Quantitative Methods in Research by Mumbi Chola - UNZA School of Public Health - Epidemiology & Biostatistics-unza

Quantitative Methods in Research Mumbi Chola Dept. of Epidemiology & Biostatistics UNZA School of Public Health mumbi24@gmail.com Outline Lecture Objectives Quantitative Methods Sources of Quantitative Data Questions in Quantitative Research Units and Variables Quantitative Designs Summary Lecture Objectives Understand what quantitative research is Understand the sources of quantitative data Understand the type of questions answered by quantitative research Know what a unit and variable is Have a basic understanding of the various epidemiological designs and be able to distinguish among them Quantitative Methods Quantitative research is ‘Explaining phenomena by collecting numerical data that are analysed using mathematically based methods (in particular statistics)’. Aliaga and Gunderson (2000) Quantitative Methods Those which generate numerical data Involve quantitative analysis using statistics to improve numerical data Like all data collection and analysis, choice of quantitative method should be driven by a question Sources of Quantitative Data Primary data Data collected to answer a specific research question Usually involves the research collecting data Secondary data Existing institutional data Routine data Usually collected for other purposed Typically analysed to answer research questions Questions in Quantitative Research There are four main types of research questions suited for quantitative research : The first type of research question is that demanding a quantitative answer. E.g. How many have been affected by the Ebola virus? 2. Numerical change can likewise accurately be studied only by using quantitative methods. E.g. Are the numbers of Breast Cancer patients at UTH rising or falling? Questions in Quantitative Research 3. Finding out about the state of something or other, we often want to explain phenomena. E.g. What factors are influencing pregnancies among teenagers in Lusaka? 4. Hypotheses testing. We might want to explain something E.g. Whether there is a relationship between disease outcome and type of birth of a child (i.e. natural and c-section) Units and variables Quantitative data is collected from someone or something which are known as Units or cases The data that we are collecting from these units are known as variables Variables are any characteristic of the unit we are interested in and want to collect (e.g. sex, age, marital status, HIV status). The label ‘variable’ refers to the fact that these data will differ between units. For example, HIV status will differ by province or level of knowledge will differ by education level Quantitative Research Designs Quantitative approaches Cross- sectional Case-control Trials Intervention randomly assigned Individually randomised Cluster randomised Observational Experimental Studies Main types of epidemiological studies Cohort Time Disease or no disease Unexposed persons Exposed persons (PROSPECTIVE) COHORT STUDY Exposed or unexposed Persons with disease Persons without disease CROSS-SECTIONAL STUDY (RETROSPECTIVE) Epidemiologic Study Designs The past The future Now CASE-CONTROL STUDY Cross-sectional studies All information usually collected at the same time Primary data collection Secondary data analyses (already existing data) Can collect information on previous exposure Prevalence studies Surveys Repeated cross-sectional studies: trend analyses Cross-sectional studies Usually population-based Often random sample from a population with variation in both exposure and outcome (disease status) Often stratified based on certain variables (age, gender) Examples of Cross sectional Studies Zambia Demographic and Health Survey Census of Zambia Living Conditions Monitoring Survey Economic Census of Zambia Time Disease or no disease Unexposed persons Exposed persons (PROSPECTIVE) COHORT STUDY Exposed or unexposed Persons with disease Persons without disease CROSS-SECTIONAL STUDY (RETROSPECTIVE) Epidemiologic Study Designs The past The future Now CASE-CONTROL STUDY Time Disease or no disease Unexposed persons Exposed persons (PROSPECTIVE) COHORT STUDY Exposed or unexposed Persons with disease Persons without disease CASE-CONTROL STUDY (RETROSPECTIVE) Epidemiologic Study Designs The past The future Now CASE-CONTROL STUDY Advantages Of Case-control Studies Advantages: 1. Useful in studying relatively rare diseases 2. Important in understanding new diseases 3. Commonly used in outbreak investigation 4. Useful if induction period is long 5. Relatively inexpensive 1. Susceptible to bias if not carefully designed (and matched) 2. Especially susceptible to exposure misclassification 3. Especially susceptible to recall bias 4. Restricted to single outcome 5. Incidence rates not usually calculable 6. Cannot assess effects of matching variables Disadvantages Of Case-control Studies Examples of Case-Control Studies Cholera outbreak in Lusaka from October 2017 in Zambia has resulted in approximately 5,900 cases and 114 deaths. (Sinyange et al, 2018) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6048949/ 1950's Cigarette smoking …