 |
| comparing the survival time between placebo treatment versus treatment |
Suppose you are conducting clinical study for a novel drug or being participated in such trials, you want to know whether new drug is doing better or not. since it is novel drug, there won't be enough data of previous study, you have to see the outcomes of events from the time of enrollment till date in different groups of people getting treatment. Hazard Ratio interpretation provides the backbone of predictions you make to answer this question.
Hazard ratio is the outcome measure used in time to event analysis (survival analysis) which is also called as instantaneous event rate. The term hazard means any bad event in literal sense but for study point of view it can be any end point or event or risk of an outcome after intervention being introduced to the participants.
- It is the ratio between hazard in the treatment arm to hazard in the controlled arm.
- The term Hazard means instantaneous event rates, which in simple term is the probability of a particular event in any given time.
- At the same time, HR gives the rate of progression of disease since it can be plotted against time.
Formula for calculating Hazard Ratio
HR = The ratio between hazard of particular event in intervention group to the hazard of event in controlled group.
Hazard in treatment arm
Hazard in control arm
Bio-statistician calculates HR, but researchers/ Investigators should be able to interpret it accordingly. Here, in this blog we will be discussing about the inferences that can be made based upon the calculated value.
Interpretation of Hazard ratio
HR =1: The probability of having an outcome is same for both treatment arms, i.e. There is lack of associations between the intervention and outcome.
HR =2: The probability of event in treatment arm is twice than that in controlled arm
twice as many patients in treatment arm are having the event proportionally to comparator group
HR = 0.5: At an time half as many patients in the treatment group are having an event proportionally to the comparator group
Why Hazard Ratio is needed instead of Relative Risk ?
- Technically these two measures are different since Hazard Ratio can be plotted for events against time whereas relative risk defines the probability of an outcome in two different group regardless of timings.
- Hazard ratio is used to interpret the result of intervention in clinical trials as survival analysis whereas Risk Ratio commonly used in observational studies.
- This plot of time against an event throughout the study period is called kaplan mier curve. The probability can be calculated from this curve.
Here, since there is death event is happening on each event, survival probability is continuously declining along with the follow ups - Those participants who do not reach the study end-point and lost follow-ups are also included in analysis. Power as well as validity of study can be maintained by analyzing number of survivors at multiples point of time.
Here, in visits 2 to 5,survival probability is constant, since there are no events happen or loss of follow ups - While Interpreting Hazard ratio, Time based parameters like Median time are taken into considerations if possible to know the magnitude of benefit of treatment to the patients.
Can you predict the positive outcome based upon findings of Hazard Ratio
Not exactly. Because Hazard ratio only gives the idea about outcome or event, which can be varied from study to study. For example, if study is ongoing for cardiovascular pathology and end point is determined as stroke or death, then High value of hazard ratio can show study drug not being effective whereas in cancer research if end point is disease free survival, higher the HR value, study drug will be considered better.
References
No comments:
Post a Comment