Such studies must use study designs and analytical approaches that consider the biases and issues that were discussed in detail above and summarized in Supplementary Table 2.
However, the spectacular effects reported in these studies are compatible with time-related biases.
Long-term metformin use is as- sociated with decreased risk of breast can- cer. It does not guar- antee the same duration of treated diabetes in cases and controls that would avoid time-window bias. A Cox proportional hazards model was used to estimate the hazard ratio HRadjusted for age, sex, insulin use, and a chronic disease score.
With respect to metformin, the scientific evidence of its potential beneficial effects on cancer would need to be reassessed critically before embarking on further long and expensive trials.
Time-related biases are common in observational studies of drug effects and have been described extensively in several therapeutic areas. Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered.
However, given our focus on the methodological aspects of the literature on this topic and the large number of included studies with null results, the effect of publication bias on the current study was likely minimal. Diabetes Care ;35 Accepted November 10, Design Case—control study nested within a BO cohort.
As a consequence, metformin received much attention as a potential anticancer agent; these observational studies sparked the conduction of randomized, controlled trials of metformin as cancer treatment.
Note the ORs were all above 1. The adjusted HR of cancer mortality for sulfonylurea versus metformin was 1. Immortal time bias can arise when the period between cohort entry and date of first exposure to a drug, during which the event of interest has not occurred, is either misclassified or simply excluded and not accounted for in the analysis.
Patients were excluded if they had a history of oesophageal cancer anytime before BO diagnosis and if they had a history of gastric cancer within 6 months after BO diagnosis. This article has been cited by other articles in PMC. These were matched to 76 control subjects, with a median of 5 controls per case IQR 4—6.
This article has been cited by other articles in PMC.
Metformin is a drug of choice for the management of type 2 diabetes mellitus 34. Acknowledgments — No potential conflicts of interest relevant to this article were reported. Immortal time bias in Pharmacoepidemiology Am J Epidemiol ; The 3, subjects who entered the cohort while using sulfonylureas and remained on these throughout the entire follow-up and the 1, who used only metformin from day 1 throughout follow-up pose no problem to the definition of exposure.
However, residual confounding may have resulted due to the presence of unmeasured confounders. This study generated great interest in metformin as an agent in the prevention and treatment of cancer, and many preclinical studies showed that metformin can inhibit the growth of cancer cells in vitro and in vivo 78.
Time-related bias Time-related bias in observational studies can produce illusory results in favour of the treatment group and may affect both cohort and case-control studies, mostly database studies.
Patients entered the cohort at the time of their first OHA prescription during — However, the follow-up started from the first insulin prescription, and as a result, the time before the last insulin prescription was misclassified as exposed when it should have been classified as unexposed.
Time-related biases include immortal time bias, a bias introduced with time-fixed cohort analyses that misclassify unexposed time as exposed; time-window bias, a bias introduced because of differential exposure opportunity time windows between subjects; and time-lag bias, a bias introduced by comparing treatments given at different stages of the disease 12.
Fourth, this systematic review focused on the association between long-acting insulin analogs and cancer incidence. Unlike immortal time bias, it is not possible, a priori, to predict the direction of this bias on the point estimates since it directly depends on the differential durations for exposure assessment in the cases and controls.
In the example of case control studies assessing such as the use of inhaled corticosteroids and death in chronic obstructive pulmonary disease patientsno clearly valid approach to data analysis can fully circumvent this bias.
It is defined in the book Modern Epidemiology K. Finally, as is true with any systematic review, there is the potential for publication bias.
In our systematic review, we also identified the lack of lag periods used and the inclusion of prevalent users as additional limitations in a few studies. Diabetes Care 2; Moreover, this immortal person-time should be classified as sulfonylurea-exposed until the start of metformin, at which point the remaining person-time can be classified as metformin-exposed.
Importantly, unlike the previous review, the present methodological assessment was conducted in the context of a systematic review.
A case-control approach that properly accounts for time produces a rate ratio of (–)—suggesting no benefit of statins on lung cancer risk. We show analytically that the magnitude of the bias is proportional to the ratio of the unequal time-window lengths. Time-window Bias in Case-control Studies.
Statins and Lung Cancer (Epidemiology ; 22 (2)) describes a case-control study which reported a 45% reduction in the rate of lung cancer with any statin use. It does not guarantee the same duration of treated diabetes in cases and controls that would avoid time-window bias.
Suissa S, Dell’aniello S, Vahey S, Renoux C. Time-window bias in case-control studies: statins and lung cancer. Epidemiology ; – 5. Time-window Bias in Case-control Studies. Statins and Lung Cancer (Epidemiology ; 22 (2)) describes a case-control study which reported a 45% reduction. The publications were classified according to the potential bias induced by the study design, including immortal time bias, immeasurable time bias, selection bias, and confounding (all described in the following text in the context of selected studies).
Objectives Time-window bias was described in case–control studies and led to a biased estimate of drug effect. No studies have measured the impact of this bias on the assessment of the effect of medication adherence on health outcomes.Time-window bias in case-control studies statins and lung cancer