Health Research Notes PDF

Health Research Notes Week 1: Introduction to Research Modes to Acquire Knowledge ○ tenacity ideas accepted as valid have existed for a long time or are often repeated ○ intuition ideas accepted as valid seem...

Health Research Notes Week 1: Introduction to Research Modes to Acquire Knowledge ○ tenacity ideas accepted as valid have existed for a long time or are often repeated ○ intuition ideas accepted as valid seem to be intuitively true ○ authority ideas have been validated by a respected person valid documents are declared as true ○ rationality idea are validated based on principles of logic/existing ideas ○ empircism gaining knowledge through experience ○ science a combo of raitonality (to develop theories) and empiricism (tests theories) Scientific Process ○ a logical/rational process which systematically collects and anlayzes data to describe, explain, predict and control a phenomenon 3 Aims of Research ○ Discovery/Description description refers to purely explain a phenomena this is done in qualitative research, we try to understand the POV of participants and interpret results description is done in quantiative through descriptive stastics ○ Exploration/Explanation exploration describes relationships between concepts/phenomena descriptive and correlative in nature explanaiton goes beyond the simple relationship by establishing relationships between concepts/phenomena and determines their purpose uses advances stastical tests can be quantitative or qualitative ○ Prediction/Control  this happens in experimental research to assess the probability of certain outcomes in a provoked situation prediction- refers to finding causal relationships between variables control- varies conditions in a research setting to produce a certain result Scientist Traits ○ doubt & curiosity- they are eager to learn more through integration into a theory/model eager to learn bc they have an underlying motivation their results should be replicable they have to be objective and impartial - should put aside personal biases they have to seek facts, not rely on personal values Theory vs Template ○ theory- an abstract explanation of relations that unite facts, concepts and propositions facts turn into a question which gathers more facts comprehensively organizes knowledge and guides research ○ template- an organized set of concepts/variables and their interrelationships less comprehensive but the same idea Scientific Fact (replicability principle) ○ a fact is scientific if observations can be repeated ○ example: scientific articles follow conclusions that other ppl have come to Paradigms of Research ○ a reference model that takes into account values, concepts, beliefs, and principles to decide how disciplines approach phenomena positivism-oriented → thinks reality is objective interpretivism-oriented → thinks reality is constructed 11 Steps in a Research Process ○ identify problem ○ research question/hypothesis ○ lit review/critical reading ○ research proposal ○ select sample ○ set measures ○ data collection  ○ data analysis ○ interpret data ○ writing ○ dissemtinaiton of results Guidelines for Successful Research ○ focus on priority problems ○ action-oriented ○ multi-disciplinary ○ participatory ○ timely ○ cost-effective ○ simple, short-term designs ○ clear results ○ honest limits ○ expressed implications and recommendations General Forms of Research Foundational vs Applied (purpose) ○ foundational- generates new knowledge to generate theories or understand a phenomenon ○ applied- finds solutions to practical problems like checking the efficacy of a drug Quantitative vs Qualitative (method)- 5 differences ○ quantitative post-positivism approach and deductive reasoning a formal, objective and systematic process used to describe and verify assumptions tests assumptions measures association between variables using correlation or cause-effect output is numerical data ○ qualitative constructivist and interpretive approach uses inductive reasoning informal, subjective, non-systematic process used to generate assumptions to understand phenomena that are difficult to access output is narrative data Experimental vs Non-Experimental (means put in place) ○ experimental one/more variables are manipulated to study cause and effect relationship and produce a certain outcome, used to predict and control the objective  we have control over the study subject only used for quantitative ○ non-experimental none of the conditions are manipulated and used to understand the discover, describe and explain phenomena used for both qualitative and quantitative Characteristics of Research general ○ should clearly state the problem, objectives, methods ○ should build on existing data ○ involve a systematic process ○ have a clear vision of the outcome who is involved ○ everyone concerned with the problem: policymakers health care providers managers community ○ how much they participate and their role depends on: level, complexity and the focus of the study a must for research to be ○ purposeful ○ targetted ○ credible ○ timely Identifying & Prioritizing Topics for Research Identification of a Topic - why is it required? ○ required to describe a health situation or problem When Problem Requires Research ○ when there is a perceived discrepancy between what is expected and reality ○ when there is no clear explanation for the discrepancy ○ when there is more than one possible solution/answer to a problem Ways to Think of a Topic ○ own personal experiences/observations ○ talk to researchers/experts ○ talk to healthcare providers/community ○ read scientific articles/journals ○ formal ads by organizations 7 Criteria to Prioritize a Research Topic ○ relevancy  ○ urgency ○ avoid duplication of results ○ application of results ○ ethical accetibility ○ political acceptibility ○ feasibility of study 3 Features of a Good Topic ○ short and precise (specific) ○ catches readers’ attention ○ must be related to the general objective of the study (the purpose) Analysis & Statement of the Problem 3 Reasons To Analyze the Problem ○ to pool and collect knowledge by doing a literature review ○ determine the core problem and factors ○ determine the focus and scope of the research 3 Steps in Analyzing the Problem ○ talk to participants to clarify the problem ○ specify the core problem (nature, intensity, size, distribution) ○ analyze the problem, the contributing factors and categorize them to make a conceptual framework Deciding Focus & Scope of Research: Points to Consider ○ Usefulness how useful is information to be collected ○ Feasible analysis of data should be feasible ○ Duplicable findings should be duplicated with previous studies - ensures validity and reliability  Week 2: Selecting the Research Topic & Searching Databases 3 Different Research Goals 1. descriptive → to describe and explain phenomena/concepts 2. explanatory → to describe links between concepts through correlation/regression analysis 3. predictive → to find cause-and-effect relationships 5 Phases of Research Process 1. conceptual → coming up with ideas 2. methodological → actual experiments/trials 3. analytical → analyzing results 4. empirical → using science to explain results 5. interpretation/dissemination → distribution of results 5 Phases of Conceptual Phase 1. choose a topic and do preliminary reading 2. do more critical reading and literature review 3. develop a research framework 4. come up with a research problem 5. come up with research objectives, questions, assumptions Phase 1: Selecting the Research Topic Research Methods ○ quali = to generate ideas and theories ○ quanti = to test theoretical propositions What is a Research Topic? ○ the topic (subject of study) should correspond to the research problem it intends to explain ○ the general and preliminary questions go to a more specific question Sources to Find a Research Topic/Subject of Study ○ personal experiences/observations ○ clinical settings ○ current topics/social issues ○ theoretical propositions (to check theories) ○ published works (research, conferences) ○ priorities (healthcare facilities) What are concepts ○ mental rep of observed facts and their interrelationships physical activity + quality of life Determine Research Question What does the research question do? ○ identifies study topic, population and concepts to go from preliminary to more specific question example: heart disease rates in women and the effect of mental health issues 3 Levels of Research Question ○ descriptive describes human phenomenon and experience through a qualitative subjective approach describes concepts and factors through a quantitative objective approach (variables and their association) ○ explanatory describes links between concepts using correlation/regression analysis ○ predictive predicts cause and effect relationships 4 Ways to Assess the Relevancy of Research Question ○ see if preliminary question is worth exploring ○ see if there’s a rationale ○ see if concepts are verifiable and observable ○ see the study’s feasibility (enough participants, time, resources) Phase 2: Literature Review provides a justification for the research question/assumptions 5 Steps of Lit Review pinpoint your research question ○ by doing preliminary readings using different sources (google, wiki, scholar, library catalogue, periodicals) ○ primary sources- written by the researcher and is a detailed description of results and method (credible) ○ secondary sources- written by another person who reformulated and reinterpreted the researcher’s work, NOT peer-reviewed decide on databases to use ○ 3 types: general, evidence, grey literature identify keywords/descriptors ○ build a research strategy to find topics related to variables/concepts/population ○ use operators and symbols: AND = connects 2 topics and targets search results OR = connects 2 topics and expands search results = searches for all concepts starting with the base word to expand the search () = groups related concepts  “” = searches for specific expression/phrase MeSH = Medical Subject Headings = allows you to search for topics under a certain heading to limit search results. increase sensitivity ○ sensitivity = there aren’t enough resources and we expand question using “OR”/* and setting MeSH to all subsections ○ specificity = there are too many references and we limit the question using AND/” “ and setting MeSH to a few sub-sections ○ large search limits can be refined by: population type of study publication language review final list ○ check relevancy of publications (by checking the number & dates) ○ review more sources if you don’t have enough info 3 Reasons to Read Critically ○ verify the validity of methods and results ○ find similarities/differences between conclusions to put the problem into perspective ○ find new avenues for research Writing Review ○ there should be a synthesis explaining results that are similar/contradictory and find factors to explain them (methods, population, country) ○ can cite using quotes, paraphrases or summary Phase 3: Research Framework an abstract structure formed of theories/concepts because they are related to the research problem theoretical framework ○ briefly explain key relationships between concepts physical activity affects mood, reducing the risk of depression conceptual framework ○ briefly explains a set of concepts physical activity has a good effect on quality of life research framework justifying ○ this involves justifying the use of theoretical or conceptual approaches to explain concepts and establish relationships between them if there are theories that explain/solve the problem → theoretical approach is better Phase 4:Formulation of the Research Problem the research problem is formulated by deductive means in which we go from general to specific questions to convince the reader that the issue needs to be studied ○ issue needs to be written in the research proposal, especially in the intro and lit review 5 steps in Formulating a research problem present the study topic ○ 1-2 sentences in the intro that are brief, decisive, capture the reader’s attention ○ states the problem, source, observed facts, why it's important present data about the topic ○ found in lit review/state of knowledge section and describes what makes up the problem ○ includes the current state of data, factors involved, who is affected, the consequences for the people affected present empirical evidence which states current knowledge from other research ○ indicates what other researchers have written about the topic (lit review documents) and supports the research question ○ includes: similarities and differences if the relationship is established and verified what the results are present a theoretical/conceptual framework which allows you to define relationships between concepts ○ justify the framework you used to describe concepts and their relationships propose solutions and expected results ○ this is the last paragraph of the literature review ○ it should summarize the problem, and current knowledge and be consistent with the purpose of the study ○ it should propose ways to improve the situation and explain why it’s necessary Phase 5: Statement of Purpose, Questions and Assumptions research objective should include ○ the specific variables ○ population under study ○ an action verb action verbs ○ describe can be used in quali or quanti used to describe phenomenona/variables ○ explore/explain used to explain relationships between variables in a correlative or quantitative study ○ predict/control researcher controls the intervention to predict and control the outcome in an experimental study descriptive research questions- which study is used to study phenomenon vs concept ○ a quali study describes a phenomenon ○ a quanti study describes concepts  assumptions (hypothesis) ○ a hypothesis is a statement that expects relationships between variables and needs to be verified empirically (knowledge through experience) types of hypothesis ○ non-directional hypothesis predicts a relationship between vairbales but doesn’t predict the exact nature or direction ○ directional hypothesis predicts the specific nature of the relationship between variables ○ association hypothesis the variables are related and a change in one, means there’s a change in the other ○ causal hypothesis one variable causes another there is a cause and effect relationship between 2 variables conditions to be fulfilled by a hypothesis ○ should have a theoretical foundation ○ plausible ○ verifiable (through research) ○ Clear Week 3: Research Proposal Structure Formulating a Statement of the Problem Why is Statement of Problem Important? ○ it’s the first major section in a research proposal it lays the foundation of the research justifies why we need it facilitates the search for more information Contents of Statement of Problem (what is it) ○ this is a concise description of the nature of the research problem (severity, distribution , nature, intensity, magnitude, consequences) ○ systemically elucidates why the research should by describing attempts to solve the problem, past success, failures and challenges ○ argues that current knowledge isn’t enough to solve the problem and justifies why results will be significant and useful Research Protocol The first stage of research ○ planning and developing a research protocol is an essential first stage What is a research protocol? ○ a max 20-page document that is a precise and clear protocol for what the researcher wants to do and how they will do it ○ has a table of contents and not too many abbreviations Structure of Research Protocol Introduction ○ introduces the topic and theme by generally describing the research area, context aims ○ tries to formulate the problem by highlighting the importance of the topic and the overall objective of the study tries to attract the reader’s attention 1-2 pages Relevant literature/conceptual framework ○ shows the current state of knowledge on the topic and critically analyzes it to justify and clarify your problem ○ critical analysis is not just a summary but finds gaps in the method + limits and strengths in the conclusion ○ what to look for in writings: writings address similar issues, work with similar variables, use relevant methods 3-4 pages ○ structure of literature review show current data and stastics then show theories of why that is and evidence why theories are true then try to find similarities/differences between literature presentation of data - shows stastics theoretical/conceptual model - helps to find current theories state of evidence - shows current knowledge available synthesis - finding similarities/differences in writings Preliminary findings Objectives ○ general vs specific objectives ○ main vs secondary objectives 10-15, half page Methodology ○ content study design specify the type of design and justify why you’re using also discuss potential biases that may threaten validity + how they’ll be minimized  types experimental quasi-experimental single case observation(case control, cohort, cross-sectional) how to choose design nature of the study the participants you have avialble internal/external validity participants (what is the population) specify the target population (if they are humans or animals) describe the eligibility criteria for inclusion vs exclusion groups talk about challenges: duration, process can be difficult, generalizability sampling and recruitment procedures sampling is forming a subset of a target population that represents them there are 2 main methods for sampling (probalistic and non-probalistic) and the choice is made based on the question, budget and participants that are available justify how, who, where and by whom data collection procedures/study flow justify the method you’ll use to collect data/observe justify who collects them, training of evaluators, where, when, time needed for each topic also mention measures that the participants won’t know about if there is an intervention → explain if the intervention is verified, who will do it, where, dosage, how to prevent drop outs and if control groups will be rewarded variable and measurement tools define variables in a detailed way (operational way) and it maybe convenient to describe the intrsument for measuring it at the same time a variable can be categorized as dependent, indpendent variable should be measurable, observable, feasible to measure and may require a certain quality (precision) measuring instruments identify and categorize the measuring instrument for each variable other measurements: pre-tests and instruments attached to appendix (protocol, data collection sheet) sample size (calculation if needed) whats the sample size based on? the sample size is based on the objectives of the study and power (a stastical analysis method) you should be able to justify and explain any parameters used stastical analysis  describe and explain the main statistical operations for analyzing data and why each one is adequate for analyzing the objective Ethics ○ make arrangements to respect participant integrity by having research ethics committee review it REC might require or suggest changes to protocol ○ consent form elements describe role of subject the right to free and informed consent and withdraw benefits and risks insist on confidentiality of results Anticipated impact and relavance ○ describe what results you expect and who they will benefit and why (will there be a sequel) talk about if there will be socioeconomic benefits ○ describe the scope of results and whether it will be applicable/generalizable to the general population by linking it to government policies dont overestimate the relevance Dissemination of results ○ describe to whom and how you will spread the results and differentiate the dissemination to public vs users Schedule ○ should identify the amount of time for each step and a chronological sequence of dates for the completion of each stage also describe the roles and responsibilities of researchers at each stage ○ this should be realistic even if ambitious References ○ should include all references you used nd be written consistently Appendix ○ this includes all the supplemental information: measurement instruments data collection forms description of intervention program to be evaluated consent form acceptance by ethical comitte letters of support Week 4: Critical Assessment of Scientific Articles & Bias association vs causation ○ association is an identifiable relationship between exposure and disease and it implies that exposure might cause disease  finding an association doesn’t make it causal ○ causation is inferred based upon associaiton and other criteria criticism in research what is criticizing? ○ an intellectual activity that involves assessing the value of a study using a set of defined criteria to appreciate the scientific interest 3 different levels ○ critical analysis requires in-depth knowledge and done by experts/professionals this could be a peer reviewed article the expert should have a sound methodological basis, long-standing clincial experience and a great rigor of reasoning ○ critical reading requires critical thinking and reasoning as well as knowledge abut the research process it examines how the study was designed and conducted and the interpretation of results ○ brief review doesn’t require in-depth knowledge of the research process/area a brief criticism of the article which can be summarized in a few sentenses different sections that can be criticized introduction ○ gives the: statement of problem literature review research framework formulates the goal, research questions & assumptions method ○ describes study design, participant selection, variables, measurement models results ○ describe the samples studies (to know which poopulation was studied) ○ results should be related to the primary objective discussion/conclusion ○ interpretation of results ○ conclusions ○ implications/recommendations (not on exam) introduction & critique  statement of the problem ○ the research problem which should be easy to find, well stated and signficant issue ○ it should also be related to the discipline we are studying literature review ○ has previous research work been described and critizied? ○ has a theoretical basis been recognized? ○ are references recent and indicate current state of knowledge? ○ do references show the evolution of knowledge? ○ is the lit review exhaustive? (is there enough references) research framework ○ what is the research framework (theoretical or conceptual) and is it appropariate? ○ are the concepts described well? ○ are the variables and concepts related? formulates goals, research questions and assumptions ○ what is the purpose/problem? is it descriptive, explanatory or predictive? ○ how will it contribute to increasing knowledge? ○ what are the questions, hypothesis and assumptions, are they well stated? do they logically follow from the research framework? methodology & critique type of study ○ specificy the type of study ○ if experiment, is there an intervention? an exposed vs unexposed group? ○ variables should be appropriate, conceptually and operationally defined confounding factors should be recognized and don’t forget any ○ is internal validity impaired? sample and target population ○ what is the target and sample popoulation? ○ methods and criteria (inclusion and exclusion) for recruiting samples? ○ are sample sizes appropriate? have we lost sight of anyone? ○ participant consent? ○ is internal/external validity impaired? measurement tools used/variables used ○ measuring instruments are the described, suitable, reliable and valid? is the variable measurement scale measured? (nominal, ordinal, continuous) data collection and analysis ○ are stastical/quantitative techniques appropriate? ○ where is it taking place? concept of biases  bias ○ a systemic deformation of study findings sources ○ researcher ○ environment ○ participants ○ measurement tools ○ data collection and analysis process validity ○ a measure that refers to relative absence of bias/systemic error in the measures external vs internal validity ○ external- refers to the idea of generalizing results obtained ○ internal- the chracteristics that affect the structure of the study types of biases (in quantitative studies) selection bias ○ there are errors when you establish comparsion groups so sample doesn’t represent the target population as a result observed results are different from the results you would’ve obtained from the target population this happens due to procedures used to select participants during recruitment and to retain them during the process in case-control (disease vs no disease) this is when you select participants based on disease status bias happens when selection/participation is related to exposure status more likely to occur bc its retropsective and outcomes already happened in cohort/experimental (exposure vs no exposure) this is when you select participants based on exposure status bias happens when selection/participation is relted to disease status types of selection bias inappropriate control selection (case-control) controls have more exposure rates than cases differential participation (case-control/cohort) people with event participate more than controls differential loss of follow up (cohort/experimental) data is lost on ppl that drop out of the study misclassification (information) bias ○ there are errors in collecting information about the exposure and event this could be due to recall bias- when participants provide information about exposure, the participants are more likely to remember exposure if they experienced the event non-differential  measurement errors are the same in exposed/unexposed, cases/controls confusing bias ○ there are other factors, other than the exposure, that are associated with the event this affects the measure of association between exposure and event ways to adjust for confounding stratification randomization matching restriction regression models (linear, logistic, cox) for exposure to CAUSE disease, the exposure has to meet 3 conditions the exposure must be a risk factor (cause) of disease the exposure must be associated with disease the exposure can’t be part of a causal chain linking exposure + disease results & critique types of results ○ descriptive analysis - describes and summarizes data ○ inferential analysis - draws conclusions based on data ○ quantitive analysis - provides more insight? main results why results matter: ○ main results and the threshold of signficiance? ○ results should be stastically signficant ○ results should be in line with the main research question presentation of data: ○ tables, figures and data should be clear ○ both positives & negatives shown discussion/conclusion & critique interpretation of results ○ interpretation should be related to the research question and in concordance with other studies ○ both sig and non-sig data should be included as well as strengths and limits of the methods used ○ can results be generalized (external validity) conclusions ○ what are the conclusions and should summarize the main results well recommendations ○ the results’ consequence on the discipline ○ what recommendations are provided for future research ○ should advance knowledge general form & critique title and summary ○ the title should be clear and precise and the summary should be well described ○ the article should be original and if not how does it stand out from others Week 5: Research Ethics ethical vs morale ○ ethical (guided by universal theories)- a branch of philosophy that systematically examines and argues for good and bad conduct through theories/principles that can be applied universally ○ morale (guided by situations) - a set of rules that decide if an actions allowed or prohibited by an individual/group based on the place, culture, time historical & research ethics world war II & nazi experiments ○ German doctors experimented on deportees at concentration camps ○ they determines suitable candidates from marginalized populations and the rest were killed in gas chambers ○ types of experiments done to improve miliatry survival (test parachutes) improve medical treatment (tested drugs) comfirmed racists ideas of Nazis because they did racial studies/sterilizations ○ the doctors trial a famous trial that targeted 23 doctors who were prosecuted for committing war crimes as a result, the nuremberg code (1947) was established the nuremberg code (1947) ○ a result of the doctors trial ○ the first formal ethical guide to conduct research with humans ○ 5 principles of the nuremberg code qualifications- researchers should have recognized qualifications consent- participants should have voluntary and informed consent practicality- results should be oriented toward the good and have practical use benefits should outweigh risks avoid harm (physical, pscyhological, diability, death, suggering) study of tuskegee (1929-1972) ○ a study was conducted on 400 black men to monitor the progression of syphilis ○ there were divided into healthy and diseased groups  ○ not treatment was found despite penicillin being commercialized in 1940 study of milgram (1960) ○ study was conducted to study obediance to authoirty ○ there was an experimenter (authority), subject (participant) and learner (actor) there were ethical issues involved because the subject were said to be delivered an electric shock (there was no shock) stanford experiement ○ study was done to study incarceration ○ a group of 18 ppl were split into prisoners or guards and there was a lot fo violence, sadicism and emotional trauma involved the helsinki declaration purposes ○ an ethical guideline says that research protocols have to be evaluated by independent experts ○ it led to establishing ethical review committees in medical research faciliities ○ it disinguished between therapetuic and non-therapuetic research therapuetic- benefits individuals non-therpauetic- advances knowledge respect for human dignity definition ○ protecting the interest of the person and their physical, pscyh, culutral integrity respect for free and informed consent ○ free consent? no means of coercion no penalties/sanctions for withdrawal ○ informed consent? adequately informed so they can udnerstand ins and outs ○ autonomy protection & assessment assesses whether ppl can provide consent and protects vulnerable ppl ○ information: completeness & concept of deception completeness: should be accurate and understable deception: withholding info for the purpose of the study, can undermine informed consent respect for justice ○ distributive justice there should be no discrimination between research participants ○ objectivity during selection while selecting participants, remain objective  be cautious not too recruit too many vulnerable bc theyre easier to recruit bc your study will end up biased respect for vulnerable persons ○ a vulnerable person is anyone under 18 or over 18 and mentally unfit ○ 3 ways to respect them dont ill treat or disciminate offer charity (special protection, compassion and equity) get substituted consent respect for privacy ○ confidentiality any personal info disclosed during the research should not be dissemintated and be careful when disclosing results ○ anonymity responses/results related to a person or their identity should be protected level of protection depends on? the subject of the study maintaining anonymity benefits results are valid and higher participation rate participants are less likely to lie advantages and disadvantages ○ balance both the benefits should outwight the risks value the well-being and respect of participants by providing informed cosent ○ reduction in disadvantages eliminate or limit any risks to avoid maltreatment and provide help when needed ○ optimization of the good offer charity or personal/social benefits to participants consent forms form ○ can be oral or presumed and provides information + circumstances about the consent content ○ inform about the nature and purpose of project ○ inform about the nature of participation and that its voluntary ○ tell them benefits and risks ○ ensure confidentiality and anonymity ○ get emergency contacts ○ signatures considerations in research ethics plagirism ○ should not copy others  conflict of interest ○ researchers/ethics committee ppl shoudl not be involved fraud use of research data ○ this can happen if they use incompletely published results/fabricated data contribution of authors ○ eveyrone should contribute equally research ethics committees (REBs) who are they ○ professional evaluare research protocols involving humans, doesn’t matter if its funded or not purpose ○ they want to protect dignity of participants ○ decide if timing makes sense ○ decide whether a research ethics certificate is awarded ○ deposit of funds is based on sine qua none condition Week 6: Qualitative Research what is it ○ doesn’t seek to measure/quanitfy data but collects verbal data to allow interprative approach ○ tries to discover, explore and describe phenomena and their essence which is not often understood till the end of the study connected to social science/sociology of health research methods ○ sees subjectivity and that reality reflects different perspectives ○ similar to interpretive and inductive type reasoning (specific to general) ○ phenomena are observed and described as they occur 3 purposes ○ explores facts, social, cultural phenomena better understanding of those that aren’t understood ○ describes problems not well known 4 types of qualitative research ○ phenomenology ○ ethnography ○ grounded theory ○ case study data collection methods ○ structured interview ○ non-structures interview  done face to face, phone, individual or in small groups ○ participant observation ○ non-participant observation observed through field notes/recording ○ this is a circular process → number of participants is not decided but depends on the data collected (saturation) comparsion of qualitative vs quantitative qualitative quantitative focus of study social phenomena biomedical issues purpose the existence and significance of measures and quantifies variable phenomena outcome understands context and study subjects causal relationships between variables assumptions create assumptions test of assumptions type of reasoning inductive deductive data collection methods interviews individual ○ structured there is a guideline of questions strictly followed ○ semi-structured  questions are open ended ○ in-depth 1-2 points are discussed in detail and questions are initiated from what the interviewee says collective ○ a focus group is interviewed consensus method ○ nominal 8-12 participants are selected, the main points are discussed then participants rank the subjects on a scale then a hierarchy is created ○ delphi method there are no limits to number of participant sor location because all discussions are done my mail or internet observations ○ direct researcher makes observations through records or notes ○ indirect (document analysis) researcher analyzes someone else’s observations through document analysis phenomenology what is it ○ studies experiences, identifies the nature of the phenomena and the meaning people give to them ○ focuses on the perspective of people and the meaning people attach to phenomona (to explore, clarify it) ○ used most often data collection method ○ researcher conducts in-depth interview to collect first-hand perspective of participants ○ data is interpreted and structures using different stratgies to reveal different themes limits ○ no limit to what the participant says ○ the participant’s POV is unique ○ may not be transferrable bc its too unique ethnography what is it ○ observes, describes, analyzes trends and lifestyles of a culture or subculture ○ experiences are considered in a cultural/environmental context (more extensive) target population ○ a culture can be any group with a common trait (religion, sex, age) purpose  ○ understanding a group, its ways of living and beliefs by investigating social interactions, behaviours and perceptions rooted in anthropology data collection method ○ researcher lives with the study group for a long duration ○ this presents a detailed picture of their social life, rituals, symbols and perspectives of the group limits ○ time and costs involved with living with the group ○ data might be too subjective grounded theory what is it ○ describes the meaning people give to social interactions and how they interpret symbols ○ focuses on psych and social processes and builds a theory based on these empirically limits ○ theory might be too complex to understand ○ requires a lot of data to analyze fully case study what is it ○ analyzes a particular situaiton to studiy the evolution and manifestation of a phenomena of interest limits ○ there is not a lot of data available ○ might be too specific to be transferrable ○ time of study might be too short research methods depends on: ○ personal orientation of researcher ○ the nature of the research question research methods ○ qualitative- to generate ideas and theories ○ quantitative- to verify theories Week 7 + 8: Quantitative Research Designs 2 branches  descriptive 1. tries to monitor state of disease of public’s health and evaluates success of intervention programs 2. helps generate a hypothesis about what causes the disease identifies and counts cases in a population according to person, place, time and conducts somple studies 3. types case reports/studies cross-sectional study ecological study analytic/scientific 1. monitors success of intervention 2. helps verify hypothesis about causes of disease compares groups systematically to determine association 3. types case-control cohort experimental clinical trial both have to consider if they are gonna be retrospective or prospective 8 steps in a general approach to outbreak investigation 1. formulate case definitions 2. conduct case conformations 3. establish background rates of disease/cases 4. examine descriptive epidemiology of outbreaks 5. generate and test hypothesis about the causes 6. collecting data and testing environemnt 7. implementing control measures 8. disseminate informaiton with press and public descriptive case report vs case series ○ case report- a detailed report on one person with a new disease or unusual symptoms ○ case series- a detailed report on a group of people with a common new disease or unusual symptoms ○ strengths simple, fast, inexpensive reognzing and describing new/old disease detecting drug side effects understanding disease mechanisms provides info to generate a hypothesis ○ limit  there is no comparsion group to compare the data to ecological studies ○ examines rates of exposure and outcome of disease in a population/group rather than one person group can be a country, state, ○ strengths cheap, fast analysis is easy (uses correlation coefficient, linear regression) good for early stage of knowledge wide range of exposure data is available from different places (good for international lvl studies) you can study ecological relationships ○ limits ecological fallacy- group-level association may not be applicable to an individual can’t adjust for confounding factors cross-sectional studies ○ examines relationships between exposure and outcome in a particular time and used in many government surveys ○ strengths cheap and fast highly generalizable to public good for immutable, long-term and historical exposures (no risk of temporal inference) ○ limits doesn’t account for confounding factors if the exposure is changeable, then we can’t associate it with the disease temporal inference- may be biased toward long duration cases (only which are prevalent at the time of the study) longitudinal survey ○ collects information from the same participants at different points in time ○ can be retrospective or prospective mostly about a particular population common in Canada: Canadian Community Health Survey analytical studies case-control study ○ a group of people with disease (cases) are compared to a group of people without it (controls) and other is no intervention, only observation ○ mostly retrospective  ○ ways to collect data questionnnaires medical record reviews interviews databases ○ how to minimize confounding factors matching participants based on age, gender ○ guidelines for selection controls have to be a part of source population controls have to be selected without knowing their exposure status to prevent misclassification bias the case and control have to have equal time at risk for disease after an event of a control, they can only be a case now ○ hospital-based cases and controls have to from the same hospital controls have to be hospitalized for a reason not related to the exposure the location of the recruitment depends on what disease we are focusing on ○ steps take a sample of diseased vs non-diseased identify exposure calculate odds ratio ○ 4 limits there may be confounding bias recall bias- diseased may remember exposure better timing of when exposure happened is not certain prevalence of disease in sample may not match the population case-crossover study ○ this is for measuring the short-term effect of an exposure ○ a period prior to the event = exposure period ○ this is compared to a previous period when the person was exposed but didn’t experience the event = reference period ○ what is it useful for? studying if there was an increase in exposure right before the event and what caused it it limits confounding variables if the research is done on the same person these exposures are usually modifiable ones prospective cohort study ○ a study that compares a group of people who have been exposed to unexposed and collects data on them over a period of time, without any intervention ○ can be prospective or retrospective ○ methods of collecting data questionnares interviews  medical record reviews ○ steps take a sample of non-diseased identify exposed vs non-exposed follow them over time to identify who develops disease calculate relative risk ○ limits we don’t have information about all the confounding factors cross-sectional study (photograph of a phenomenon) ○ studies the association between exposure and disease at a point in time ○ not prospective or retrospective ○ limits hard to know if the exposure or disease came first confounding factors ○ a variable that is associated with both the risk of disease and exposure ○ it could cause errors in the conclusion of associaitons if not taken into account ways to eliminate ○ matching ○ stratification ○ randomization ○ multiple regression (neutralizes confounding factors) ○ Restriction Week 9: Experimental Studies Randomized Clinical Trial ○ a group of interest with a condition are randomly split into an intervention or control group and there is a follow up done control group- could receive no meds/an old one intervention group- could receive treatment/new ○ features parallel- there will be a group of ppl with different health conditions (ex. diff cancers) they receive the same medication to observe it will be beneficial in one group over another comparative- one group receives drugs and one receives a placebo people will have the same health condition single-blinded study participants will not know if they are in control or intervention groups reduces any biases because they feel they’ve received better medication and drop out of the study double-blinded study  participants and researchers don’t know which group is the control or intervention group the way the researcher administers might influence the participant’s compliance (to respectfully study protocol) there are 1-4 phases clinical trial phase 1- studies done in animals to see side effects in a lab phase 2- done in humans phase 3- verifies dose that can be tolerated and if its useful phase 4- done after several years to verify any health effects of medication ○ randomized trial studies of cohorts in which the researcher allocates the exposure and their decision is based on the needs of the study people will be randomly allocated into either group ○ objectives RCT: evaluates the effect of drug/intervention for a disease in a population prevention trials: assess the effect of a prevention measure on a disease occurrence in a non-diseased population vaccination vs no vaccination to see who develops disease screening test: assess the effect of a screening program for complications of a disease in a non-ill population randomized into screened regularly vs not screened unless abnormal they are followed over time and see who can be diagnosed earlier things to note: ○ many clinical trials are not randomized there is a selection bias issues ○ randomized trials are not only done on patients but also on animals steps ○ recruit sample with no disease ○ assign drugs to one group, placebo to another ○ monitor over several years to assess the outcome ○ calculate relative risk benefits ○ good internal validity there is less selection, confounding and missclasfiicaiton downsides ○ external validity results might be too specific to be generalizable ○ very expensive to conduct ○ sometimes impossible due to ethical issues we cant ask if fracture increases depression Types of Experimental Studies  Before & After Trial with Control Group ○ we recruit a number of people and assess the outcome in one group vs the other ○ one group receives medication and the other placebo ○ measurement of the dependent variable is done before and after the intervention ○ benefits internal validity is good bc of randomization and there is no selection bias & confounders are balanced ○ limits external validity is weak if study population is too specific and results cant be generalized After Only Trial w Control Group ○ one group receives a drug and the other receives a placebo randomly and we compare influence on outcome (biomarkers, survival, recurrence of disease) ○ measurements are only taken after the intervention Factorial Design ○ combines different treatment regimens ○ there is more than one treatment and various doses ○ we need several comparison groups so a higher sample size Crossover Trial ○ people in group 1 receive medication 1 and we assess outcome (BP) ○ after assessing outcome, we give them medication 2 and we asess outcome again ○ people in group 2 receive med 2 then med 1 and assess outcome after each step there might be a week between each medication delivery so there is no crossover effect this lets the first med effect washover and they return to baseline rate Quasi-Experimental Study researcher controls what, when and how the subjects are distributed (no randomization) control groups are not randomized benefits ○ explain relationships between variables and why events happen ○ provides a good way to see cause-effect and predict phenomena downsides ○ there might biases if we are selecting participants and eliminating variables we don't want Before & After Trial w One Group ○ only one group’s outcome is assessed before and after the intervention ○ intervention (researcher has no control over it) could have 2 levels and represents how much time has elapsed: yes or no ○ ex: smoking ban policy rate of cancer in 2014 vs 2024 to see if the outcome was impacted  used to study policy changes and their effect on disease at the population level or individual ○ downsides there is no comparison group so it's vulnerable to internal validity (maturation, historical factors, instrumental, statistical regression) causal relationships are limited Before & After Trial w Two Groups ○ 2 groups are randomly assigned to treatment or no treatment ○ outcomes are measured before and after treatment ○ there is some control ○ downsides internal validity is weak because of the interaction of selection with historical factors and maturation After Only Trial w 2 Groups ○ 2 groups are distributed in a non-random manner and evaluated after treatment ○ downsides internal validity is weak bc of selection bias no pre-treatment assessment is a barrier Interrupted Single Time Series ○ a single group is assessed repeatedly before and after treatment ○ trends of disease are related to time intervals ○ downsides internal validity is weak bc of historical factors ○ benefits multiple pre and post-tests provide control and most other factors are controlled fairly well Multiple Interrupted Time Series ○ 2 groups of subjects that are not equal are divided (not randomly) ○ several measurements are done before and after ○ similar to interrupted simple time series ○ downsides internal validity is weak bc of selection bias and historical factors Mixed Design combines quanti and quali methods to describe theory why would you choose it? ○ choice is driven by purpose and question of research: if one component is exploratory but another requires data analysis depends on how complex phenomenon is and the limits of other methods 3 top reasons to use it ○ need for quali data to interpret quanti results ○ need quanti data to interpret quali results  ○ need to explore (quali) and measure (quanti) explanatory sequence ○ quanti then quali we want to get enough info from quanti and better explain relationships of a complex phenomenon through quali downsides can take a lot of time bc of its depth there could be unexpected results sequential exploratory ○ quali to quanti ○ you need to first understand phenomena in-depth first then quanti to explore relationships, confirm them and generalize them ○ downsides time-consuming simultaneous with triangulation ○ both done at the same time to explore complementary aspects of the same question and there may be similarities, differences or combinations ○ downsides time consuming for both researcher and participant transformational simultaneous ○ either both or just one ○ the objective is to support perspective change and empower participants to offer a benefit ○ limitation not a lot of experts in this type of design and matching results from other studies is a challenge interpretation of results with the goal of advancing equity or addressing social justice concerns. strengths ○ answers complex questions that you need a lot of data for ○ provides a lot of data limits ○ time-consuming ○ not a lot of experts in this field ○ difficult to get published Historical factors are external events that occur between the pre-test and post-test (or during the intervention) that may affect the outcome.  These events influence the dependent variable and may be unrelated to the intervention itself, leading to incorrect conclusions about the intervention’s effectiveness. ​ ​