Limitations and Biases in Cohort Studies

Abstract

Good practice in research involves considering diverse sources of biases when designing a study for later validation of results. If they are recognized beforehand, it is possible to minimize or avoid them. Selection biases may originate at the time of enrolling the subjects of study, making it necessary to clearly state the selection criteria of the exposed and nonexposed individuals. If people get lost from the original sample, bias may be introduced by the consequences of reducing the sample. Biases of information could originate in loss of evidence at the moment of recording the data. The definition of follow-up protocols may also help to keep registers of all variables, so information will not be missed from the individuals under study or from the observers who conduct the follow-up. It is necessary to apply the same protocols and instruments for measuring and evaluating the health outcomes in exposed and nonexposed individuals in order to avoid biases of missclassification. Confusion biases can be avoided at the time of designing the study, with the inclusion of confounding variables from the onset. Matching by age and gender is strongly recommended, and finally, adjustment techniques are used at the time of the data analysis.

High body mass index and sleep problems during pregnancy: A meta‐analysis and meta‐regression of observational studies

“…This result is probably because the follow‐up duration was not sufficient, and inadequate follow‐up of cohort was found in low‐quality cohort studies in this review (Fung et al, 2013; Miyagawa et al, 2011; Telerant et al, 2018). Biases of information could originate from inadequate follow‐up in exposed and non‐exposed women, which may induce underestimation of the association between BMI and sleep problems (Ramirez‐Santana, 2018). Long‐term follow‐up to identify sleep problems during pregnancy is warranted in order to avoid biases of misclassification (Ramirez‐Santana, 2018).…”

“…Biases of information could originate from inadequate follow‐up in exposed and non‐exposed women, which may induce underestimation of the association between BMI and sleep problems (Ramirez‐Santana, 2018). Long‐term follow‐up to identify sleep problems during pregnancy is warranted in order to avoid biases of misclassification (Ramirez‐Santana, 2018).…”

Multiple Micronutrient Supplementation during Pregnancy and Increased Birth Weight and Skinfold Thicknesses in the Offspring: The Cambridge Baby Growth Study

“…In this analysis, missing data were dealt with by case or listwise deletions in order to prevent the introduction of inaccuracies resulting from data imputation. However, this may have introduced biases which are unfortunately common in this type of cohort study [ 31 ].…”

Cardiovascular disease subtypes, physical disability and workforce participation: A cross-sectional study of 163,562 middle-aged Australians

“…Our study population was randomly sampled from a whole-of-population database, and included ~ 10% of the entire population in the target age-group and the response rate was ~18%, consistent with cohort studies of this nature. Generally, participants in cohort studies are healthier than the general population [ 44 ]. Though we could not find a comparable age-group similar ours for prevalence of CVD in Australia, the workforce participation rate in our study was 9% higher (74.6% vs 65.2%), than that reported for Australia for the same age group during the same period (2007–08) by the Australian Bureau of Statistics [ 45 , 46 ].…”