JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD Pharm. D IV Year BIOSTATISTICS AND RESEARCH METHODOLOGY Previous Question Paper (April/May-2023 - R08)

A case study is an in-depth, detailed examination of a particular case or a small group of subjects within a real-world context. It involves the systematic collection of data through multiple sources such as interviews, observations, medical records, and historical documents to provide a comprehensive understanding of a complex issue.

Case studies are conducted for several key reasons:

• To explore rare diseases or unique clinical phenomena where large-scale statistical analysis is not feasible.
• To generate hypotheses for future, larger-scale quantitative research.
• To provide a holistic view of patient care, including the influence of environmental, psychological, and social factors on health outcomes.
• To illustrate the application of theoretical concepts in clinical practice.

The process typically involves defining the research question, selecting the case, collecting data through triangulation, analyzing the evidence, and reporting the findings to highlight specific insights that might be overlooked in broader surveys.

Determining the appropriate sample size is critical to ensure that a study has sufficient power to detect a meaningful effect or to estimate a parameter with the desired precision.

For simple comparative experiments, the sample size (n) is determined based on the desired power (1-β), the significance level (α), and the minimum clinically important difference (Δ) between two groups. The formula for comparing two means is:

where Z_α/2 is the critical value for the significance level, Z_β is the critical value for the power, and σ² is the population variance.

For confidence intervals of specific width, the objective is to estimate a population mean with a margin of error (E). The sample size is calculated by rearranging the confidence interval formula:

In this case, the width of the confidence interval is 2E. As the desired precision (smaller E) increases, the required sample size increases quadratically. If the population standard deviation (σ) is unknown, a pilot study or previous literature is used to estimate it.