{"id":4128,"date":"2023-03-14T07:44:17","date_gmt":"2023-03-14T07:44:17","guid":{"rendered":"https:\/\/www.editage.com\/insights\/best-practices-in-reporting-survival-analyses-a-guide-for-biomedical-researchers\/"},"modified":"2025-01-15T06:21:08","modified_gmt":"2025-01-15T06:21:08","slug":"best-practices-in-reporting-survival-analyses-a-guide-for-biomedical-researchers","status":"publish","type":"post","link":"https:\/\/www.editage.com\/insights\/best-practices-in-reporting-survival-analyses-a-guide-for-biomedical-researchers","title":{"rendered":"Best practices in reporting survival analyses: A guide for biomedical researchers"},"content":{"rendered":"

As scientists, we have many statistical tools at our disposal. Similar to how a regression analysis helps us understand relations between an independent variable and its covariates, survival analysis allows us to analyze time-to-event <\/a>data within or between groups. This allows us to plot a survival curve within a group over a given period, understand how effective an intervention is by comparing groups, and identify factors that may affect time-to-event.\u00a0\u00a0<\/p>\n

What is survival analysis?\u00a0<\/strong><\/p>\n

A survival analysis allows us to understand time-to-event data for events such as death, system failure, or recovery from a disease. Survival analyses are useful for modeling the probability of an event occurring over time and identifying factors affecting its occurrence. Survival analyses are vital in medical research, engineering, and many sciences when we need to understand the survival and failure rates of different groups or populations.\u00a0<\/p>\n

Uses of survival analysis\u00a0<\/strong><\/p>\n

By conducting survival analyses, we can identify factors that are associated with increased or decreased survival rates. Survival analysis also enables us to compare how different treatments and interventions influence survival outcomes, thereby measuring their effectiveness. This information is critical for developing new treatments and improving patient outcomes.\u00a0<\/p>\n

How do I conduct a survival analysis?\u00a0<\/strong><\/p>\n

Suppose we are investigating a drug that apparently reduces heart attack risk. We can use survival analysis to assess how well it works in reducing mortality. We would select a study cohort, divide them randomly into blinded treatment and non-treatment groups at the beginning of the study, collect their data, and analyze differences in mortality between groups. This will produce two survival curves that let us visualize the differences between these groups and provide compelling evidence for drug effectiveness.\u00a0<\/p>\n

Here is a brief step-by-step overview of the process:\u00a0<\/p>\n

    \n
  1. \n

    Define a research question, including the outcome of interest and the covariates that will be included in the analysis.\u00a0<\/p>\n<\/li>\n<\/ol>\n

      \n
    1. \n

      Collect data according to defined criteria.\u00a0<\/p>\n<\/li>\n<\/ol>\n

        \n
      1. \n

        Consider what comparisons are being made and select an appropriate statistical method<\/a> for the data, such as Kaplan\u2013Meier.\u00a0<\/p>\n<\/li>\n<\/ol>\n

          \n
        1. \n

          Check assumptions and fit, for example, by plotting the Schoenfeld residuals if you are using a Cox\u2019s proportional hazards model.\u00a0<\/p>\n<\/li>\n<\/ol>\n

            \n
          1. \n

            Provide hazard ratios<\/a> for each explanatory variable, with a confidence interval to help understand risk, particularly for censored data.\u00a0<\/p>\n<\/li>\n<\/ol>\n

              \n
            1. \n

              Provide clearly readable plots to visualize your data.\u00a0<\/p>\n<\/li>\n<\/ol>\n

                \n
              1. \n

                Ensure that inferences made from these data are summarized in your discussion.\u00a0<\/p>\n<\/li>\n<\/ol>\n

                Best practices in conducting survival analysis\u00a0<\/strong><\/p>\n