Research Methods Post-Midterm
Outbreak investigation Steps:
1. Confirm existence of an outbreak 2. Verify the diagnosis 3. Prepare for field work 4. Construct a working case definition 5. Find cases systematically and record information - Conduct surveillance using case definition - active surveillance - ex. review medical records, try to identify all cases - Collect data on case-patients - Create line listing 6. Perform descriptive epidemiology 7. Develop hypothesis 8. Evaluate hypotheses and execute additional studies as necessary 9. Refine hypotheses and execute additional studies as necessary 10. Reconcile with lab/environ studies 11. Implement control/prevention controls - Implement as early as possible - Usually targeted to one or more - eliminate or treat the source, interrupt transmission, reduce susceptibility - Work with regulators, industry and health educators to institute control measures - Create mechanism to evaluate both short and long-term effectiveness of control/prevention measures 12. Initiate or maintain surveillance - Monitor for additional cases - Evaluate for effectiveness fo control measures 13. Communicate findings - Reasons - summary of investigation, documentation of findings and action, justification for recommendations, documentation for legal purposes - Methods - trip report, PH bulletin, Peer-reviewed journal, oral presentation
Steps to Sampling plan development
1. Define objectives 2. Identify environmental population of interest 3. Obtain background data - physio-chemical characteristics, weather info, etc - usually done w/ a lit review 4. Identify measurement techniques - may influence sample collection techniques 5. Design field sampling plan - how many samples, where, when - consider sample strategies to be used, consider statistical analyses to be later applied
Pillars of empirical control in environmental research
1. Physical control: physically controlling data collection so as to reduce potentially distracting factors 2. Assignment control: creation of at least one of the variables in study so as to test its effect on other variables 3. Statistical control: analysis of any variable in the study that may have an effect on the variable of interest
Clinical Research: Phase One
20-100 patients, Study length: several months Purpose: Safety and Design - Determine number needed to treat - the amount of people you need to treat before you see a beneficial effect - small number of people is what you want - Number needed to harm - number of people you have to treat before you see harm - large number of people is what you want
Clinical Research: Phase Three
300-3,000 patients, Study length: 1-4 years Purpose: Monitor adverse effects
Experimental
= assignment control + physical control + statistical control
Non-experimental
= physical control + statistical control
Stages of Studies: Stage Three
Clinical Research Four phases
Preventing trials
test alternative ways to prevent disease or disease recurrence
Treatment Trials***
test new treatments, drug combinations or surgery/radiotherapy - Most COMMON RCT, used by pharmaceutical companies to test new drugs
Phases of PAR: Phase II
Collect baseline data
Stages of Studies: Stage One
Discovery and Development - Safety is central to this stage! Is the drug safe? Side-effects? Dosage?
Spatial systems
Do not change much with time - Ex: small homogenous physically isolated field contaminated w/ persistent pesticide can use a truly random equidistant selection of sample spots
Stage Four- FDA Reviews
(comes after phase three) - Review drugs and reports of problems - Occurs before marketing of drug
Availability of Existing information
- "Does the data I need already exist via another source? - If yes, you may not even need to conduct primary research, secondary research can be done using available data
Phase II Trials - Participants increase!
- 100-300 patients, evaluates the efficacy and further assess safety - Drug must pass these studies before entering Phase III - Researchers conduct these
Focus Groups: What is a focus group?
- A focus group is the use of group interaction to produce data and insights that would be less accessible without the interaction found in a group - Use focus groups to gather info from several people (8-12 is ideal) at one time - Encourage people to talk to each other, ask questions and exchange anecdotes - Group processes can help the subjects to explore and clarify their views in ways that a one-on-one interview can't - Groups help tap into different forms of communication that people use in day to day interactions (jokes, anecdotes, teasing and arguing)
Study Designs
- All epidemic studies classified as either experimental (interventional) or non-experimental (observational) - Experimental studies use a placebo which should look and taste like the drugs the other patients are getting
Maintaining Confidentiality:
- Be sure your interview is held in a private place - Do not discuss any patient's history or comments with another person outside the project - Be cautious about acknowledging your subjects in the community. If you see the subject at IGA the next day, don't recognize them unless they approach you - Be cautious about sharing information or giving your opinion - Refrain from giving medical advice or suggestion unless you are a licensed doctor: DO: offer to get the subject help if they ask. Be sure you have a contact ahead of time, that you know will help if needed
Principles of Equipose
- Benefits of a drug should not be known, there should be genuine uncertainty whether or not the outcomes will be beneficial - It is unethical knowing if one group will benefit over the grou - Important principles in creating clinical trials
Disadvantages of PAR
- Big commitment of time and effort - Process can be messy and take a long time - Money and an organizational structure is needed - Research goals are not always feasible
Phase I Trials - Specific for Drug Trials
- Clinical pharmacologic studies - toxicology, pharmacokinetics - Drug must pass these studies before entering Phase II - Researchers conduct these
Challenges of Photovoice:
- Close examination of an issue of concern can cause negative feelings...such as depression - Participants continuously make choices about: What they select as subject matter for their photos What is not influenced in their photos - The choices influence the research findings
Superiority Studies
- Compare new medications to determine if they are MORE effective than currently available interventions
Equivalence Studies
- Comparing new and cheaper medication to older and more expensive medications to see if intervention is as effective and equivalent - Also called a Non-Inferiority study
Statements that help get more info out of the interviewee:
- Could you tell me more about that? - That must have been unusual - tell me more? - If I were in that situation I would be afraid... - Some people have said... I want to hear more about that but we have to get through this part in the time limit. I'll make a note of that and can you tell me the rest later? (This helps you keep the interview on track, in case you get a talker)
Quantitative methodologies
- Cut and dry results - If you wish to collect quantitative data you are probably measuring variables and verifying existing theories or hypotheses or questioning them - Data is often used to generate new hypotheses based on the results of data collected about different variables - Many questions that need to be asked, a survey w/ quantitative questions may be best, quickly to answer - Online survey tools can be used, interactive and prevents survey fatigue - Questionnaires: logical and easy option as a way of collecting information from people, difficult to design, response rate is nearly always going to be a problem, sample limited by how long the questionnaire is and the types of questions asked, close or opened questions can be used, layout can be a problem, scale responses may not be accurate, experts are needed to help set up questionnaires, respondents shouldn't have to pay to return post the questionnaires, reduce response rates, lengthy ones reduce response rate also, questions should be clear, reliable methods should be used to collect and manage data, questionnaire can be set up to be read by an optical mark reader; saves time
Performing descriptive epidemiology
- Describe clinical characteristics of the illness - Describe demographic characteristics of those affected - Identify or infer population at risk - Provide clues to etiology, modes of transmission - Guide interventions
Goals of science
- Description: describe and categorize basic phenomena - Prediction: predict phenomena based on past and present observation - Explanation: explain described and predicted phenomena (explain observed patterns), infer causality - to be able to explain, to draw valid conclusions on observed phenomena - establish causality, Empirical control over phenomena is needed - any method of increasing ability to establish causality from empirical data - Control: try to control the phenomena in order to bring about desired changes
Environmental sampling key steps:
- Develop sampling PLAN (when, where, how samples will be collected) - Collect samples - Preserve samples (for transport and storage) - Analyze samples
Consort Database
- Every clinical trial MUST be registered here to increase transparency - Displays every clinical trial and what stage it is in - Guidelines/Flow-Chart - Enrollment-Eligibility and Exclusion, Allocation-Allocation of intervention to which group and Reasons, Follow-Up-Reasons for loss to Follow-Up and discontinued intervention, Analysis-Analyzed versus Excluded from Analysis
The Interview Procedure:
- Explain the study - Explain the consent form - State time it takes - conditions - If subject agrees to the interview, explain any compensation they will receive - Ask for permission to record the interview - Thank constantly for their time... before, during and after - Be prepared: have all your forms ready...bring extra pen and paper - know where everything is...be organized, know where the bathrooms, water fountains and telephones are, just in case your subject needs them
Put PAR to work:
- Form a research team of interested community members - Reflect and decide upon research question that is important to community - Plan a research approach that will answer that question - Collect data using the chosen research approach - Analyze the data to generate research findings - Share the research findings with community members and stakeholders
How many samples should be collected? Depends on:
- Homogeneity or lack of it in system - Countries' regulations that establish sample quantities per area
Data Collection and analysis:
- How are you going to collect your data? Who is your target sample? - How will the data be analyzed? Will you use a software program? - Where will the data be stored? Password protected computer? Locked filing cabinet?
Sample Size
- How big does your sample need to be to make a confident conclusion based on your results? - Larger sample size, that will eliminate time-consuming, smaller-format methods such as focus groups and in-person interviews
Timing
- How quickly do you need results? - Can time consuming methods of data collection be used? - Time also affects the screening criteria for participant selection
Focus Groups: Benefits
- Identify shared and common knowledge - Often used in cross cultural research and work with ethnic minorities - Participants can provide mutual support in expressing feelings that are common to their group but they consider to deviate from mainstream culture - Actively facilitate the discussion of taboo topics because the less inhibited members of the group breaks the ice for shyer subjects - Participants may develop certain perspectives as a consequence of talking with other with similar experiences NOTE: she said...focus groups are more challenging.. Some folks may feel uncomfortable. It can be difficult to stay on topic. Having a note taker in the room may compromise confidentiality and cause some of the subjects to not want to chime into the conversation
Tips for developing questionaires
- Identify your research questions - Identify your sample - Draw up a list of appropriate questions and try them out with a colleague - Pilot them - Ensure questions are well laid out and it is clear how to 'score them' (tick, circle, delete) - Ensure questions are not leading and confusing - Code up the questionnaire so you can analyse it afterwards - Gain permission to use questionnaires from your sample - Ensure they put their names or numbers on so you can identify them but keep real names confidential - Hand them out/post them w/reply paid envelops - Ensure you collect as many as possible - Follow up if you get a small return - Analyze statistically if possible and/or thematically
5 Key concepts of Photovoice:
- Images teach - Pictures can influence policy - Community members ought to participate in shaping public policy - Influential policy-makers must be audience to the perspective of community members - Photovoice emphasizes individual and community action
Advantages of PAR:
- Increases feeling of participant ownership of process/program. This allows for sustainability of the changes being made - High possibility of efficiency - Produces practical outcomes - Increases likelihood that data will be used - Increases likelihood that resulting program or intervention will meet the needs of the stakeholders and be culturally appropriate - Participants develop skills and confidence. They gain knowledge and information and thus become empowered - Focus on positive change
Analysis of Results
- Kaplan-Meier Survival Analysis - Intention to Treat - Explanatory
Phase III Trials
- Large scale effectiveness and safety - RCT occurs here - Drug must pass phase III to be approved by FDA and licensed for marketing - Researchers conduct these
Recruitment and Retention of Study Participants
- Major challenge is to recruit eligible participants who are willing to be randomized - Without a sufficient number of volunteers, study won't be able to yield statistically significant results
Quantitative research
- Measures level of occurrence - Asks "how many" - Studies actions - Is objective - Provides proof - Is definitive - Measures levels of actions and trends - Describes
Summary of PAR
- Method used to involve community residents, clients and other constituents in social change oriented research - Role of Researcher - Consultant; Partner with participants - Participants work with a facilitator to identify a community problem, develop research methodology, collect data, and analyze findings - Main Purpose - Social Change - Participants (and researcher) achieve learning within larger group - Participant issues, actions and learning highlighted/No Expert - Researcher works with "participants" - Actions taken through process - action is incorporated into research itself - The data is used to make recommendations for resolving the problem - Participants advocate for funding, legislation, or government action to adopt the findings - Emergent property: creativity - Outcome - Alleviate Oppression or improve community or service quality
Phase IV Trials (also called Post-Marketing Surveillance)
- Monitoring new agents as they become more widely used - Carried out by physicians
Levels of Blinding - Reduces information and selection bias
- Non "Blinded-Study" - participants, researchers and biostatisticians know who's getting what - Single-Blinded study - only participants don't know who's getting what - Double-blinded study - participants and researchers don't know who's getting what - Triple-blinded study - participants, researchers and biostatisticians don't know who's getting what - gold standard for RCT and studies ****
Outbreaks are usually recognized by:
- Observation of a single event or cluster of events by a clinician or lab personnel - Report of one or more patients - Routine surveillance activities - Ex. Yambuku Mission Hospital Sept 19, 1976 - 17 patients and one midwife died since 1st of Sept, Symptoms: fever, vomiting, abdominal pain, bloody diarrhea, hemorrhaging rapidly progressing to death, illness appears to be spreading to hospital staff and areas around hospital
Challenges and Solution for PAR:
- PAR efforts take more time than other types of research. Gathering data from multiple sources, delegating responsibilities and reaching consensus require greater amounts of time - Pacing the activities helps - Communities that don't have strong leadership might need a facilitator to coach them through the process - Funding is needed for things like conference calls, meetings and stipends (if any are given) - Members must be willing to share in the work and rewards - Consumers must be strongly represented and heard
Key features of PAR
- Participants examine their own knowledge and understanding - The research is self-directed rather than others directing it - Participants are co-researchers...it is up to the Public Health worker to train those in the community to do basic research - Practical - Critical - Emancipatory - Reflexive
Assumptions about the process:
- People who experience the problem are in the best position to conduct research on the issue - All people can learn basic research skills - Participants can establish equal partnerships with researchers that can be used to address community problems - PAR related activities help empower member of the powerless groups. Participation in the PAR process is a critical component to community interventions
Considerations when sampling
- Perfect lab and statistical analyses cannot rectify problems caused by faulty sample collection - Handling must not affect chemical/biological composition of samples - avoid contamination of samples, adequate storage, preservation, transport, sampling must be done in function of analytical requirements in mind
Randomized Controlled Clinical Trials *** Gold Standard - Triple Blind RCT study
- Prospective/Longitudinal study, exposure occurs before the outcome - Measures incidence through relative risk/risk ratio and NOT odds ratio! - Odds ratio used in case-control or cross-sectional studies - Patient are randomly assigned to an intervention/treatment group - control group receives a placebo or a standard treatment method, intervention group receives drug or treatment being tested
Qualitative research
- Provides depth of understanding - Asks "why" - Studies motivations - Is subjective - Enables discovery - Is exploratory - Allows insights into behavior, trends, etc - Interprets
Advantage of an in-depth interview:
- Provides much more detailed info that what is available through other data collection methods...such as a survey - Provides more relaxed atmosphere in which to collect info
Basic Tenets of Qualitative Research
- Purposive rather than random sampling - Data emerges as the research progresses - Subjects plays role in interpretation - listen carefully to the responses - Typically not generalizable- because each response is unique to the subject - Interviewers/moderators/data analysis require training - Should contain measures of reliability and validity - Inter-rater reliability is a key factor - Analysis is systematic
Randomization in RCT
- Randomization eliminates selection bias - Examples: simple randomization, stratification
Kaplan-Meier Survival Analysis
- Relative risk and Hazard Ratio used - P-value more than alpha 0.05 - fail to reject null hypothesis - P-value less than alpha 0.05- reject null hypothesis
Expressing Results of Randomized Trials
- Risk of developing disease or complication - Efficacy can be calculated through rates of developing disease
Good interviews are Sales
- Selling the subject on participating - Communicating your interest in the subject's response - Creating a safe environment for disclosure
Photovoice participants:
- Share ideas and concerns about their experiences and their community - Take on the role of photographer where they have the responsibility of capturing photographic evidence - Assume the role of co-researcher
Construct a working case definition
- Simple objective criteria - Develop a case definition using: symptoms or laboratory results (clinical), time frame (1-2 months preceding first known case), Affected population (Person) (any), Location (place) - Loose (sensitive) vs. Tight (specific) case definition - ex. Gastoenteritis - sensitive: nausea or diarrhea, specific: laboratory-confirmed Salmonella, Shigella, Norovirus, etc - Categories of certainty - confirmed, probable, possible
Comparing Effectiveness
- Some trials compare new therapy to placebo - Some trials compare new treatment with an older accepted treatment to determine if new treatment is superior
The Research Onion
- Start with a research philosophy: develop an approach - inductive or deductive - Develop a research strategy - Make a research choice - can be mono-method or a mixed method - Time Horizon: longitudinal or cross sectional studies - Data Collection and analysis
Types of Interviews:
- Structured - Semi-structured - Informal
Design of a Randomized Controlled Trial
- Study Population - Selection by defined criteria-Enrollment - Study Sample - Informed consent needed from those invited to participate - Randomization into Two Groups-Allocation - Control (nonexposure), Treatment Group (exposure) - Follow-Up - Analyzed (End up with 4 groups) - Control-developed outcome and No outcome, Treatment-Developed outcome and No outcome
Develop hypothesis
- Subject matter knowledge - known sources, vehicles, transmission modes - Review descriptive epidemiology - what would account for most - Outliers - unique exposure opportunities - Talk to case-patients - what do they think - What do local health officials think
Methods used for PAR:
- Surveys - Observations - Interviews - Focus groups - Community meetings - Photovoice
Ending the interview:
- Thank the subject - Explain follow up, if there will be one - Prepare your notes on your interview immediately - At the end of your shift, make copies of all forms - Turn in forms to project manager
Characteristics of PAR:
- The "problem" originates within the community - The research goal is to fundamentally improve the lives of those involved through structural transformation - The people of the community or workplace are involved in controlling the entire research process - Participatory research plays a role in enabling by strengthening people's awareness of their own capabilities - The researcher with specialized training may be outsiders to the community, but are committed learners in a process that leads to militancy (fighting for change) rather than detachment
What is an interview?
- The art and science of exploring an individual's thoughts - The subject being interviewed is the expert - Qualitative method - Skill learned by doing - practice makes perfect
Smith's (1997) Five-Stage PAR Model
- The researcher collects information about the community and its problems - The researcher uses dialogue to engage the group in a process of problem-identification - Group members develop an understanding of the social, economic, and political context or origins of the problem. They identify questions that they want answered - The group identifies theories about problems, designs data collection methods and generates possible solutions to address the problem - The group takes action - PAR is considered an alternative to quantitative research in which the researcher "know all"
What is the purpose of PAR?
- To minimize power differences between researchers and participants/constituents - To increase the knowledge of participants - To reduce harmful effect of oppression by involving members of powerless groups in the construction of knowledge - To facilitate action to address social problems
Condition for successful PAR project:
- Trusting relationships among members must be developed in order for a consensus about project goals, data collection, methods, and an analysis of findings, and recommendations can be reached - Training about research methods, data collection and analysis must be provided for the participants - Establishing a good organizational structure to support the work team - The provision of strong administrative support and adequate resources for the project - A skilled facilitator to coordinate the process
Qualitative methodologies
- Used to understand meanings, look at, describe and understand experience, ideas, beliefs and values, intangibles such as these - Focus groups: captures consumer insights and open-ended reactions - Highlighting tools and heat maps: consumers to indicate areas of strong like or dislike - Face to face interview: enable discussion with human subjects; you can take notes, tape interview or reply on memory; use open or close ended questions or both - Close ended questions gives you an opportunity to give your respondents some options and makes data more quantifiable. THIS LIMITS THE RESPONSES YOU GET - Open ended questions elicits an almost endless number of responses. Gives you a better idea of one's feelings and views. Difficult to quantify. - If you decide to use interviews: identify your sample, find appropriate questions, do start w/some basic closed questions (name, etc), don't ask leading questions, try them out w/ a colleague, pilot them, then refine the questions so that they are genuinely engaged w/ your research object, contact your interviewees and ask permission, explain the interview and its use, carry out interviews and keep notes/tape, transcribe, analyze results and related these findings to others from your other research methods
Evaluates Hypothesis
- Usually analytic epidemiology - cohort study and risk ratio, case control study and odds ratio - Otherwise, compare facts with hypotheses
Importance of environmental sampling
- We not have the technology to scan entire planet or an entire subsystem in the plant to get a complete analysis - One small sample does NOT reflect properties/characteristics of the entire area - because environmental systems are usually VERY HETEROGENOUS, RARELY HOMOGENOUS
Statistical Significance
- When a result is statistically significant, it means that it is highly unlikely that the result occurred by chance alone - Looking for general statistically significant data or definitive and clear? - Definitive, clear, black and white data requires quantitative data and a large sample size - Important for extrapolating results
Tips for good data collection:
- When in doubt, refer to the research questions -Pre-test the interview/focus group guide - Revise the process or guide if it's not working - Talk to the right people (target audience and key stakeholders) - Use trained interviewers/moderators
When are in-depth interviews appropriate?
- When you want detailed info about a person's thoughts and behaviors or want to explore new issues in-depth complex subject matter and knowledgeable respondents - In place of focus groups if the potential participants many not be included or comfortable talking openly in a group - To refine questions for future surveys of a particular group - Highly sensitive subject matter, e.g. a study among women who have had an abortion - Geographically dispersed respondents: a study among policy makers in 8 countries - Peer pressure: a study of male teenagers to explore their attitudes about sexually responsible behavior. You don't want to put a bunch of teen boys in a group talking about sex...
Representative sample:
- sample that collectively reflects properties/characteristics in a study population - is representative of the environmental characteristics AT TIME of sampling - Multiple small samples used to obtain information about composition of entire study area
Early steps of an outbreak investigation
--> Do not have a particular order Confirm the existence of an outbreak - Epidemic vs. endemic - Outbreak vs. epidemic Verify the diagnosis - Will make the decision about conducting field investigation - Prepare for field work
What do we do if?
A person decides not to participate half way through? - Go back to informed consent. You can ask them why, but you can't force them to participate The respondent becomes too ill or tired to complete the interview? - Get help if they are ill - If they are too tired. Try to reschedule for another time The subject appears "off"? The subject refuses to answer some questions? - Try to rephrase the question, or skip it. Try to come back to it later The subject ask for medical or psychosocial help (or appears to need it)? - Offer to get them in contact with someone that can help them...don't you do it Practice your interview several times before you go live. Use your family and friends to help hone your interviewing skills
Science
A personal and social endeavor in which ideas and empirical evidence are logically applied to create and evaluate knowledge about reality
Phases of PAR: Phase III
Action plan
CBPR:
Addresses an issue, concern or topic raised by the community - For example: - (health) Why are there high rates of cancer? - (education) Why do children drop out of school? - (social) How can we address housing needs? - (economic development) Should we develop a wind farm? - (environmental) How can we make our neighborhood safer? - (culture) what will happen to our knowledge of the water if it is polluted? Community (co)-led, designed, conducted, disseminated - For example: - The project is headquartered in the community - Community members decide what kinds of methods work best - Community members carry out the research - Community members are co-authors on articles - Reports are shared with community before other scholars Is aimed at change - For example: - (structural) change power relations or institutions of society - (behavioral) change lifestyle choices Creation or production of knowledge based on systematic study - For example: - (hypothesis-driven) women who do not have a regular primary care provider are more likely to develop cervical cancer - (grounded-theory) interviews with women discovered that experience with sexual abuse prevented them from seeking gynecological care
Explanatory
Analyze according to treatment actually received, regardless of randomization - Did treatment make a difference even if randomization was no considered?
Health-related quality of life trials
Approaches aimed at improving comfort and quality of life through pain relief, improved patient function or boosting well being
Data analysis software
Atlas ti, QSR, Nvivo, MAXqda, HyperRESEARCH, Dedoose
Judgemental
Based on prior knowledge of temporal and spatial variations in the system
Mixed methods
Can be 2 different quantitative approaches, 1 quantitative and 1 qualitative or 2 different qualitative approaches
Photovoice:
Community members share their words and photos as a way to reach decision-makers and implement positive change in their home communities - A PAR method that employs photos and group dialogue as a means for marginalized individuals to deepen their understanding of a community issue or concern - The visual images and stories are the tools used to reach decision makers - Photovoice equips people with cameras so they can create photo evidence and symbolic representations to help others see the world through their eyes - The aim is to improve conditions by making changes at the community level
Health economic evaluation trial
Compare cost data between similar treatments, healthcare costs
Evidence-Based Review of CBPR
Context for the review: - Health disparities persist - Knowledge is not always translated into appropriate action in communities of need - Community-based participatory research offers an alternative approach to translating research into practice
General Principles of Exclusion Criteria/Eligibility Criteria
Defines study population so results can be generalized to entire population but sometimes it can be difficult to generalize - Example: ACE inhibitors vs. Ca+ channel blocks - African-Americans, AfroCaribbean and British Africans can't use ACE inhibitors. They do not work as effectively as Ca+ blockers so sometimes studies covering this difference cannot account for the entire population only for this particular group Deviations from criteria are not allowed because they can jeopardize the scientific integrity Criteria include age, sex, type and stage of disease, treatment history and other medical conditions
Phases of PAR: Phase I
Develop community focus
Types of Clinical Trials
Diagnostic trials Screening trials Preventing trials Health economic evaluation trial Health-related quality of life trial Treatment trials
Differences between experimental and non-experimental research designs to quantitative environmental research
Experimental manipulation is indispensable when establishing causality as opposed to relying on correlation In experimental research: BOTH methods are used and valid
Running the focus group
Forms - Encourage subjects to sign the registration log - gives you contact info - Distribute the demographic/background info form - offer to fill out if they are illiterate - Collect all forms prior to the start of the focus group Procedure: - Explain study, review consent form, remind them of the expected time frame Tape recorder: - Before group begins: record yourself giving the date, time, location of the group - Have extra batteries and make sure the recorder is charged - Make sure everyone is ok with being recorded Ensuring Confidentiality: - Assign name tags to each participant - Prior to starting, write numbers for each subject next to their name. Give name tag with name and number on it. During the group session, refer to the subject according to their number...not their name Facilitating: - Let them know that can get more info/ask questions after the session - Encourage participants to focus on each other and you - Encourage them to talk about differences in beliefs/understands from others in the group - Urge debate - Have a note taker so you can focus on the group. The note taker should be recording things like body language and other non-verbal communications
Role of good sampling strategies in environmental research
Foundations of environmental research method same as all scientific fields - Logic: sound reasoning based on strict principles of validity - Empirical evidence: results of systematic observation
Inductive approach
Gather data and develop a theory
External Validity
Generalizability from study population to general population in a randomized trial
Six factors for determining research approaches
Goals: - What your project needs to accomplish? - Will you have time for follow up research or is it a one time thing? - What type of information do you need to make a conclusion? - Should be specific
Informal interview
Group, individual, motivational
Structured interview
Has a set list of questions
Mixed methods
Helps to triangulate the data Basic characteristics: - Design can be based on either or both perspectives - Research problems become research questions or hypothes based on prior knowledge, literature, experience or research process - Sample size varies - Data collection involves any available technique - Continuous interpretation
What happens with the data?
Hermeneutic method: How are the statements related? What patterns do you see? Which ones are a function of a variable we predicted? Which ones seem to follow a pattern? - Coding and visual assessment: hand coding...if you don't have a software package that does this - watch out for inter-rated reliability Data Analysis: - Reading: data emersion - read for content, not quality and identify patterns - Coding: identify emerging themes- choose and use a software package, create codes, build theme-related files - Displaying: - identify the principal sub-themes that emerge from the data, examine the evidence that supports each sub-theme - Reducing: distill info to highlight essential concepts and relationships - interpretation, Synthesizing findings: gaps and connections Establishing Trustworthiness: - Credibility - focuses on the confidence in the truth of the findings, look for negative cases, test rival explanations and seek reasons for inconsistencies - Dependability - incorporate a team approach, use multiple independent coders or analyst (remember that inter-rater reliability thing from before) - Confirmability: use and audit trail - Transferability: draw conclusions carefully and provide comprehensive description of context - Triangulation: gather info on the same topic from different sources
Focus Groups: Conclusion
Interactions b/t participants can be used to achieve 7 main aims: - Highlight the respondents' attitudes, priorities, language, and framework of understanding - Encourage research subjects to generate and explore their own questions and develop their own analysis of common experiences - To help to identify group norms and cultural values - To provide insight into the operation of group social processes in the articulation of knowledge - To encourage open conversation about embarrassing topics and to permit the expression of criticism - To encourage a variety of communication from participants-tapping into a wide range and form of understanding - Facilitate the expression of ideas and experiences that might be left underdeveloped in an interview and to illuminate research participants' perspectives through the debate within the group
Skills needed for use of PAR:
Interpersonal skills such as engagement - Cultural competence - Group facilitation and consensus building - Research skills - Knowledge of service delivery systems, community processes, power, and economic and political systems
What is participatory action research?
It is an approach/strategy to research that brings together researchers and community members: - To identify problems that the community faces - To empower community members to research and create solution to those problems - To improve conditions in the community - the community is involved in all aspects of the research from the planning stage all the way through data analysis and implementation of any action plan - PAR emphasizes collaboration between researchers and community members - it is a process of "investigating reality in order to change it" - focuses on creating social change, it is research with, rather than research on or for, it is rooted in resistant values such as mutual aid, care, communalism, intentionality, and reflective action
Critiques of PAR
Main Points: Depoliticization of participation - Failure to take sufficient account of the wider power dynamics which constrain possible impact - Participation needs to be re-politicized and re-scaled beyond the local level Roles and motives of outside facilitators - The restriction that the outsider's role to "facilitation" may be as problematic as one who takes over - The assumption that whatever "local people" say is valid, is as patronizing as its opposite - Financial, political, and/or furthering one's career Over-emphasis on formulas and techniques - structure vs. structurelessness - Structure: fails to take sufficient account of the complexities of real people's lives - Structurelessness: participation built on personal relationships and informal networks generate its own conclusions - The structure of participation has consequences on who participates, how they do it, and how effective participation is likely to be Problems with public participation - Reinforcement of existing privledges - Group dynamics that are likely to lead to 'dysfunctional group consensus' - Creation of group identities that may, themselves, be exclusive - Demands a conscious effort to avoid 'groupthink' Dichotomization of participation - Participation associated with 'salvation' and non-participation with guilt - In some cases, the refusal to participate might defend the ideal of genuine participation better than 'participation' itself - Where is the cutoff point beyond which the rejection of 'participation' is the only way of affirming the idea of meaningful participation? - Is 'participation' even the most appropriate response to inequity and marginalization?
Good Communication Skills:
Maintain eye contact - most of the time - If respondent cries and is not looking at you - don't force it - Nod to show you are engaged - If you have to interrupt, explain why - Face your body toward the person - Use light touch or contact if appropriate (wrist, arms, shoulder). Be aware of the cultural norms, though. In some societies, it is improper to touch others
Ethical Guidelines:
Many argue that PAR is inherently ethical, because this type of work involves: - Placing a lot of responsibility on the participants - Demands continuous reflexivity on the participants - Demands sensitivity to emergent and ethical issues as research unfolds - Research participants are considered less as subjects and more as research partners At the core of the PAR approach are principles of democracy and humanity within research, involving: - Respect for persons participating in the research - A duty of care to vulnerable participants - An effort to limit risk and maximize participants collective and individual benefits - Opportunities for self-representation - Ongoing responsiveness to the needs for the research partners - Frequent review by those who are involved - Continuous reflection about potential ethical dilemmas by the academic research - Reciprocity - quid pro-quo...participating in a project for someone because they participated in your project...this should be avoided - Ethical accountability - Action and publication - Informed consent - Anonymity and confidentiality - Data storage and access
Phases of PAR: Phase IV
Measure change
Intention to Treat
Most RCT use this analysis method - Use incidence rate and include everyone initially assigned by randomization in calculation even if lost to followup
Products of PAR:
PAR presents research finding in ways that will engage community members and other stakeholders and prompt social change: - Powerpoint or poster presentation - YouTube video or documentary - Art instillation or exhibit - Theatrical production - Publication in an academic journal - Presentation at an academic or industry conference - Anything else the collaborators choose
How does PAR happen?
PAR uses a cyclical process - Pre-cycling: the community and the researchers come together to prioritize the needs of the community - Reflecting: as a group, on how the research is going - Planning the research activities and next steps - Acting out the planned research activities - Observing the collected data and research process
Types of RCT
Parallel study - two groups followed over time - Four comparison groups Cross-Over Trial - intervention group and control exists for some time but then there is a "wash out period" to clear body of drugs based on half-life and serum concentration then the control group becomes the intervention group and the intervention group becomes the control group - This design adds more credibility to the intervention because it essentially two prospective studies in one
Dynamic systems
Physical, chemical, biological changes are continuous - Ex. most rivers can use systematic and equidistant sampling of points, usually based on criteria and repeated over time - Dynamic system more commonplace in environment as environmental systems are usually very dynamic and heterogenous
The process for conducting in-depth interviews:
Plan - ID what info is needed and from whom - ID stakeholders who you would like to recruit (health professionals) - List stakeholders to be interviewed and then ID individuals within those groups - Ensure research with follow international and national ethical research standards, including review by ethical research committees Develop Instruments - Dev. an interview protocol - What to say to interviewees when; including informed consent - What to say to interviewees in concluding the interview - What to say during the interview - What to for follow-up Recruitment - Send out invitation letters to potential participants - Follow-up with phone call, using phone script - Make sure that inclusion criteria from participations are met - Choose an agreed upon location for the interview that is private and quiet At the interview: - Have the subject complete the demographic form - Assign a number or fictitious name to the interviewee and write it on a name tag - Seek informed consent. Re-explain the purpose, why they were chosen, expected time of the interview, confidentiality, and the use of a note taker or tape recorder Collect Data: - Once the subject has consented: turn on recorder. Record in your voice: starting date and time, location, fictitious name/number After the interview: - Summarize key data for your report immediately following the interview - Verify info. Given in interviews as necessary - Solicit feedback from interviewees..revise..finalize
Stages of Studies: Stage Two
Pre-clinical Research - considered Phase Zero, NO HUMANS tested here - Animals and test-tubes, bench research
Limitations to an in-depth interview:
Prone to bias - program or clinic staff might want to "prove" that the program is working, their interview responses might be biased - Responses from community members and program participants could also be biased due to their stake in the program Can be time-intensive: - Interviews can take a long time, then they have to be transcribed and analyzed Interviewer must be appropriately trained in interviewing techniques - They should be comfortable and interested in what they are doing - They must be sure to use effective interviewing techniques...don't use leading questions --> use open ended questions. Have a list of steps that are to followed for every interview. Keep your opinions to yourself. Not generalizable - Generalizations about the results are not usually able to be made because small samples are chosen and random sampling methods are not used
What is an in-depth interview?
Qualitative research technique that involves conducting intensive individual interviews with a small number of respondents to explore their perspectives on a particular idea, program or situation
Experimental Design
Randomized participants in study
PAR in Public Health
Referred to as community-based participatory research (CBPR) Has made important in-roads in Public Health: - Diaster relief - Community based rehabilitation - Accident prevention - Hospital care - Drug prevention - Occupational health and safety - IOM, in 2001, identified PAR as 1 of the 8 new competency areas
Phases of PAR: Phase V
Report accomplishents
Haphazard
Sampling location is chosen arbitrarily
Semi-structured interview
Set list of questions but allows for probes to elicit more info
Scientific research methods
Set of techniques and procedures for empirical scientific investigation, and logic and conceptual foundations that tie scientific investigations together with substantive theory
Clinical Research: Phase Two
Several hundred patients, Study length: months-2 years Purpose: Efficacy, side-effects
Clinical Research: Phase Four
Several thousand patients, study length: varies, purpose: safety and efficacy
Random
Specified number of samples, where each has an equal chance of being selected
Deductive approach
Start with a theory and then prove it
Purposes of an outbreak investigation:
Stop the outbreak - Ensure PH/prevent the spread of disease - identify the agent, reservoir, source and/or mode of transmission (hard to establish), determining who is at risk for the disease, place, time, identifying the exposures or risk factors that increase risk of disease - Prevent future outbreaks - Improve surveillance and outbreak direction
Systematic
Taken at locations and/times according to a predetermined pattern
Screening trials
Test best way to detect diseases/health conditions (Ex. Colonoscopy, Fecal Occult Blood tests for colon cancer screening)
Involvement of PAR Teams:
The PAR teams as a whole: - Determines the research questions - Designs research project, including evaluation tools - Carries out the research activities - Analyze and interprets the data - Shares results in a meaningful way with stakeholders
Basic interviewing skills
The interviewer is listening and paying attention to verbal and nonverbal cues - Alert to know when to ask a follow-up question probe or ask for an example - Creates trust and empathy - Conveys this verbally and nonverbally - Adopts the interview to the speech code and comfort level of the respondent - Paces - knows when to skip and when to repeat questions - Confirms what she/he thinks was said without restating - yes - so you were frustrated during this part? - No - you think this is a waste of time? - Be yourself: be honest, share your thoughts if they agree with the subjects, dress properly - look professional, be respectful: no gum chewing or cell phone calls...get off of Facebook!
Rationale for essential components of an epidemiologic case definition
Three pronged approach - Epidemiologic - most common - Laboratory testing - need to know what is being tested - Environmental assessment
Stratified
Used in systems w/very distinct areas, requiring random sampling in each strata
Internal validity
Validity within the trial - how well has the study been conducted? Free from bias? Randomization? Is it valid for the study population?
Sampling Strategies
Where, when, how considering implications of study objectives, variability of environmental system and cost of study
Focus Groups: Recruitment
You want a variety of subjects so you can get different opinions - Recruit from different geographic areas (from each parish in Grenada) - Keep a log of each recruit, and update it as necessary - Book your location (quiet and private) ahead of time - Have refreshments: creates relaxed environment, ensures folks are hungry...plan ahead
Basic principles of the scientific methods
a. Empirical evidence: capable of being verified or disproved by observation or experiment b. Logic: sound reasoning based on strict principles of validity NOTE: using conclusions based on inductive reasoning, deductive reasoning is used in the scientific method to reach a conclusion
Grab sample
discrete sample collected at a specific location and specific point in time
Diagnostic trials
identifying better procedures for diagnosing disease/condition
Do a check at each phase....
if it's not working, then go back and re-work your plan
Quasi-Experimental Design
non-randomized participants in study
Composite sample
samples made by throughly mixing several grab samples taken at different times and/or locations