this systematic review, it may be concluded that the use of sugar-free chewing gum as an adjunct to toothbrushing provides a small but significant reduction in plaque scores. Chewing sugar-free gum showed no significant effect on gingivitis scores. In the absence of brushing, no effect on plaque and gingivitis scores could be established.
Season of birth and hand preference were examined in a sample of 42 (7 males, 35 females) individuals who were identified as schizotypic based on their scores on selected scales of the Chapman Psychosis Proneness Scales (CPPS) and a matched comparison sample of 42 individuals with non-deviant CPPS scores. Presence or absence of schizotypy was analyzed using chi square tests of independence with the presence or absence of each risk factor serving as the independent variable. Further analyses incorporated independent means t tests to examine mean scores on the CPPS with the presence or absence of each risk factor again serving as the independent variable. Results supported the hypothesis that winter/early-spring birth would be associated with psychometric schizotypy, although the results for mixed-handedness fell just short of statistical significance. However, mixed hand preference was associated with higher scores on MagId and PerAb scales of the CPPS, but not the RSA scale, which suggests that mixed laterality is associated with the more cognitive-perceptual aspects of schizotypy. Results are discussed in relation to previous literature and their relevance to the prediction of schizophrenia-related psychosis.
Growing interest in personalised medicine and targeted therapies is leading to an increase in the importance of subgroup analyses. If it is planned to view treatment comparisons in both a predefined subgroup and the full population as co-primary analyses, it is important that the statistical analysis controls the familywise type I error rate. Spiessens and Debois (Cont. Clin. Trials, 2010, 31, 647-656) recently proposed an approach specific for this setting, which incorporates an assumption about the correlation based on the known sizes of the different groups, and showed that this is more powerful than generic multiple comparisons procedures such as the Bonferroni correction. If recruitment is slow relative to the length of time taken to observe the outcome, it may be efficient to conduct an interim analysis. In this paper, we propose a new method for an adaptive clinical trial with co-primary analyses in a predefined subgroup and the full population based on the conditional error function principle. The methodology is generic in that we assume test statistics can be taken to be normally distributed rather than making any specific distributional assumptions about individual patient data. In a simulation study, we demonstrate that the new method is more powerful than previously suggested analysis strategies. Furthermore, we show how the method can be extended to situations when the selection is not based on the final but on an early outcome. We use a case study in a targeted therapy in oncology to illustrate the use of the proposed methodology with non-normal outcomes. Copyright © 2012 John Wiley & Sons, Ltd.
PURPOSE: To analyze quality of life (QOL) in a randomized, placebo-controlled phase III trial concluding that the addition of concurrent and maintenance bevacizumab (Arm 3) to carboplatin and paclitaxel prolongs progression-free survival in front-line treatment of advanced ovarian cancer compared to chemotherapy alone (Arm 1) or chemotherapy with bevacizumab in cycles 2-6 only (Arm 2). PATIENTS AND METHODS: The Trial Outcome Index of the Functional Assessment of Cancer Therapy-Ovary (FACT-O TOI) was used to assess QOL before cycles 1, 4, 7, 13, and 21; and 6 months after completing study therapy. Differences in QOL scores were assessed using a linear mixed model, adjusting for baseline score, and age. The significance level was set at 0.0167 to account for multiple comparisons. RESULTS: 1693 patients were queried. Arm 2 (p<0.001) and Arm 3 (p<0.001) reported lower QOL scores than those in Arm 1. The treatment differences were observed mainly at cycle 4, when the patients receiving bevacizumab (Arm 2 and Arm 3) reported 2.72 points (98.3% CI: 0.88 ~ 4.57; effect size=0.18) and 2.96 points (98.3% CI: 1.13~4.78; effect size=0.20) lower QOL respectively, than those in Arm 1. The difference in QOL scores between Arm 1 and Arm 3 remained statistically significant up to cycle 7. The percentage of patients who reported abdominal discomfort dropped over time, without significant differences among study arms. CONCLUSION: The small QOL difference observed during chemotherapy did not persist during maintenance bevacizumab.
OBJECTIVE: To understand the nature and long-term fate of supranormal differential renal function (snDRF), analysis of a post-pyeloplasty database was performed. MATERIALS AND METHODS: Preoperative diethylenetriaminepentaacetic acid (DTPA) diuretic renal scans (DRS) identified 33 children with snDRF (>55%). Postoperatively, all showed improved hydronephrosis in ultrasound imaging or normalized drainage pattern in DRS during more than 5 years of follow-up. The fate of snDRF was described, and the nature of snDRF was speculated by associating it with clinical and ultrasound-related variables. RESULTS: Median age at operation was 22 months. Preoperative hydronephrosis was grade III in 27 patients (82%) and grade IV in 6 (18%). Hydronephrosis was normalized or improved, showing relief of obstruction. Overall fate of snDRF revealed significant reduction of DRF in 23 patients (70%), of whom 18 (81%) showed normalization during 5 years of follow-up. Stationary change of snDRF was seen in one-third of patients. Comparison of clinical and ultrasound-related variables revealed lower age, lower renal parenchyma thickness (RPT) increase, and higher preoperative DRF in the normalized group compared with the other group. The normalized group also showed a higher number in Society of Fetal Urology grade 4 hydronephrosis and lower RPT, although they reached borderline statistical significance. CONCLUSION: Long-term follow-up revealed that 70% of snDRF significantly decreased and normalized in most patients. The normalized group was younger and showed more hydronephrotic changes. Hydronephrotic change is suggested as a possible cause of snDRF.