China
Africa
Explore chapters and articles related to this topic
Cancer Epidemiology
Published in Trevor F. Cox, Medical Statistics for Cancer Studies, 2022
Cohort studies: Advantages: as the main study outcomes do not influence the selection of study individuals, incidence, risk and relative risk can be estimated; can assess exposures that are rare. Disadvantages: possibly long follow-up time and high cost for prospective cohort studies, which is not such a problem for historical studies; no control over variables in historical cohort studies; no randomisation and so imbalances can occur; possible selection bias; rare cancers would need a very large cohort
Study Limitations to Consider
Published in Lisa Chasan-Taber, Writing Grant Proposals in Epidemiology, Preventive Medicine, and Biostatistics, 2022
Prospective cohort studies are less prone to selection bias because they enroll participants who do not have the disease of interest and follow them for disease incidence. Therefore, the disease (outcome) of interest is unknown at the beginning of the study (baseline) and should not influence selection of participants into exposed and unexposed groups.
Infectious Disease Data from Surveillance, Outbreak Investigation, and Epidemiological Studies
Published in Leonhard Held, Niel Hens, Philip O’Neill, Jacco Wallinga, Handbook of Infectious Disease Data Analysis, 2019
Bias in data from cohort studies can occur in several ways. First, when people who are lost to follow-up differ in terms of determinants or characteristics from those who remain in the study, the resulting data is not representative of the initial cohort. This outcome can lead to a biased assessment of the effects of a certain determinant, when the loss to follow-up is related to both to the determinant and the outcome of interest. A second cause of bias can result from unequal assessment of health outcome status between exposed and unexposed individuals, or, vice versa, from unequal assessment of the disease status between exposed and non-exposed individuals. Another important source of bias occurs when the exposure status of study participants is dependent on factors which also are related to the health outcome of interest (confounding by indication, e.g., frailty bias in influenza vaccine studies). Other sources of bias, e.g. those discussed in 3.2.3, can also occur in cohort studies.
ADM-assisted prepectoral breast reconstruction is not associated with high complication rate as before: a Meta-analysis
Published in Journal of Plastic Surgery and Hand Surgery, 2023
Jiaheng Xie, Ming Wang, Yuan Cao, Zhechen Zhu, Shujie Ruan, Mengmeng Ou, Pan Yu, Jingping Shi
Among the articles we selected, the publication time ranged from 2018 to 2020. Two of the studies were prospective cohort studies and the rest were retrospective analyses. The total study span ranged from 2009 to 2019. Given the presence of bilateral breast reconstruction in some studies, we put the number of breasts in parentheses after the number of patients. Significantly that the complication rates are all based on the number of breasts, not the number of patients. Two of the articles used ADM combined with Vicryl. And the rest applied ADM only. The ADM and Vicryl used in these articles were in the form of mesh. Eleven articles used direct implantation (one-stage breast reconstruction), 12 articles used two-stage breast reconstruction related to the expander, and five articles used both methods (Table 3).
Loss to follow-up after direct-to-implant breast reconstruction
Published in Journal of Plastic Surgery and Hand Surgery, 2023
Eun Key Kim, Soo Hyun Woo, Do Yeon Kim, Eun Jeong Choi, Kyunghyun Min, Taik Jong Lee, Jin Sup Eom, Hyun Ho Han
Selection bias due to loss to follow-up is inevitable in most cohort studies, and its effect has been investigated in the field of epidemiology. Dettori suggested that a follow-up loss of less than 5% leads to little bias, whereas a loss of more than 20% poses a serious threat to validity; however, generally, the recommended follow-up threshold is around 60–80% [11,21]. Missing data were categorized into missing completely at random (MCAR; the probability that a subject remains in the study does not depend on the exposure, confounders, or outcome) or missing at random (MAR; the probability of a subject remaining in the study depends on the exposure or confounders but not outcome), and missing not at random (MNAR; the probability of being lost to follow-up depends on the outcomes to be measured and cannot be completely explained by the covariates) mechanisms. Kristman et al. found no notable bias when loss to follow-up was related to MCAR or MAR mechanisms. However, they found seriously biased estimates with even low levels of loss to follow-up when observations were lost to follow-up based on the MNAR mechanism [21]. Little’s test of MCAR is known to be useful for testing the missing mechanism [22]. Our results appeared to be related to the MNAR mechanism.
Association between nut consumption and cancer risk: a meta-analysis
Published in Nutrition and Cancer, 2022
Chang Cao, Xinyan Gan, Yan He, Shiqi Nong, Yonglin Su, Zheran Liu, Yu Zhang, Xiaolin Hu, Xingchen Peng
All prospective cohort studies concerning the relationship between nuts intake and cancer risk or mortality were assessed for eligibility. Candidate studies were included if they met the following criteria: 1) prospective cohort studies or case-cohort studies; 2) considered intake of total nuts (including peanuts and tree nuts), tree nuts (including almonds, Brazil nuts, cashews, hazelnuts, macadamia, pecans, pistachios, pine nuts, and walnuts), peanuts and peanut butter as exposure; 3) considered the risk of cancer or mortality as outcomes; 4) reported estimate of hazard ratio (HR) or risk ratio (RR) with the corresponding 95% CIs. If the same cases from the same cohort were reported in more than one study, only the most recent study or the study reporting the most cases was included. If articles included the cases from the same cohort but assessed different exposure or outcomes (e.g., different cancers), they were included in the meta-analysis and dose-response analysis.