Understanding Biostatistics

Biostatistics Definition

• Statistics refers to statistical data or methods
• It involves studying and using theories and methods to analyze data from random processes
• Statistics provides tools and techniques for the scientific method, including forming hypotheses, designing experiments, gathering data, summarizing it and drawing inferences

Statistics Types

• Can be mathematical or applied
• Mathematical Statistics focuses on the study and development of statistical theory and methods
• Applied Statistics involves applying statistical methods to solve real-world problems with randomly generated data.

Biostatistics

• A branch of applied statistics for health sciences and biology
• Biostatistics differs from biometry depending on whether applications are health-related or in broader biology fields like agriculture, ecology, and wildlife biology

Why Biostatistics?

• Some statistical methods, are more common in health applications, like survival and longitudinal data analysis
• Using health examples makes the subject more appealing to those in health-related fields
• It illustrates how methodology applies to similar real-life problems

Characteristics of Statistical Data

• They must be aggregates, meaning they are 'number of facts'
• Data needs to be affected by multiple causes
• Data must be enumerated or estimated with reasonable accuracy to be reliable
• Collection of information must be systematic and for a clear purpose by the enumerator
• Data points must be comparable, relating to each other in time, space, or condition

Studying Statistics: Rationale

• Statistics provides a way to organize information formally, beyond anecdotes
• Quantification is increasingly used in medicine and public health providing intrinsic variation in biological processes
• Quantitative information is becoming more common in public health and medicine due to technology
• Statistics is essential for interpreting the results of data manipulation

Statistical Technology & Medical Research

• Planning, conduct, and interpretation of medical research increasingly relies on statistical technology
• Statistics help determine if new treatments are better, assess side effects, determine patient numbers, and understand clinical measurements

Statistics in Medical Literature

• Statistics pervades medical literature due to the quantitative nature of public health and medicine
• Interpretation of data with variability is central to statistics

Statistical Literacy in Nursing: Relevance

• Nursing relies on scientific research for positive patient outcomes
• Statistics are integral to nursing, impacting patient care and policies
• Numerical data informs and influences, requiring statistical knowledge

Statistics Role in Enhancing Patient Care

• Statistics help nurses prioritize treatments and assess patient needs
• Nurses use statistics to identify trends in vital signs, make informed decisions, and document medication timing

Leveraging Statistics for Patient Understanding

• Statistics quantify issues and challenge nurses to respond effectively
• Nurses can expect patients to be aware of statistics surrounding critical care

Statistics Supporting Evidence-Based Practices

• Nursing practice relies on empirical evidence for effective protocols
• A basic understanding of statistics is needed to interpret relevant literature

Statistics help nurses to develop professionally, improve practice, and lobby for policy changes. Florence Nightingale demonstrated improved sanitary conditions can reduce deaths.

Statistics Role

• Helps make informed decisions in critical care settings
• Emotional instinct alone can be unreliable in difficult circumstances
• Can teach nurses to be cautious in their approach
• Can remind nurses that outcomes are complex, requiring consistent study over time
• Helps nurses to understand that study results can vary and it is important to be discerning

Association Versus Causation

• Nurses should distinguish association from causation in statements
• Terms like 'factor,' 'influence,' and 'related' often indicate association, not causation

Levels of Measurement

• Scales used to express theories of scale types
• A variable has one of four levels of measurement: nominal, ordinal, interval, or ratio

Nominal Level

• Numbers classify data
• Letters, words, and alphanumeric symbols can be used

Ordinal Level

• Depicts an ordered relationship between variables

Interval Level

• Classifies and orders measurements
• Specifies equal distances between each interval on the scale

Ratio Level

• Observations have equal intervals and a value of zero

Data Sources

• There are two main types: statistical and non-statistical
• Statistical sources include censuses and official surveys
• Non-statistical sources include data collected for administrative purposes

Statistical Survey

• Normally conducted using a sample
• Used to collect sample data and analyze it using statistical methods

Census

• Based on all items of the population
• Provides comprehensive data collection for a specific period

Register

• Storehouses of statistical information for data collection and analysis
• Data from here is reliable

Primary and Secondary Data

• There are two main types of data
• As the name suggests, primary data is taken by the surveyor from first-hand experience
• Secondary data is that which has already been taken and recorded by an organisation

Primary Data Collection Methods

• Personal investigation: the surveyor collects the data themself
• Collection Via Investigators: trained investigators are employed to contact reoponsdants
• Questionnaires: can be used to ask specific questions
• Telephonic Investigation: questions are asked over the phone for accuracy

Secondary data Types

• Is opposite to primary data
• Already collected and published by some organization

Statistical Variables

• All experiments examine some kind of variable(s)
• A variable can be measured, manipulated and controlled for
• There are independent and dependent variables

Independent and Dependent Variables

• An independent variable is sometimes called experimental or predictor
• It can be manipulated in an experiment

Experimental and Non-Experimental Research

• Examines variables in different ways

Experimental research

• The aim is to manipulate a variable and observe its impact on a dependent variable(s).

Non-experimental research

• The researcher can not manipulate variables, which can lead to unethical practices.

Categorical and Continuous Variables

• Categorical variables are also known as discrete or qualitative variables
• Continuous variables can be further categorized as either interval or ratio variables.

Nominal variables

• Variables that have two or more categories, but which do not have an intrinsic order

Dichotomous variables

• Nominal variables which have only two categories or levels

Continuous variables

• Otherwise known as quantitative variables
• Continuous variables can be further categorized as either interval or ratio variables.

Interval variables

• Central characteristic is that they can be measured along a continuum and they have a numerical value

Ratio variables

• Are interval variables, but with the added condition that 0 (zero) of the measurement indicates that there is none of that variable.