Health statistics form the empirical foundation of public health. To measure the health status of a community, plan healthcare delivery, and evaluate interventions, reliable data is paramount. The Health Information System (HIS) is a mechanism designed for the collection, processing, analysis, and transmission of information required for organizing and operating health services.

Primary Sources of Health Information

The primary sources of data include the Census (a decennial total population count), Registration of Vital Events (births, deaths, marriages), and the Sample Registration System (SRS) which provides reliable estimates of birth and death rates at national and state levels. Other critical sources include epidemiological surveillance, hospital records, and disease registries.

Once data is collected, it must be organized to be comprehensible. Tabulation is the systematic arrangement of data in rows and columns. Good tabulation is concise, self-explanatory, and follows strict rules regarding titles, stubs (row headings), and captions (column headings).

Graphic and Diagrammatic Methods

Depending on whether the data is quantitative (continuous/discrete) or qualitative, different visual tools are used. For continuous quantitative data, Histograms, Frequency Polygons, and Ogive curves are standard. For qualitative or discrete data, Bar Charts, Pie Charts, and Pictograms are preferred.

To summarize large datasets, we use single numerical values that represent the dataset. These are the Measures of Central Tendency: the Mean (arithmetic average), the Median (middle value), and the Mode (most frequent value).

However, central tendency alone doesn’t show how scattered the data is. For this, we use Measures of Variability (Dispersion). The most common are Range, Mean Deviation, Variance, and Standard Deviation (SD). The coefficient of variation (CV) is used to compare the relative variability of two different distributions.

Studying an entire population is often impossible, hence we take a Sample. Sample size determination depends on the desired confidence level, margin of error, and estimated prevalence. Sampling techniques are divided into Probability (Random, Stratified, Cluster) and Non-Probability (Convenience, Purposive).

Tests of Significance (like t-test, Chi-square, Z-test) help determine if observed differences are true or due to chance. Correlation measures the strength of the relationship between two variables, while Regression predicts the value of one variable based on another.