VIEWS: 0 PAGES: 18 POSTED ON: 12/14/2011
Introduction to Statistics and Biostatistics and Definitions REFERENCE BOOKS 1. WA 900 D184 1991 Daniel, Wayne W. Biostatistics: A Foundation for Analysis in the Health Sciences 2. WA 950 D272 2004 Dawson, Beth Trapp, Robert G. Basic & Clinical Biostatistics Lange Medical Books/ McGrow-Hill NewYork 2004 3. WA 950 G545 2002 Glantz, Stanton A. Primer of Biostatistics McGrow-Hill NewYork 2002 4. WA 950 A733 2002 Armitage, P. Statistical Methods in Medical Research Blackwell Science Oxford 2002 This course teaches the basic skills needed to critique the medical literature by providing a fundamental understanding of biostatistics. Course Description The primary purpose of this course is to teach you basic skills to critique the medical literature and need for statistical consultancy when designing, conducting and reporting a research. As future physicians you have an obligation to remain current in your field of practice and to treat patients according to generally accepted standards of care. To do this well you will need to read those journals that are considered the most important sources of new information impacting on your field of medicine. For example, internal medicine physicians generally read the New England Journal of Medicine (NEJM) which is published by the Massachusetts Medical Society. Perhaps the most influential medical journal, its articles are widely reviewed by the media. Most surgeons read the Annals of Surgery. In short, each field of medicine has its own specialty journals. As your patients hear media reports about the latest research results, they will most certainly ask for your opinion: Should adult patients take antioxidants to help prevent heart disease? Do the potential benefits of hormonal replacement in postmenopausal women outweigh the potential risks? STATISTICS and BIOSTATISTICS Statistics is the art and science of data. It deals with •Planning Research •Collecting Data •Describing Data •Summarizing- Presenting Data •Analyzing Data •Interpreting Results • Reaching decisions or discovering new knowledge Biostatisitcs is the application of statistical methods to health sciences. Basic tasks of statistics •To describe •To draw inferences concerning the underlying population Classification Descriptive statistics: These are devices for organizing data and may be: (i) tabular (ii) graphical (iii) numerical Inferential statistics: Much of the statistical investigations are carried on samples. Conclusions about the population are thus drawn from the observations carried out on these samples. Statistical methods which are used for making inferences about the population are based on probability theory and are called inferential statistics. Data : Set of values of one or more variables recorded on one or more observational units. Observation (case): Individual source of data. Variable: This is a quantity which varies such that it may take any one of a specified set of values. It may be measurable or non-measurable. Population: A collection, or set, of individuals, objects, or measurements whose properties are to be analyzed. Sample: A subset of the population, selected in such a way that it is representative of the larger population. Parameter : A summary value which in some way characterizes the nature of the population in the variable under study. Statistic : A summary value calculated from a sample of observation. Data Sources of data 1. Routinely kept records 2. Published data sources 3. Data on electronic media 4. Surveys and Experimental research 5. Census 6. Generated or artificial data Types of Data 1. Qualitative Data Results from a variable that asks for a quality type of description of the subject. 2. Quantitative Data Results from obtaining quantities-counts or measurements. Scale of measurement Qualitative (Categorical) Data Nominal Ordinal Severity of disease Two More than two Income level categories categories Marital Status Binary Dichotomous Disease – or + Gender(M;F) BMI(<23;>=23) Quantitative Data Discrete Continuous # of birth # of death Interval Ratio Temperature Age IQ Weight Height Why should medical students learn biostatistics? 1. Medicine is becoming increasingly QUANTITATIVE. • The aim is to improve the Health Status of the population. • We have to clarify the relationships between certain factors and diseases. • Enumarate the occurances of diseases • Explain the etiology of diseases (which factors cause which diseases) • Predict the number of disease occurence • Read, understand and criticize the medical literature. Why should medical students learn biostatistics? 2. The planning, conduct and interpretation of much of medical research are becoming increasingly reliant on statistical methods. Planning: How many patients must be treated? How do we have to allocate the subjects to treatments? What are the other factors which may influence the response variable? Conduct: Under which conditions must the study be conducted? Is matching necessary? Is blinding (single blinding or double blinding) necessary? Is there a need for a control group? Shoud the placebo effect be considered? Which experimental design technique is more appropriate? Interpretation Example 1 Distribution of Women with a Diagnosis of Distribution of Healthy Women Among Tromboembolism Among Blood Groups Blood Groups Blood Group Frequency % Blood Group Frequency % A 32 58.2 A 75 51.7 AB 4 7.3 AB 8 5.5 B 8 14.5 B 19 13.1 O 11 20.0 O 43 29.7 Total 55 100.0 Total 145 100.0 Example 2 In a follow up study 609 males between ages of 40 to 76 free of coronary heart disease (CHD) were examined from 1990 to 1999. At the end of the period 71 new cases of CHD were identified. The primary goal of this study was to evaluate the association between obesity and the incidence of CHD. The results were as follows: Risk Ratio= R Obesity 27 / 122 Disease Yes No Total R= 44 / 487 CHD+ 27 44 71 CHD- 95 443 538 The risk of CHD is 2.45 Total 122 487 609 times higher among obese subjects.
Pages to are hidden for
"Introduction to Statistics and Biostatistics and Definitions"Please download to view full document