sample size calculation for Bio Medical research

The significance of sample size calculation lies in its ability to ensure that research studies produce reliable, credible, and meaningful results. Here's why it's so important.
Accuracy and Precision: Sample size calculation helps researchers determine the right number of participants needed for their study. This ensures that the results are accurate and precise, reflecting the true characteristics of the population being studied.
Generalizability: A well-calculated sample size ensures that the study findings can be generalized to the broader population. By including a representative sample, researchers can draw conclusions that are applicable beyond the study sample, enhancing the external validity of the research.
Resource Optimization: Sample size calculation helps researchers optimize the use of resources such as time, money, and manpower. By determining the minimum sample size required to achieve the study objectives, researchers can avoid wastage of resources on overly large or underpowered studies.
Ethical Considerations: Calculating the appropriate sample size is essential for ethically justifying the involvement of participants in research studies. It ensures that the potential benefits of the study outweigh any potential risks or burdens to participants, aligning with ethical principles of beneficence and non-maleficence.
Regulatory Compliance: Regulatory agencies often require justification for sample size decisions in research studies, particularly in clinical trials and studies involving human subjects. Sample size calculation provides a transparent and evidence-based rationale for study design, facilitating regulatory compliance and approval.
Reducing Bias and Error: Adequate sample size reduces the impact of random variation and sampling errors, enhancing the internal validity and reliability of study findings. It minimizes the risk of biased or spurious results, allowing researchers to draw more accurate conclusions about the relationships or effects under investigation

1. Sample size Calculation for Medical Research Dr. Swati Patel , M.Sc (Statistics)PhD Statistician Department of Community Medicine SMIMER

2. Study Population Sample Sample is subset of population

3. Descriptive Study – For qualitative variable When study parameter is single ,qualitative variable then calculate sample size by using ,   Or Use Open Epi & Epi tools Software to calculate sample size when population size /study population is known

4. Descriptive Analytical It is used to describe the occurrence of disease by time ,place and person. It deals with to estimate the population parameter. Two commonly used parameters are mean and proportion To examine the etiology and strength of causal associations.

5. Objective

6. Assuming the prevalence non communicable disease among the diabetic patients …………...  

7. What we need to calculate sample size P = Proportion of DM in HIV Infected patients previous study or Pilot survey ………\ Q= 1-P E= Allowable Error = 1% to 9%...... = Standard normal value at 95% level=1.96  

8. P= 6.4% from previous study N= 575 by formula  

9. Descriptive Study – For quantitative variable Suppose the researcher want to know the average BMI of adult for the same city then below mentioned formula should be used ,

10. The researcher is interested in knowing the average BMI in adult age group of that city , and based of pilot survey researcher found that SD of BMI 0.7339 and allowable error 0.05 units , then formula for sample size calculation will be  

11. Alpha: Chance of concluding that the experimental treatment is more effective when in fact it is not Beta: Chance of claiming no difference when a difference exists Types of Error :  and β H True H False Accept β (FN) Reject  (FP) Reality Decision P+Q = 1 Q = 1 – P

12. Null Hypothesis( Ho ) is true Patient has cancer Null Hypothesis( Ho ) is false Patients does not have cancer Reality: Participant does not has cancer Type I Error (model predict cancer but patients does NOT have cancer) False Positive Correct outcome (model predicts no cancer and patients does NOT have cancer) True positive Reality: Participant has cancer Correct out come (model predicts and patients does have cancer ) True negative Type II error (model predicted no cancer but patients does have cancer) False negative

13. sample size calculated by considering the proportion/prevalence of HTN in patients on HAART and HAART naïve patients of 17% and 2% respectively ,from previous study(19), Level of significance/confidence level =95% Power =80%. Information required to determine Sample size for Cross sectional Analytical – For two qualitative variable

14. Objectives: 1. Estimate of serum calcium level in Diabetes and non- diabetes. 2.Compare the serum calcium level in Diabetes and non- diabetes.

15. Sample size sample size calculated by considering the mean calcium level in Diabetic is 8.87±1.453 and non- diabetic patients is 9.51±0.4 from previous study. Level of significance/confidence level =95% Power =80%.

16. Exercises

17. A researcher wants to perform a study for hypertension among diabetics and non-diabetic patients to find out………………. From previous studies, percentage of hypertension among Diabetics was 70% and among non diabetics was 40%

18. An investigator want to conduct a study to find out whether there is any difference in effect of pollution on lung function by studying force expiratory volume (FEV1) between traffic police and general population…….. The null hypothesis that there is no difference in mean FEV1 in traffic police and general population. From the previous study it is known that mean difference of FEV1 is 2.5 l/min and SD of FEV 1 is 3.5 l/min and 5 l/min among traffic police and general population respectively . How many subject from traffic police and general population required for study?

19. If a researcher want to see the impact of smoking on cardiovascular mortality ,then he/she will take two groups one having subject who smoke and other consisting of who do not smoking. At the end of the study both groups are to be compared for cardiovascular mortality . According to the previous studies the proportion of cardiovascular death in who do smoke is 43.5 % and who do not is around 20% .

20. Thank You