Satyajit Das

Bias is the difference between the measured values against the actual value. Bias can be positive or negative. If the difference between the measured values and the actual value is positive then you have positive bias. If the difference between the measured values and the actual values is negative then you have negative bias.
 
Three types of bias can be distinguished: information bias, selection bias, and confounding
 

 
 
Reporting bias means that only a selection of results are included in any analysis, which typically covers only a fraction of relevant evidence. This can lead to inappropriate decisions (for example, prescribing ineffective or harmful drugs), resource waste and misguided future research.
 
Types of Reporting Bias
1.      Citation bias: basing your analysis on studies that you find in the citations of other studies.
2.      Language bias: ignoring studies not published in your native Language.
3.      Location bias: certain reports or studies are harder to find than others. For example, studies that are published in journals might be indexed higher in databases.
4.      Duplicate publication bias: studies that are published in more than one place might get more weighting than other studies.
5.      Outcome reporting bias: selective reporting of certain outcomes, such as outcomes that paint a company in a good light.
6.     Publication bias: studies with positive findings are more likely to be published — and published faster — than studies with negative findings or no significant findings.
7.      Time lag bias: some studies take years to be published, especially if they show no effect or have unwanted results. Studies that are positive or newsworthy are published much faster.
 
Reporting Bias can occur in the life cycle of many research and are as follows:

Reporting Bias are widespread phenomena in the medical literature. So, the organisations took some preventive measures to safeguard itself from reporting bias. Transparency is the most important action to safeguard health research. 
Tips to avoid different types of bias during a trial are given below:
 
 
1. Clearly define risk and outcome, preferably with validated methods and standardize data collection.
2. Samples should originate from the same general population.
3. Standardize interviewers interaction.
4. Use prospective studies and avoid using historical controls.
5. Use objective data sources whenever possible.
6. Carefully design the plan for lost to follow up.
7. Clearly define exposure prior to study.
8. Validate measures as primary as primary outcome.
9. Consider cluster stratification to minimize variability.
10. Register trial with an accepted trial register.
11.  Unknown confounders can only be controlled with randomization.