NATURE OF BIOSTATISTICS ALVIN DUKE R. SY, MSPH, RN Learning Objectives At the end of the session, the learners should be able to: explain the role of biostatistics in public health, differentiate the different types of data, and ® discuss common terms associated with biostatisticsTDP ea Data is Everywhere! Data is utilized and summarized frequently... ...in research literature OF 156 and 155 participants in usual care and clinical pathway periods, during their initial care, 60 (38.5%) usual care participants were hospitalized, and 27 (17.6%) pathway participants were hospitalized (difference = 20.8 percentage points [PP], 95% confidence interval [Ci] = 10.4 to 31.2 PP). Over 44 days, 70 (44.9%) usual care and 44 (28.8%) pathway participants were hospitalized (difference 16.1 PP, 95% Cl = 4.9 to 27.4 PP).* Data is utilized and summarized frequently... ~» in research literature =, is populormndig A report from the CHO HIV/AIDS Core Team showed that the confirmed cases in the city already increased to 112 as of the end of September, a significant increase from the 44 detected last year. A total of 27 new HIV/AIDS cases were reported on July, 23 cases in August, and 18 in September, with the youngest involving 15-year-old male. * Data is utilized and summarized frequently... ~»...in research literature +... popular media * Data can be analyzed and summarized to provide useful information. » BAD data can be analyzed and summarized to provide incorrect or harmful or non-informative information. SEE ee | 6Stal'sbicy isa science dealing with the collection, organization, analysis, and interpretation of numerical data. Public Health Statistiex refer to quantitative data needed as basis for planning, monitoring, and evaluation of health services * Vital statistics data on vital events (births, deaths, marriage) * Health statistics data on morbidity (causes and frequency of illness) hospital and clinical statistics (no. of admissions and discharges, be occupancy, length of stay, etc.) ~ service statistics (no. of baby immunized, no. household with sanitary toilet, no. of women practicing family planning, etc.)Brostalistes .wis the scientific discipline focused with the application of statistical principles concerning biological problems _-specific applications include public health, medical sciences, agriculture, environmental and ecological sciences Brostabsbes __can be useful in cerms of being: ¥ adata processing & analytics technique, ¥ a tool for objective appraisal and evaluation, ¥ a tool for decision-making, or ¥ a tool to design and analyze research well“assessment of the health status in the community (risk factors) = prevention and control of disease = monitoring and evaluation of health interventions ® provision of health policy formulation and health care planning Descriptive + summarize and present data ~* measures of central tendency measures of dispersion and location ~* tabular presentation ~* graphical presentation — 2 = determination of disease causation ® Inferential * predictions, generalizations, and conclusions ~» estimation of parameters ~~» hypothesis or significance testingDesersptive Statistics >» ..tabulation of number of probable dengue cases per region in the Philippines >» .ause of histogram to present frequency distribution of newborns based on their birthweight >> calculation of average length of stay of patients in the hospital B Population population Sample Select a sample from the populationEstimation woprocess by which a statistic computed from a random sample is used to approximate or estimate the corresponding parameter ..generation of point or interval estimates Hypothesis testing ..Set of procedures that culminates in the rejection or non- rejection of the null hypothesis based on the probability of occurrence of the sample results if the null hypothesis were true =~ Nature Cy Statistical Data21/08/2023 * Statistical data must have the following characteristic: ~ expressed numerically, ~* treated as a mass or group of observations, and ~~» subject to variation Vaniabilly «tendency of a measurable characteristic or behavior to change .-Statistics is necessary to analyze variabilityVariable ..refers to the measurement or characteristic of interest Conta ..a phenomenon whose value remains the same from person to person, from time to time, or from place to place *= pull of gravity = 6.674x10-" m&kgr hs? = speed of light = 299,792,458 meters per secondA study was done among UP Manila students to determine the proportion of current smokers School Smoking status Constant Variable ony 2 A study was done among universities in Manila to determine the proportion of students who are current smokers School Smoking status Variable Variable cI er | 2Qualitative non-quantifiable categories numerical representation as labels or codes only sex educational level disease status eye color Quantitative values indicate quantity or amount either discrete or continuous discrete variables assume only integral values or whole numbers continuous variables attain any value including fractions and decimalsNombenl, + ..a classificatory scale used only as labels * ..represent a set of mutually exclusive and exhaustive classes ~* sex (male or female) ~+ blood groups (i.e., A+, B+, AB+, O+) ~+ medical diagnosis (e.g., HA Pneumonia, CAP, etc.) 25 Gulia, » shas similar characteristics to nominal variables, but the categories can be ordered or ranked -+ disease severity (e.g., Mild, Moderate, Severe) “+ age groups (e.g., Infant, Child, etc.) ~ rating (e.g., Agree, Disagree, etc.) 26Ledewal « ..has similar characteristics to ordinal variables, however the distances between all adjacent classes are equal * ...these variables do not have a true zero point (arbitrary) ~* temperature in Celsius » calendar time * standard IQ test 27 Rufio avith similar characteristics to variables of interval scale, except it has a meaningful zero point (that exists) + ..the “ratio” of two ratio variables can be meaningfully computed and interpreted ~+ blood pressure * number of doctor visits ~» weight and height 28¥ A variable can either be quantitative or qualitative at one point in time, but not both, at the same time. ¥ A variable can be classified under one scale or level of measurement only at a single point in time ¥ A variable can be quantitative or qualitative depending on the objective of the data collection ¥ Likewise, the scale of measurement can vary depending on how the variable is treated Se ee 30* ...to make sense of massive amount of information * .needed to avoid reaching invalid conclusions out of limited data ~» Research has shown that the ability to generalize from a sample to a population is hardwired into our brains and has been observed in eight-month-old babies * ..essential to avoid being fooled by seeing apparent patterns among random data : qh, certain pveTay Celaya Daniel WW & Cross CL Biestaistics: A Foundation for Analyssin the Health Science, roth ed. New York: Wiley, 2013. Gubulani R. Group worved over surge of HIV/AIDS cases in GenSan, Philippine News Agency, Oct 7, 2021 Ittps:{fiww.yma gov.phtarticles/n55859 -MedinaJRC. Lecture Notes on Bistalistis, Department of Epidemiology & Biostatistics, Collegeof Public Health, UP Manila 2022. ‘Mendoza OM, Borja MP, etal Founuations of Statistical Analysis forthe Health Sciences, UP Manila: DEBS, ‘alan DA, Mower WR, etal Pathway wth single-dose long-acting intravenous antibaticreduees emergency department hospitalizations of patients with skin infections, Acad Em Med, 20%; 28(0):108-17