The DOI matches its Crossref record; 75 of the 76 references it cites were found in a registry or an index.
Using multiple genetic variants as instrumental variables for modifiable risk factors
Open accessAbstract
Mendelian randomisation analyses use genetic variants as instrumental variables (IVs) to estimate causal effects of modifiable risk factors on disease outcomes. Genetic variants typically explain a small proportion of the variability in risk factors; hence Mendelian randomisation analyses can require large sample sizes. However, an increasing number of genetic variants have been found to be robustly associated with disease-related outcomes in genome-wide association studies. Use of multiple instruments can improve the precision of IV estimates, and also permit examination of underlying IV assumptions. We discuss the use of multiple genetic variants in Mendelian randomisation analyses with continuous outcome variables where all relationships are assumed to be linear. We describe possible violations of IV assumptions, and how multiple instrument analyses can be used to identify them. We present an example using four adiposity-associated genetic variants as IVs for the causal effect of fat mass on bone density, using data on 5509 children enrolled in the ALSPAC birth cohort study. We also use simulation studies to examine the effect of different sets of IVs on precision and bias. When each instrument independently explains variability in the risk factor, use of multiple instruments increases the precision of IV estimates. However, inclusion of weak instruments could increase finite sample bias. Missing data on multiple genetic variants can diminish the available sample size, compared with single instrument analyses. In simulations with additive genotype-risk factor effects, IV estimates using a weighted allele score had similar properties to estimates using multiple instruments. Under the correct conditions, multiple instrument analyses are a promising approach for Mendelian randomisation studies. Further research is required into multiple imputation methods to address missing data issues in IV estimation.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source doesn’t support the claim as written.
In the citing paper
“All three of these SNPs have previously been shown to be strong instruments for MR analysis as measured by F-statistics > 10 [8,18]”
From Associations of genetically determined iron status across the phenome: A mendelian randomization study (2019) · cites this paper as [18]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredIn the citing paper
“Interpretation of results from MR studies relies on four conditions.12,24”
From Mendelian randomization (2022) · cites this paper as [24]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
(2001) · RePEc: Research Papers in Economics · Cited by 27,577
Specification Tests in Econometrics
Hausman, J. A. (1978) · Econometrica · doi:10.2307/1913827 · Cited by 18,820
Statistical Analysis With Missing Data
Lahiff, M., Little, R. J. A., & Rubin, D. B. (1989) · Journal of the American Statistical Association · doi:10.2307/2289883 · Cited by 17,385
Large Sample Properties of Generalized Method of Moments Estimators
Hansen, L. P. (1982) · Econometrica · doi:10.2307/1912775 · Cited by 13,467
Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls
The Wellcome Trust Case Control Consortium, Management Committee, Burton, P. R., Clayton, D. G., Cardon, L. R., Craddock, N., Deloukas, P., Duncanson, A., Kwiatkowski, D. P., McCarthy, M. I., et al. (2007) · Nature · doi:10.1038/nature05911 · Cited by 9,442
Self-citationStatistical Analysis with Missing Data
Little, R. J. A., & Rubin, D. B. (2002) · Wiley Series in Probability and Statistics · doi:10.1002/9781119013563 · Cited by 8,473
Genetics of Food Intake Self-Regulation in Childhood: Literature Review and Research Opportunities
Faith, M. S., Carnell, S., & Kral, T. V. E. (2013) · Human Heredity · doi:10.1159/000353879 · Cited by 7,633
Instrumental Variables Regression with Weak Instruments
Staiger, D., & Stock, J. H. (1997) · Econometrica · doi:10.2307/2171753 · Cited by 7,081
Lippa, N. C., & Sanderson, S. C. (2013) · Human Heredity · doi:10.1159/000353712 · Cited by 6,095
Davey Smith, G., & Ebrahim, S. (2003) · International Journal of Epidemiology · doi:10.1093/ije/dyg070 · Cited by 5,882
Flagged first, then by citations
1 of the 76 references couldn’t be matched to a known paper.
Research works printed but not linked: 3
- Youngman LD, 2000, Circulation, 102, 31
- Pierce BL, 2010, Int J Epidemiol
- Davey Smith G, 2010, Genes Nutr
Cited by
Sorted by citations
Items 801–900 of 1,000
Yan, P., Yao, J., Ke, B., & Fang, X. (2024) · BMC Medical Genomics · doi:10.1186/s12920-024-01950-3 · Cited by 2
“R 2 ) [11].”
Jia, Z., Liu, X., & Liao, W. (2024) · Microbiology Spectrum · doi:10.1128/spectrum.03948-23 · Cited by 2
Liu, Y., Ying, X., Li, Y., Zhu, X., Jing, W., Wang, X., & He, Z. (2024) · Translational Lung Cancer Research · doi:10.21037/tlcr-24-480 · Cited by 2
Yang, Q., Wang, X., Liu, Y., Liu, J., & Zhu, D. (2024) · Frontiers in Endocrinology · doi:10.3389/fendo.2024.1404747 · Cited by 2
Chen, X., Li, A., Zhou, W., & Yao, L. (2024) · Frontiers in Neurology · doi:10.3389/fneur.2024.1408758 · Cited by 2
“Further, SNPs demonstrating significant associations with the exposure variables were evaluated using the F -statistic, identifying strong IVs by the criterion F > 10 (26).”
MethodYu, J., Xie, W., & Wang, P. (2024) · Brain Research · doi:10.1016/j.brainres.2024.149206 · Cited by 2
Ma, X., Chang, L., Li, S., Gu, Y., Wan, J., Sang, H., Ding, L., Liu, M., & He, Q. (2024) · BMC Gastroenterology · doi:10.1186/s12876-024-03383-9 · Cited by 2
“The F statistic was computed using the formula: F = β 2exposure /SE 2exposure [35].”
MethodCai, X., Wang, D., Ding, C., Li, Y., Zheng, J., & Xue, W. (2024) · Frontiers in Oncology · doi:10.3389/fonc.2024.1381803 · Cited by 2
“F-statistic values were estimated to avoid bias from weak genetic instruments (14, 15).”
BackgroundZhao, H., Yang, C., & Xing, F. (2024) · Journal of Alzheimer s Disease · doi:10.3233/jad-240715 · Cited by 2
Sleep Duration and the Risk of Atherosclerosis: A Mendelian Randomization Study
Xu, X., Huang, Y., Liu, J., & Han, X. (2024) · Arquivos Brasileiros de Cardiologia · doi:10.36660/abc.20230813i · Cited by 2
“20 To limit the influence of possible weak IV bias, an F statistic greater than 10 was expected to be of sufficient strength for the main study.”
BackgroundShen, B., Pu, Y., Zheng, X., Liu, Y., Yang, L., Liu, J., & Li, Z. (2024) · Clinical Epigenetics · doi:10.1186/s13148-024-01743-6 · Cited by 2
“In addition, we calculated F value through evaluate the intensity of IVs by utilizing the formula beta 2 /SE 2 with F value > 10 were selected [26, 30] (Tables S5 and S6).”
MethodDuan, H., Arbeev, K. G., Holmes, R., Bagley, O., Wu, D., Akushevich, I. V., Schupf, N., Yashin, A. I., & Ukraintseva, S. V. (2024) · Frontiers in Aging · doi:10.3389/fragi.2024.1442017 · Cited by 2
“The F value can be calculated by the following formula (Palmer et al., 2012): where n is sample size, k denotes given number of SNPs (here is k (cid:1) 1 since we calculated this value for each SNP), and R 2 denotes the coef fi cient of determination, which is the proportion of variation explained…”
MethodZhang, C., Ren, J., Xu, X. Z., Lei, H., Deng, G., Liu, J., Gao, X.-M., Li, J., Wang, X., & Wang, G. (2024) · Frontiers in Nutrition · doi:10.3389/fnut.2024.1433754 · Cited by 2
Jia, C., Yi, D., Ma, M., Xu, Q., Ou, Y., Kong, F., & Jia, Y. (2024) · Frontiers in Molecular Biosciences · doi:10.3389/fmolb.2024.1391419 · Cited by 3
Circulating micronutrients levels and their association with the risk of endometriosis
Zhang, Y., Li, M., Zhang, F., Lin, J., Hong, Y., & Nian, Q. (2024) · Frontiers in Nutrition · doi:10.3389/fnut.2024.1466126 · Cited by 2
“By utilizing genetic variations strongly linked to exposure as instrumental variables, this approach establishes associations between exposure and outcome with greater reliability (26).”
MethodWang, Z., Shu, Q., Wu, J., Cheng, Y., Liang, X., Huang, X., Liu, Y., Tao, Z., Wang, J., Bai, F., et al. (2024) · Cytokine · doi:10.1016/j.cyto.2024.156788 · Cited by 3
Wang, H., Chen, G., Gong, Q., Wu, J., & Chen, P. (2024) · International Immunopharmacology · doi:10.1016/j.intimp.2024.113466 · Cited by 2
Zhou, H., Yang, X., Yang, Q., Cai, Z., Hu, K., & Huang, T. (2024) · Medicine · doi:10.1097/md.0000000000040230 · Cited by 2
Chen, X., Liu, S., Liu, C. X., Huang, Y., Hou, X., Zhuang, J., Luo, Y., Yu, N., Zhuang, J., & Yu, K. (2024) · Cornea · doi:10.1097/ico.0000000000003741 · Cited by 2
“11 Notably, SNPs with an F-statistic greater than 10 were considered robust instruments.”
Liu, X., Jin, J., Wang, B., & Ge, L. (2024) · Frontiers in Endocrinology · doi:10.3389/fendo.2024.1437999 · Cited by 2
“…the strength of each SNP using the F statistic, which is a function of the magnitude and precision of the genetic effect on the trait: F = R2(N2)/(1-R2),where R2 is the proportion of the variance of trait explained by the SNP and N is the sample size of the GWAS of SNPs with the trait (22).”
Jia, G., Guo, T., Liu, L., & He, C. (2024) · Journal of Clinical Hypertension · doi:10.1111/jch.14932 · Cited by 2
Li, J., Wang, W., Qiu, L., Ren, Y., Li, M., Li, W., & Zhang, J. (2024) · Medicine · doi:10.1097/md.0000000000040618 · Cited by 2
“…instrumental variables from biasing the outcome, we evaluated the completed SNPs for screening based on the F statistic, which was calculated as [18] : In the formula, MAF = minor allele frequency, β = effect value, SE = standard error, n = sample size, and k = number of instrumental variables.”
Liu, J., Luo, W., Hu, Z., Zhu, X., & Zhu, L. (2024) · Medicine · doi:10.1097/md.0000000000041106 · Cited by 3
“[10] F statistic < 10 indicating that it is a weak instrumental variable and was excluded.”
Causal Links Between Bone Diseases and Temporomandibular Disorders
Zhang, L., Zhang, Q., & Zhao, D. (2025) · International Dental Journal · doi:10.1016/j.identj.2025.01.010 · Cited by 2
Ding, X., Ren, S., Wen, H.-Z., Zhang, Z., Ye, J.-A., Pan, W.-K., & Ye, J. (2025) · Cancer Control · doi:10.1177/10732748251320842 · Cited by 2
Xiang, J.-C., An, Y., Sun, J., Xu, J., Xiong, Y., Wang, S., & Xia, Q. (2025) · Discover Oncology · doi:10.1007/s12672-025-02021-4 · Cited by 2
“The F-statistics were computed using the following formula [22–24].”
Cao, Y., Li, X., Gao, J., Zhang, N., Zhang, G., & Li, S. (2025) · Depression and Anxiety · doi:10.1155/da/3131579 · Cited by 2
Zhu, Q., Liu, D., Liu, H., Pan, X., Shi, Y., Guo, J., Tong, N., Chen, J., & Zeng, H. (2025) · Discover Oncology · doi:10.1007/s12672-025-02219-6 · Cited by 2
“An F-statistic less than 10 was considered indicative of a weak instrument and was excluded from the following analysis [22].”
Ding, T., Lin, Q., & Qu, X. (2025) · Discover Oncology · doi:10.1007/s12672-025-02434-1 · Cited by 4
“…10000 kb; (3) Exclude palindrome sequences and SNPs with inconsistent alleles between two samples; (4) F statistics: F = R 2 (N − 2)/(1 − R 2), to evaluate the strength of SNPs, The F-statistic value ≥ 10 indicates a lack of strong evidence of weak instrumental bias (supplement: Table S5) [24–26].”
The causal relationship between blood cell traits and aging: a Mendelian randomization study
Chen, P., Zhou, Y.-H., Wei, X. C., Wei, H.-G., Li, B., & Ouyang, L. (2025) · BMC Geriatrics · doi:10.1186/s12877-025-06051-z · Cited by 2
“An F-stat of at least 10 indicated a relatively lower risk of instrument bias in the MR analysis [44].”
Dong, X., Aveyard, P. N., Guess, N. D., Huang, N., Huang, T., & Gao, M. (2025) · Journal of Affective Disorders · doi:10.1016/j.jad.2025.119594 · Cited by 2
Chen, G., Jin, Y., Chu, C., Zheng, Y., Yang, C., Chen, Y., & Zhu, X. (2025) · Scientific Reports · doi:10.1038/s41598-025-05290-9 · Cited by 4
Wang, X., & Zhu, K.-F. (2025) · Medicine · doi:10.1097/md.0000000000042883 · Cited by 2
“[21,22]”
Gut microbiota and breast cancer risk: a Mendelian randomization study stratified by ER status
Yue, J., Gu, L., Chen, S., Wen, F., & Huang, G. (2025) · Discover Oncology · doi:10.1007/s12672-025-03379-1 · Cited by 2
“1 MR analysis design and assumptions for gut microbiota-breast cancer associations by ER status threshold of ≥ 10 was adopted to effectively control for weak instrument bias, thereby ensuring the robustness of subsequent causal inference [12].”
Li, J., Shi, J., Chen, Y., & Guo, Y. (2025) · Medicine · doi:10.1097/md.0000000000043997 · Cited by 2
“[23]”
Causal Effects of Immune Cell Populations on Cognitive Performance: A Mendelian Randomization Study
Fu, J., Yang, M., & Zheng, Q. (2025) · Brain and Behavior · doi:10.1002/brb3.70861 · Cited by 1
“We systematically assessed IV strength by calculating both the proportion of phenotypic variation explained (PVE) and corresponding F -statistic to address potential weak instrument bias (Palmer et al. 2012).”
Zhou, J., Dou, X., Wu, X., Li, Y., Wang, S., & Kang, J. (2025) · Clinical Rheumatology · doi:10.1007/s10067-025-07693-5 · Cited by 2
“…(1) Rel-evance assumption: IVs must be strongly associated with the exposure (RA); (2) Independence assumption: IVs must be independent of confounders; (3) Exclusion restriction assumption: IVs can affect the outcome (OP) only through the exposure (RA) and not through other pathways [26].”
Sun, D., & Ye, Q. (2023) · BMC Pulmonary Medicine · doi:10.1186/s12890-023-02788-8 · Cited by 2
“The F -sta- tistic is considered a measure of instrument strength, and a value greater than 10 indicates a sufficiently strong instrument [34].”
BackgroundBai, X., Zhuo, C., Tong, R., Dao, L., Yu, L., Xu, L., & Li, L. (2023) · Research Square · doi:10.21203/rs.3.rs-3441830/v1 · Cited by 2
Feng, Z., Chen, J., Zhang, Y., Jia, C., Du, L., & Chen, D. (2023) · Research Square · doi:10.21203/rs.3.rs-3719075/v1 · Cited by 2
Le, N. N., Tran, T., McClure, J., Gill, D., & Padmanabhan, S. (2024) · medRxiv · doi:10.1101/2024.08.22.24312458 · Cited by 2
Causal effect of beta-blockers on the risk of lung cancer: a Mendelian randomization study
Lu, Y., Luo, J., Huo, Z., Ge, F., Chen, Y., Chen, Y., Zhang, Q., Li, C., Wang, J., Gan, J., et al. (2023) · Journal of Thoracic Disease · doi:10.21037/jtd-23-1098 · Cited by 1
Du, Y., Lu, Z.-W., Li, K., Wang, Y., Wu, H., Huang, R., Jin, X., Wang, Y., Wang, J., Geng, A., et al. (2024) · Frontiers in Public Health · doi:10.3389/fpubh.2024.1373044 · Cited by 1
“Palindromic SNPs with intermediate allele frequencies and SNPs with F -values 10 were removed (18).”
BackgroundJing, N., Gao, X., Ding, H., Wang, Y., Zhang, Y., Liang, G., & Gao, M. (2024) · Journal of Vascular Surgery Venous and Lymphatic Disorders · doi:10.1016/j.jvsv.2024.101889 · Cited by 1
“31 Finally, potential pleiotropic SNPs were removed using the MR pleiotropy and residual sum and outlier (MR-PRESSO) method.”
MethodZhang, X., Fu, Q.-Q., Cai, Y., Li, X., Chen, L., Jiang, Y., & Chen, Y. (2024) · Human Molecular Genetics · doi:10.1093/hmg/ddae041 · Cited by 1
Liu, H., Wei, X., Wu, W., Zhou, G., & Yuan, J. (2024) · BMC Nephrology · doi:10.1186/s12882-024-03590-2 · Cited by 1
“Generally, an F statistic of ≥ 10 suggests a relatively low risk of bias caused by weak IVs in MR analysis [19].”
BackgroundGenetically predicted asthma and the risk of abnormal spermatozoa
Li, Z., Teng, Z., Han, Z., Zhang, Y., & Wang, Y. (2024) · Frontiers in Genetics · doi:10.3389/fgene.2024.1377770 · Cited by 1
Wu, X., Hu, C., Wu, T., Du, X., Peng, Z., Xue, W., Chen, Y., & Dong, L. (2024) · Human Genomics · doi:10.1186/s40246-024-00622-8 · Cited by 1
“When the F-statistics was 10, we usually assumed that the genetic variation used was a weak IV, which may introduce bias to the results [70].”
MethodEffects of Sjogren's Syndrome on essential hypertension: a two-sample mendelian randomization study
Liu, T., Wang, L., Zheng, H., Lyu, Z., & Wang, B. (2024) · Minerva Cardiology and Angiology · doi:10.23736/s2724-5683.24.06522-0 · Cited by 1
Cao, D., Zhang, S., Zhang, Y., Shao, M., Yang, Q., & Wang, P. (2024) · BMC Cancer · doi:10.1186/s12885-024-12787-5 · Cited by 1
“R^2) [28].”
Li, J., Zhang, Z., Li, F., Liu, Y., Yin, P., Wang, X., Huang, S., Zu, J., Zhang, S., Dong, L., et al. (2024) · European Neurology · doi:10.1159/000541288 · Cited by 2
“Using the formula F = (beta/se) 2 to calculate SNP ’ s F statistic value [28].”
Liu, Y., Huang, M., Sun, Y., & Dai, W. (2024) · Frontiers in Cardiovascular Medicine · doi:10.3389/fcvm.2024.1401384 · Cited by 1
Ma, X., Zhu, P.-P., Yang, Q., Sun, Y., Ou, C., & Li, L. (2024) · Healthcare · doi:10.3390/healthcare12181882 · Cited by 1
“References [73–76] are cited in the Supplementary Materials.”
BackgroundCausal Relationship of Skin Microbiota on Psoriasis: A Mendelian Randomization Study
Chen, Y., Zhuang, Z., & Rao, Z. (2024) · Clinical Cosmetic and Investigational Dermatology · doi:10.2147/ccid.s484366 · Cited by 1
“19 Additionally, we utilized the PhenoScanner V2 database of human genotype–phenotype associations to analyze the potential associations of SNPs with confounding variables.”
MethodChen, C., Yan, S., Wan, B., Yu, Y., Zeng, J., Tan, L., & Lu, J. (2024) · PubMed Central · doi:10.11817/j.issn.1672-7347.2024.240103 · Cited by 1
“MR analyses typically require a threshold of F -statistics>10 to quantify the strength of instrumental variables [19] .”
BackgroundMethodInfluentialChen, P., Ni, S., Liu, Q., & Ouyang, L. (2024) · Frontiers in Nutrition · doi:10.3389/fnut.2024.1434648 · Cited by 1
“An F-stat value of at least 10 suggests a reduced likelihood of instrument bias (IB) in MR analyses (43).”
BackgroundLiu, Z., Zhang, Z., & Xie, P. (2024) · Medicine · doi:10.1097/md.0000000000040220 · Cited by 1
Jin, X., Xu, K., Wu, J., Yang, C., Bao, J., Du, L., Chen, B., Shao, X., & Zhou, C. (2024) · Medicine · doi:10.1097/md.0000000000040180 · Cited by 1
Investigating the causal and genetic relationship between migraine and Parkinson's disease
Deng, M., Zhou, X., Liu, F., Wang, K., Luo, L., Zhang, M.-J., Feng, Q., & Liu, J. (2024) · Headache The Journal of Head and Face Pain · doi:10.1111/head.14881 · Cited by 2
“26,27 For the exclusion restriction assumption, the retained SNPs were clumped for linkage disequilibrium (LD) using European samples from the 1000 Genomes Project as the reference panel, 28 applying the PLINK tool (version 1.9).”
InfluentialSun, Y., & Hao, S. (2024) · International Journal of Molecular Sciences · doi:10.3390/ijms252413485 · Cited by 1
Peng, R., Wang, X., Wang, W., Li, Z., Sun, Y., & Yang, M. (2025) · Hereditas · doi:10.1186/s41065-025-00374-y · Cited by 1
Causal role of plasma liposome in diabetic retinopathy: mendelian randomization (MR) study
Yin, K., Ding, L., Li, X., Zhang, Y., Song, S., Cao, L., Deng, R., Li, Z., Li, Z., Xia, Q., et al. (2025) · Diabetology & Metabolic Syndrome · doi:10.1186/s13098-025-01612-z · Cited by 1
“To this end, weak IVs were detected using the F statistic, calculated as , where R 2 represents the cumulative explained variance, and N denotes the sample size for the selected SNPs [19–21].”
Bidirectional Mendelian Randomization identifies plasma proteins associated with urticaria risk
Zhu, X., & Wu, W. (2025) · Archives of Dermatological Research · doi:10.1007/s00403-025-03927-3 · Cited by 1
“R 2 )] reater than 10 to eliminate weak instrumental variables, thereby minimizing bias associated with weak instruments [16]; and (3) ensure the independence of each IV by reducing linkage disequilibrium (LD).”
Chen, W., Jing, N., Liu, Q., Hong, M., Wang, X., Chen, B., & Wang, Y. (2025) · Medicine · doi:10.1097/md.0000000000041466 · Cited by 1
“[21] Potential pleiotropic SNPs are removed using the Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) method.”
Zhang, L., Chu, Q., Jiang, S., & Shao, B. (2025) · PLoS ONE · doi:10.1371/journal.pone.0309805 · Cited by 3
Niu, M., Pan, J., Wang, K., Zhang, L., Lin, Z., & Zhang, F. (2025) · Vascular Health and Risk Management · doi:10.2147/vhrm.s509753 · Cited by 1
Liang, R.-S., Su, J. Q., Wu, X., Wang, Q., Cai, Y.-M., Su, H.-Y., Tang, J., & Yao, C. (2025) · Scientific Reports · doi:10.1038/s41598-025-94761-0 · Cited by 1
“R 2 ), where “MAF” is the SNP used as the IV secondary allele frequency, “n” is the sample size, “k” is the number of IVs used 16 .”
Huang, J., Zou, W., Lv, Z., Han, H., Huang, J., & Su, H. (2025) · Scientific Reports · doi:10.1038/s41598-025-97429-x · Cited by 1
“The IVW method was chosen as the primary technique for estimating causal effects, as it meta-analyzes individual Wald-type ratios of IVs using a multiplicative random-effects model 26,27 .”
Mitochondrial proteins and congenital birth defect risk: a mendelian randomization study
Li, X., Li, D., Li, Y., Xu, Z., Zhang, Q., Chen, X., Xu, F., & Zhang, R. (2025) · BMC Pregnancy and Childbirth · doi:10.1186/s12884-025-07562-8 · Cited by 1
“The F-sta-tistics were utilized to assess the strength of the IVs, probing whether the effect estimates of the causal associations were likely to be swayed by weak instrument bias [24].”
Physical activity and risk of epilepsy: A 2‐sample Mendelian randomization study
Yang, C., Wang, M., Chen, X., Yao, J., Ma, Y., & Lian, X. (2025) · Epileptic Disorders · doi:10.1002/epd2.70024 · Cited by 1
Assessing the causal effects of type 2 diabetes and obesity-related traits on COVID-19 severity
Seo, J.-E., Kim, G., Park, S., Lee, A., Liang, L., Park, T., & Chung, W. (2025) · Human Genomics · doi:10.1186/s40246-025-00747-4 · Cited by 1
“A threshold of F statistic greater than 10 indicated lower risk of weak instrument bias [31].”
de Ruiter, S. C., Tschiderer, L., Grobbee, D. E., Ruigrok, Y. M., Willeit, P., den Ruijter, H. M., Schmidt, A. F., & Peters, S. A. E. (2025) · Diabetes Obesity and Metabolism · doi:10.1111/dom.16406 · Cited by 1
“27 The number of selected variants for each exposure and per estimate with corresponding F -statistics are provided in Table S4.”
Wang, Q., Zheng, S., Ye, W., Zhu, L., Huang, Y., Wang, Z., Liu, C., Sun, F., Luo, Z., Li, G., et al. (2025) · Journal of Affective Disorders · doi:10.1016/j.jad.2025.119439 · Cited by 1
Li, Y., Yao, Y., & Wu, Y. (2025) · Journal of Alzheimer s Disease · doi:10.1177/13872877251345151 · Cited by 1
“34”
Yang, T., Xiong, T., Wang, L., Luo, L., Mei, H., Yang, J., Yao, X., Xiong, Y., Zhang, X., & Huang, C. (2025) · Medicine · doi:10.1097/md.0000000000042593 · Cited by 1
“[20] To validate the effectiveness of the chosen IVs, the initial analysis utilized the inverse variance weighted (IVW) method, which assesses the overall validity of the instruments while computing weighted average effects based on the P -values of associated SNPs. [21] Additionally, Cochran Q…”
Liu, R., Huang, Y., Jiang, M., Xu, F., Pei, Q., Ma, J., Li, Y., Shen, S., Zhang, B., Guo, X., et al. (2025) · Mechanobiology in Medicine · doi:10.1016/j.mbm.2025.100137 · Cited by 1
“23,24 We dropped SNPs that were missing in the outcome, and 34,844 SNPs that were significantly associated with metabolites were finally retained.”
Chong, Y., Li, X., Peng, Z., Wang, Z., Fu, D., Huang, Z. L., & Chong, T. (2025) · Annals of Human Genetics · doi:10.1111/ahg.70002 · Cited by 1
“…moderate allele frequency were excluded (Özsoy et al. 2023); the exposure and outcome sites were coordinated through the Harmonise function (Ergün et al. 2024); instrumental variables with an F > 10 were calculated and retained (Burgess and Thompson 2011; Palmer et al. 2012; Pierce et al. 2011).”
Wu, C., Tong, S., You, Y., Li, Y., Cai, R., Fang, J., Fang, X., & Zhang, X. (2025) · Journal of Affective Disorders · doi:10.1016/j.jad.2025.119781 · Cited by 1
Smoking can increase the risk of osteoarthritis in European women
Mao, B., Li, H., Zhong, J., Li, X., & Sang, H. (2025) · Scientific Reports · doi:10.1038/s41598-025-09546-2 · Cited by 1
“A value of F 10 was considered to have a weak instrument bias thus removed 29,30 .”
Yang, Z., Xiu, X., Liu, F., Li, J., Tang, Z., Zhong, W., Xu, S., Ma, S., Wang, Y., Zuo, R., et al. (2025) · Medicine · doi:10.1097/md.0000000000043198 · Cited by 1
“[18] At the same time, SNPs with echo structures should be excluded when coordinating exposures and endings using the harmonise_data function.”
Mao, W., & Wu, L. (2025) · Journal of Pain Research · doi:10.2147/jpr.s516682 · Cited by 1
“23 Furthermore, as the genotype of the organism is established before the disease begins and is not influenced by how it developed, undesirable defects caused by potential confounders or reverse causal relationships can be reduced to a minimum.”
Xue, J., Zhou, N., Li, Q., Wang, R., Li, Y., Zhu, H., & Lv, C. (2025) · Shock · doi:10.1097/shk.0000000000002633 · Cited by 1
Sheng, J., Zhang, Y., Dong, L., & Yao, J. (2025) · Medicine · doi:10.1097/md.0000000000042496 · Cited by 1
“20]”
Fatima, T., Överdahl, C., & Maglio, C. (2025) · Journal of Scleroderma and Related Disorders · doi:10.1177/23971983251352341 · Cited by 1
“36”
Dib, M., Meena, D., Yarmolinsky, J., Azzo, J. D., Salman, O., Tavolinejad, H., Gan, S., Beeche, C., Pourmussa, B., Gill, D., et al. (2025) · Journal of the American Heart Association · doi:10.1161/jaha.124.037596 · Cited by 1
“27”
Cao, D., Huang, X., Luo, L., Liu, Y., Wen, J., Zeng, L., Han, J., Li, J., & Li, Y. (2025) · Annals of the New York Academy of Sciences · doi:10.1111/nyas.70019 · Cited by 1
Risk relationship between osteoporosis and plasma proteins
Chen, C., Zeng, Q., Ye, Q., & Jin, F. (2025) · Medicine · doi:10.1097/md.0000000000044105 · Cited by 1
“…using the F test, calculated as , where R 2 represents the extent to which IV explains exposure, N represents the sample size of the GWAS study of exposure factors, and k represents the number of SNPs. [24] Screening SNPs with F > 10 to quantify IV strength and indicate a strong correlation.”
Cheng, J., Dong, X. S., Yang, Y., Qin, X., Zhou, X., Fu, Y., Wang, J., Wang, Y., & Zhang, D. (2025) · BMC Pregnancy and Childbirth · doi:10.1186/s12884-025-07860-1 · Cited by 1
Maimaiti, A., Pan, L., Liu, Y., Huang, M., Ding, W., Yang, W., Sun, K., Wang, Y., & Wang, Z. (2025) · Cerebrovascular Diseases · doi:10.1159/000546508 · Cited by 1
“Moreover, a study by Palmer et al. [65] investigated the genetic determinants of drug response using MR analysis, providing insights into personalized medicine approaches.”
Dissecting the shared genetic architecture between anxiety and cognitive function
Yang, M., Yu, Z., Wei, Z., Wang, J., Wang, N., & Zhang, L. (2025) · Journal of Affective Disorders · doi:10.1016/j.jad.2025.120306 · Cited by 1
Yang, J., Liu, Z., Li, X., Qiu, Y., Liu, Q., Huang, X., & Lian, L. (2025) · Brain and Behavior · doi:10.1002/brb3.70809 · Cited by 1
Liu, S., Zheng, W., Chen, H., Tu, M., Zhong, Y., Lou, Y., Wen, Y., & Chen, L. (2025) · Food Science & Nutrition · doi:10.1002/fsn3.70965 · Cited by 1
“…for univariable analyses was calculated using the formula , N represents the number of participants, EAF represents the effect allele frequency, and β is the estimated effect of the SNP to assess its ability to uniquely predict the outcome (Gill et al. 2019; Levin et al. 2020; Palmer et al. 2012).”
Chen, L., Hu, X., Wu, M., Zeng, Z., & Wang, Y. (2025) · Medicine · doi:10.1097/md.0000000000044619 · Cited by 1
Jin, W., Zhang, Z., Zhang, H., Xiaoli, J., & Wang, J. (2025) · Discover Oncology · doi:10.1007/s12672-025-03590-0 · Cited by 1
“To ensure strong instruments, only SNPs with an F statistic greater than 10 were included to reduce weak instrument bias [28].”
Xiao, Z., Liu, X., Tang, W., Lv, Y., Zhang, T., Zhan, X., Sun, Q., Omindo, W. W., Wang, Q., Zhang, R., et al. (2025) · Human Genomics · doi:10.1186/s40246-025-00829-3 · Cited by 1
“Accordingly, SNPs with F-statistics below 10 were excluded from further analysis to mitigate the risk of weak instrument bias [27].”
Shi, X., Tang, Y., Zhang, Y., Yu, C., Ma, Y., Yan, F., & Liu, T. (2025) · Biomedicines · doi:10.3390/biomedicines13102433 · Cited by 1
“To screen for genetic variants suitable for a MR analysis and ensure the validity of IVs, we applied stringent selection criteria and rigorous quality-control procedures; details are provided in the Supplementary Note S1 [25–29].”
Wang, M., Liu, B., Ren, J., Yuan, Q., Cai, J., Ye, D., Qian, Y., Mao, Y., & Li, Z. (2025) · Clinical Rheumatology · doi:10.1007/s10067-025-07726-z · Cited by 1
“The F -sta- tistic was calculated using the following formulas: where “ R 2 ” was the variance in protein concentration explained by single nucleotide polymorphisms (SNPs),“MAF” was the minor allele frequency, “ k ” was the number of IVs, and “ n ” was the sample size of the GWAS [20, 21].”
Liu, L., Liu, B., Liao, M., Zeng, B., Qin, L., & Li, M. (2025) · Metabolomics · doi:10.1007/s11306-025-02357-6 · Cited by 1
Zhang, D., Li, L., Zhang, Y., Wang, P., Lei, L., Hu, Y., Zhou, C., Yang, S.-S., Liu, W., & Huang, X. (2025) · Journal of Translational Medicine · doi:10.1186/s12967-025-07241-4 · Cited by 1
Educational Attainment as the Primary Socioeconomic Determinant of Heart Failure
Ye, M., Gu, Z., Guo, S., Xu, H., Dong, Y., Zhu, W., & Sun, J. (2025) · JACC Advances · doi:10.1016/j.jacadv.2025.102268 · Cited by 1
Metadata
- Authors
- Palmer, T. M., Lawlor, D. A., Harbord, R. M., Sheehan, N. A., Tobias, J. H., Timpson, N. J., Smith, G. D., & Sterne, J. A.
- Affiliations
- University of Bristol (GB)
- University of Leicester (GB)
- School of Clinical Sciences, University of Bristol, Bristol, UK
- Corresponding author
- Palmer, Tom M
- ORCID
- Venue
- Statistical Methods in Medical Research
- Publisher
- SAGE Publications
- Published
- 2011-01-07
- Volume
- 21
- Issue
- 3
- Pages
- 223-242
- ISSN
- 0962-2802 (ISSN-L)
- 1477-0334
- PMC
- PMC3917707
- OpenAlex
- W2151096860
- Cited by
- 1,289
- References
- 76
- Citations by year
- 2026: 92 · 2025: 240 · 2024: 394 · 2023: 164 · 2022: 83 · 2021: 67 · 2020: 44 · 2019: 33 · 2018: 22 · 2017: 20 · 2016: 27 · 2015: 44 · 2014: 20 · 2013: 19 · 2012: 14
- FWCI
- 11.61
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- Statistical Methods in Medical Research (publishedVersion)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Adelaide Research & Scholarship (AR&S) (University of Adelaide) (submittedVersion)
- Lancaster EPrints (Lancaster University) (submittedVersion)
- Figshare (submittedVersion, cc-by)
- Leicester Research Archive (University of Leicester) (submittedVersion, cc-by)
- Bristol Research (University of Bristol) (submittedVersion, cc-by)
- Funders
- Wellcome Trust (G0600705, 092731)
- University of Bristol (G0600705)
- Medical Research Council (G0600705)
- Economic and Social Research Council (ES/E00234X/1)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic Associations and Epidemiology
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetics and Physical Performance
- Keywords
- Mendelian randomization, modifiable risk factors, weak instruments, finite-sample bias, fat mass, bone mineral density, children, causal effect estimation, continuous outcomes
- MeSH terms
- Female, Genotype, Humans, Male, Risk Factors, Genetic Variation, Mendelian Randomization Analysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
