Exam 1 Intro to Research Methods

Goal of Study: Compare Dif. Tooth Whitening Methods Whitness was scored on a 1-5 scale (1 lightest, 15 darkest) One man in study had a "baseline" color of 1 (before the study). What was wrong with INCLUDING him?

"Tip#2: get people who have the condition you're studying" aka "people who need tooth whitening" -> this man should not have been included. Next time, add inclusion rule: people with baseline scores over 5.

Dental Examples of statistics - fix.. write these out

#1 toothache Pain: a comparison of visits to physcans, emergency departments adn other dentists. - ONE QUESTION = who is more likely to visit the dentist? men with toothache or women with toothache? Analyze: 1) % who say "red": - men - women 2) % who visit dentist: - men - women #2 "Caring for African-America patients in private practice: disparities and similarities in dental procedures and communication" ONE QUESTION - who is more likely to chat with their dentist? patient who are the same race as their dentist? or patients who are a dif. race from their dentist? Analyze: 1) % who say "red" - men - women 2) % who chat with dentist: - same race - dif race QUESTION: DR. X THINKS THE CHANCE OF CHATTING DEPENDS ON WHETHER THE PATIENT AND DENTIST ARE THE SAME RACE => we take a SAMPLE of each situation: - in "same race" situation, 78% chat. - in "different races" situation, 44% chat. WHAT WOULD BE MORE EVIDENCE SUPPORTING DR. X'S HYPOTHESIS? a sample size of A) 9 per group or a sample size of B) 900 per group? answer: sample size of 900 per group. - In both A) and B), had 78% vs. 44% chatting, which is pretty big difference. - A) 9 per group -> results are not reiable [hard to draw definitive conclusions, the difference could be random chance]. - B) 900 per group -> results are reliable [the trend (78% vs 44%) held up over a big sample. we're pretty sure the difference is REAL]. IMP POINT^: in statistics, virtually always better to get higher sample size. THE HIGHER THE SAMPLE SIZE, THE MORE RELIABLE YOUR RESULTS WILL BE. Sample size is also referred to as "N"

What are the "Stepping Stones of the Research Process"?

- developing a new conclusion from the evidence - designing a method to collect data about the research question. - collecting and integrating current knowledge. - asking a question.

Four TYPES of Foreground (PICO) Questions:

1) Therapy/prevention questions - To determine effect of treatments/interventions that r imp. outcomes for patients (e.g. improve function, worth the effort and cost, prevent harm or avoid adverse effects). 2) Diagnosis questions (dont have example on this one on a dif fc so know it) - To determine degree to which a test is reliable and useful (e.g. establish the power of an intervention to differentiate btwn those with and without the condition or disease). 3) Harm, etiology, and causation questions - To identify causes of a disease or condition and determine relationships btwn risk factors, potentially harmful agents, and possible causes of disease or condition. 4) Prognosis questions - To estimate the clinical course or progression of a disease or condition over time, anticipate likely complications and prevent them.

Want the sample and population to be similar in terms of:

1) demographics: % in each gender, race, etc. 2) medical/dental characteristics: % who smoke, who visit dentist more than once/yr, etc. .

Ways to improve your study, regardless of sampling TYPE?

1) use proper inclusion/exclusion rules 2) Report information about representativeness - readers will want to know how representative your sample is. True for simple random samples and for convenience samples. -> e.g. IF SAMPLE IS DIF. FROM THE POPULATION (remember that table), LIST THAT AS LIMITATION. Note: many times tho we only know some of the information in the table (we know sample percentages, but not population percentages). Include sample percentages in your paper. 3) Avoid extrapolating or over-generalizing results. EXTRAPOLATION: - when you try to generalize ur findings too far. - apply your results inappropriately. - examples: Kids vs. Adults (see below); If you do an "in vitro" study, dont assume same results will apply "in vitro!"; if something happens in a test tube, dont claim it will also happen in a living organism (DONT EXTRAPOLATE)! - example: suppose everyone in ur sample is a child, but you claim that the results apply to adults too -> this is not smart. when you extrapolate, often draw conclusions that are rediculous.- Studying growth patterns in children, take sample of 100 kids ages 6 to 10, kids average 3 inches of growth per year, extrapolate to adults: claim they also grow 3 inches per year...it is SO wrong to conclude from this that adults will grow to 20 feet. (DONT EXTRAPOLATE TO ADULTS; adults teeth/bodies are not like kids!)

What are the "three distinct stages" in ALL types of research?

1. PLANNING - developing research question, research background and rationale, form research hypothesis, chose applicable methodology, writing resarch protocol/proposal, obtaining funding, gaining ethical and organizational approval. 2. DOING - prep. for data collection and management, recruitment participants/consents, analyzing the data. 3. PUBLISHING - planning how to publish the results, presenting the data at scientific conference, writing manuscripts for peer reviewed journals.

Three Basic Concepts for Critically Appraising a Research Article:

= Messsage, Validity, Utility. 1. Message! - "WHAT ARE THE MAIN RESEARCH RESULTS AND CONCLUSIONS?", what's the take-home message? what's the "bottom line"? How does this work support my research plans? 2. Validity! - "CAN THE CONCLUSIONS BE JUSTIFIED BASED ON THE APPROACH AND FINDINGS?" Was there a clear statement of the research aims? Was the method appropriate for the research questions? Are the key findings discussed and justified? How significant was the treatment effect? Has the research team explored source of bias? 3. Utility: - "IS THE RESEARCH RELEVANT TO UR STUDY AND WILL THE RESULTS BE APPLICABLE TO UR PATIENTS?" How do the findings fit in w current knoeldge, and what do they add to it? Are new areas of study raised? Can the new information change my current practice? Can the results be applied to the local population? How different are the study patients from your patients? - Additional things to pay attention to: Ethics, Funding Sources, Appraising the statistics.

What is Critical Appraisal? When would we do this? Why do We Need to Critically Appraise Research Articles? Three Basic Concepts for Critically Appraising a Research Article?

= a method that allows you to increase the effectiveness of your reading and reviewing the literature. -> exclude (OUT!) articles that are of poor quality and non-informative; & (IN) evaluate good articles for their strengths, weaknesses and key discoveries. - this step comes after you've found 1,000 articles in the area you are searching (and thus you've got your research question). Why do We Need to Critically Appraise Research Articles? CRITICAL APPRAISAL WILL HELP YOU TO: 1. Decide how applicable the research you're reading about is to your study, and if it would help you improve your clinical practice. 2. Refine your research question and study design. 3. Evaluate the published literature as part of the submission of research protocols for ethics committees (IRB, IACUC) and funding agencies. CRITICAL APPRAISAL SKILLS PROGRAM (CASP) - THERES AN APP FOR THAT! CASP CHECKLISTS. ITS ALL ABOUT A CAREFUL, HEALTHY BALANCE: btwn: Assuming that every published article is written by a pathological liar? versus assuming that every published article is good bc it is published! Do NOT assume that "Just Because the Article is Published It must be Useful" - you need to be CRITICAL and SKEPTICAL, but TO A POIINT. 1) "What is the article looking for?" - reviewing the resarch question, hypothesis, or specific aim are usually at the end of the introduction. 2) "Are the methods relevant" - reviewing the methods section to see if the approaches used are relevant for the rsch question and study design. 3) "Is the Article Useful for You?" - Reviewing the reuslts and conclusions. If the information is not stated clearly, is ambiguous and is not helpful, leave it and move to the next article.

How do you get a representative sample?

A) SIMPLE RANDOM SAMPLING: choose sample "at random" from the population. Each subject has the same chance of inclusion (its fair). A random sample is likely to be similar to population (its representative). = EXAMPLE: You want to know % of TUSDM students who actually floss -> make list of all students, pick 20 at random, and write down whether they floss -> sample should be REPRESENTATIVE = IMP COMMENT: SIMPLE RANDOM SAMPLING may or may not be possible for you (can use in some studies not other: CAN use it in small population that is easily accesible (tufts dental students), CANT use it when population is large and/or not easily accessible (kids ages 6 to 10). B) CONVIENCE SAMPLE = not as good as simple random sample, but still worthwhile.

REVIEW (BRIEF) HISTORY OF RESEARCH ETHICS BEFORE AND AFTER WW2: (lecture 2) A) Eugenics Movement B) Nuremberg Trials: Nuremberg Code (1947) C) United States of America v. Karl Brandt, et. Al D) Henrieta Lacks E) Tuskegee Syphilis Research (1932-1972) F) National Research Act of 1974 - be FAMILIAR with this one at least... G) "Experiments in Torture: Evidence of Human Subject Research and Experimentation in the "Enhanced" Interogation Program" H) Belfast "Good Friday" Agreement and No. Ireland I) Huntington's Disease & Search for Cause

A) genetic superiority - foundations in US in at turn of 20th century. - led to forced sterilization in those "unfit" to reproduce -> target populations = poor, uneducated, native americans and immigrants. - Sterilization in Vermont: targeted "3 D's": delinquency, dependency, and mental defect; practice continued well into late 1950's, Abenaki Indians & French Canadians. B).Nuremberg Code: 1. Voluntary Consent of the human subject is absolutely essential. 2. To the Benefit of Society and not obtainable by other means 3. Designed and based on results of animal experiments and knowledge of the disease or problem that anticipated results will justify the experiment. 4. Avoid all un-necessary physical and mental suffering and injury. 5. No prior expectation of death or disabling injury* 6. Degree of risk never exceed humanitarian importance of the problem. 7. Proper planning and facilities protect against any possibility of injury, disability, or death. 8. Conducted only by qualified persons. 9. During experiment, subject may withdraw at any time. 10. Scientist must be prepared to stop research if continuation of the experiment is likely to result in injury, disability or death. - The accused (against nuremberg code) faced four charges including: 1. "War Crimes": from performing medical "experiments" without subjecs' consent, or plannign and performing mass murder of prisiners of war and civilians. 2. "Crimes against Humanity": commiting crimes described under count 2. C) United States of America v. Karl Brandt, et. Al - first of 12 trials for war crimes after world war 2. - 20 medical doctors, 3 Nazi officials - all accused of involvement in gruesome human experimentation. - trial held october 1946 to dec. 1947. - 7 perceived death sentence, 9 to prison, 10 to life, 7 acquitted. D) Henrieta Lacks - samples taken WITHOUT CONSENT -> "immortal life" , HeLa cancer cell line. - born: august 1, 1920 - Dx'd w cervical cancer in 1951 - sampels taken by George gey W/O consent. - Source of HeLa cell line. - "the immortal life of henrietta lacks" - inmates at ohio state prison (1950's-1960's) , terminally ill patients, and 300 were injected with live HeLa cancer cells without being told. E) Tuskegee Syphilis Research (1932-1972) - funded by US public health service to study natural history of syphilis. started in 1932 when no effective tx avaliable. enrolled 399 indigent and uneducated AA sharecroppers in alabama. By mid 1940's pencillin known effective but actively withheld from participants who beleived they were receiving treatments. Only stopped after whistleblower brought to media attention in 1972. $10 million payout followed. By end of experiment, 28 men died directly of syphilis, 100 men dead of related complications, 40 of their wives had been infected, 19 of their children had been born with congenital syphilis. F) National Research Act of 1974 - created The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. Belmont Report (1979). Established THREE ethical principles in conduct of human subjects research. - BELMONT [applying ethical principles] A) Respect for persons = AUTONOMY => informed consent, voluntary participation, freedom to withdraw w.o penalty, protect privacy and confidentiality. B) BENEFICENCE: "Do unto others..." => risks are minimized and benefits are maximized, risks are justified by potential benefits, conflicts of interest are managed or eliminated. C) JUSTICE: Distribute risks and benefits => protection of vulnerable populations [i.e. pregnant women, neonates, fetuses; children; prisoners], those who may benefit are not excluded tho. G) "Experiments in Torture: Evidence of Human Subject Research and Experimentation in the "Enhanced" Interogation Program" - Terror Investigations: Following sept. 11, 2011 attacks, US gvt stepped up intellegence gathering. Justice Dept Office of Legal Council re-defined certain acts (waterboarding, extreme sleep deprivation, stress positioning) as "safe, legal and effective" under rules of "enhanced interrogation techniques" - Enhanced Interrogation: required medical personnel to monitor EIT, ostensibly to ensure did not cause "severe mental or physical pain." Accusation that CIA systematically used information obtained to enhance "safety" and effectiveness. H) The troubles - secessionist movement in No. Ireland pitting predominantly Nationalist/Catholic groups favoring independence from England (IRA) versus predominately Unionist/Protestant groups (orange order) favoring the UK. - Violent protests, bombings, executions, reprisals, and recriminations over 3 decades. -> BELFAST "GOOD FRIDAY" AGREEMENT: negotiated by George Mitchell (D), ME. Ratified by popular referendums in Northern Ireland and Republic of Ireland in 1998. Negotiated half to hostilities and disarming of militant groups. Integration into political system. -> "Belfast Project" = oral history project at boston college initiated by journalist Ed Moloney and formed IRA member Anthony McIntyre begun in 2001. Interview key players on both sides. Interviews granted on condition that would not release interviews until participant had died. Subject of interview (Dolours Price) reportedly confessed involvement in murder of Jean McConville, accused informant for British Gov. in 1972. Revealed involvement in BP in violation of confidentiality agreement with BC. British gov. re-opened case and subpoenaed interview transcripts and tapes under terms of mutual aid treaty with US. BC lost intitial attempt to block subpoena and did not file appeal. Researchers resisted on basis of perceived danger to themselves and research subjects and First Amendment privilege. Reoped of old wounds. Legal procedures continue across the two continents, has involved US State Dept and US Supreme Court. I) Huntington's Disease & Search for Cause: - autosomal dominent genetic disorder of Htt gene results in progressive neuro deterioration by age 30's in all affected. No cure, treatment palliative, death inevitable. Symptoms can appear after peak childbearing age (-> to have children or not?). - Research: Genetic polymorphism association studies hit on the 12th try, localized to short arm Chr4 in 1983. Gene identified and mutation defined in 1993. Research continues to understand pathogenesis of disease and potential treatments. - Huntington's Kindreds: sucess of genetic linkage analysis relies on size and quality of family information. limited kindreds identified in US. Largest kindred lives on shores of Lake Maracaibo in Venezuela with 3000 members sine early 1800's traceable to single ancestor. - Huntington's Progress: researchers spent months obtaining blood samples and pedigree information. Identification of gene led to development of genetic test to predict HD [avaliable widely in US and Europe, Costs ~ $2,000]. Potential treatments include siRNA and other $$$ (i.e. profitable) approaches. - Who ahs benefited from this research? Are venezuelan families likely to benefit from future treatments? Nancy Wexler, MD has raised $6M for care in Venezuela; Hereditary Disease Foundation runs respite/clinic in VZ.

"An approach to oral health care that requires the judicious intergration of systematic assessments of clinically relevant scientific evidence, relating to patients oral and medical condition history, with the dentists' clinical expertise and the patient's treatment needs and preferences"

ADA definition of Evidence-based practice.

Background and Foreground Questions (EBD is a five-step process that starts with a question - a PICO question!)

BACKGROUND questions are general knowledge inquiries that ask who, what, where, when, how, or why. They help narrow a broad scope and search about a topic in order to find the details needed for a foreground/specific question. Examples of background questions: - What causes ___ (identify disease OR condition)? - What is ____ (identify product or therapy)? - How are patients with _____ (id disease or condition) managed? - What are the suggested therapies for ____ (id disease/problem)? FOREGROUND question often ARISES FROM a problem or patient/client question. It is a specific question that is structured to find a precise answer and phrased to facilitate a computerized search. A "WELL-BUILT" FOREGROUND QUESTION IS DESIGNED AS A "PICO" QUESTION.

So you have a good research question, you have reviewed the literature and critically appraised articles, and you have revised the rsch question... now what?

Chosing the right research design to answer your question!

Three components of "Evidence-Based Practice"?

Dentists expertise, scientific evidence, patients needs and preferences.

How many steps is evidence-based practice? What does it start with?

EBD IS A FIVE-STEP PROCESS THAT STARTS WITH A QUESITON. 1. QUESTION: articulate a clear q based on patient's clinical needs (PICO quesiton). 2. FIND: conduct comprehensive search for latest relevant research. 3. APPRAISE: critically assess the relevant evidence. (critical appraisal) 4. ACT: use best identified valid and relevant evidence in patient care. 5. EVALUATE: assess how well the previous steps worked.

What are elements of a good sample?

Elements of a good sample: a) Quantity: N (sample size) count should be HIGH ENOUGH to draw reliable conclusions (the higher the sample size, the more reliable your results will be). b) Quality: should choose a VALID group of subjects.

"Based on the results, can we CONCLUDE that X and Y are really different?" "Based on the results, can we conclude that the chance of something depends on X vs Y?" What would make your results more valid (versus less valid) / better evidence?

Greater difference in sample % (the statistic is a number that comes from a sample) means more evidence that the difference between X and Y are REAL [and not due to random chance, given that sample is portion of population]. Sample size with greater amt ppl per group will have more reliable results because you can be more sure the difference is real (easier to draw def. conclusion) and difference is less likely to be random/by chance. Subjects (group of ppl) need to be valid: (1) use good inclusion/exclusion rules, (2) report information about representativeness in you paper (e.g. must list as limitation if groupA% in your sample differs from %groupA in population; the closer sample%'s are to population %, the better), (3) don't extrapolate results (e.g. in vitro doesn't imply in vivo results; results in kids doesn't imply same in adults).

What are some ways YOU can be involved in research? and why should YOU (dental student) consider being involved in research?

How can You Be Involved? - critically appraise the literature to decide if there is evidence to change in practice. - become part of research team to develop and implement a project. - help carry out a research project. - lead a research program, have an academic career. Why should YOU consider being involved in research? - ALL HEALTH CARE PROFESSIONALS ARE OBLIGATED to provide care based on the best evidence avaliable, which shows that a particular therapy, intervention, or technique actually works.

Why do we need evidence-based practice?

Large range variations in practice; slow translation/integration of scientific evidence into practice (e.g. dental sealants); managing the information overload; changing educational competencies that require students to have skills for lifelong learning. WHY DOING RESEARCH? EVIDENCE-BASED PRACTICE is necessary in medicine and related professionals in order to ensure that the treatments and the processes that we use are sound [research provides evidence}

Hierarchy of Research Evidence - the traditional hierarchy of resarch designs that will provide the most reliable evidence for effectiveness of intervention:

Most powerful evidence (1) based on ability to control for bias and to show cause and effect in humans = multi-center clinical trials. Least powerful evidence (8) = case reports and opinions. 1. SYSTEMATIC REVIEWS & META-ANALYSIS: Systematic review combines the results of a number of original studies that have been systematically searched out and judged on a defined inclusion criteria. A meta-analysis is the combo of results from several independent studies. 2. RCTs: particpants are assigned randomly to a control or a treatment group who receive a specific intervention. 3. COHORT studies: two or more groups of ppl who have a dif. exposure to a specific agent (i.e. smoking, surgery) are selected and folloewd up to find the differences btwn the groups for specific outcomes. 4. CASE-CONTROL studies: retrospecfive analysis of individuals w a specific condition (cases) that are matched with those without that condition (controls). 5. Case series, cross sectional studies/surveys: an examination of a sample of the population of interest at a specific point in time. 6. CASE REPORTS: a report based on a single patient. 7. Expert opinions: a consensus from experts in the field. 8. Animal and laboratory research: highly controlled and precise experiments that have questionable relevance to human condition.

IRB - understand basic tenets of human subjets research and the role fo the IRB -> discuss application of these principles to recent controversial research activities

Origins of the IRB: - National Research Service Award Act (1974) Authorized creation of the IRB framework: -- DHS promulgated 45 Code of Federal Regulations (CFR) 46 (1980's, revised 2001). -- 45 CFR 46 adopted by 17 federal agencies: the "Common Rule" (1992). --Compliance is condition of receiving Fed funds. Role of the IRB: - protects rights and welfare of research and subjects at the INSTITUTIONAL level. - assist investigators in meeting federal, state and institutional regulations and policies. - Review of "Human Subjects Research." - Makes decision: "is this research?" Federal Wide Assurance (FWA) - Dept. of Health and Human Services holds purse strings for NIH funding. - FWA is institution's agreement to have IRB review all NIH-funded research. - Decisions regarding "research" assigned exclusively to IRB. Common Rule (45 CFR 46) - Defines: Research, Human Subjects Research, & Three Categories of HSR = Exempt, Expedited (not greater than minimal risk), & Full board review required. Essential Elements of IRB Review: 1. Risk are minimized. 2. Risks are reasonable. 3. Selection of subjects is equitable. 4. Informed Consent will be sought. 5. Adequate plan for monitoring data. 7. Adequate plan for protecting Privacy. When is IRB review required? If research involves human subjets AND is not otherwise exempt, the Common Rule Applies: - Informed consent is required - IRB must review and approve - IRB must review annually. Who makes up the IRB? (open invite) - medical and dental professionals spanning many disciplines. - Scientists/scientific staff - community representatives e.g. lawyers, lay community members, prisoner advocate. - Professional office staff (director of IRB operations, IRB blue cordinator, IRB red cordinator).

4 aspects of PICO question (give example)? (lecture 3)

P = Problem or Patient - what is the problem or patient group you r interested in. I = intervention or exposure - what is being done to the patient or what possible exposure is of interest? C = Comparison - what is the usual treatment or factor you want to compare with? O = outcome - what is the desired outcome of the treatment? Example of PICO question: "For a patient with tooth discoloration due to coffee and tobacco, will crest whitestrips, as compared to custom trays for use with an at-home whitening bleaching system, be as effective in whitening his teeth within 1 week?" ANother example of Pico question: "Does the use of powered toothbrush for six months (I) reduce the plaque index (O) of patients with chronic periodontitis (P) as compared to the use of manual toothbrush (C)? Note: FOREGROUND question often ARISES FROM a problem or patient/client question. It is a specific question that is structured to find a precise answer and phrased to facilitate a computerized search. A "WELL-BUILT" FOREGROUND QUESTION IS DESIGNED AS A "PICO" QUESTION.

Quantitative Research vs. Qualitative Research (be able to define, was in red) - what are the main approaches used in each type?

QUANTITATIVE RSCH = involves the analysis of NUMERICAL DATA in order to explain a pre-defined research hypothesis. The researcher aims to remain objective and control the conditions of the research. - main approaches = these both test theories by producing NUMERICAL DATA that are statistically analyzed. the use of a SAMPLE OF A POPULATION in order to make extrapolation about the WHOLE POPULATION from the result of the smaller sample. Data can be collected PROSPECTIVELY (from start of study forward) or RETROSPECTIVELY (searching back thru data bases). Data can be analyzed to study CHANGES OVER TIME (longitudinally) or to observe an OCCURENCE AT A SPECIFIC TIME point (cross-sectional) within the study sample. a) EXPERIMENTAL STUDIES try to establish CAUSE & EFFECT (e.g. evaluation of new therapies or procedures) b) OBSERVATIONAL STUDIES do not include a reseach intervention; they explore association/CORRELATION BTWN FACTORS to establish the development of a condition (smoking and cancer). QUALITATIVE RSCH = involves analysis of WORDS, PICTURES, OR OBJECTS in order to explore and describe a situation, process or experience. The researcher interprets the data as it unfolds during teh collection process, and the subjective nature of the process is embraced. Provides a more HOLISTIC VIEW, the "BIGGER PICTURE" of the problem that is investigated, INCLUDING information that cannot be reduced to numbers. - main approaches: based on the philosphy that there may be MORE THAN ONE WORLD VIEW and different researchers can interpret the same data in different ways; these appraches allow the resarcher to interact w the research subjects in their own language and on their own terms, allowing a DEEPER UNDERSTANDING of the problem under investigation. The approach to data collection, study design, subsequent analysis and data interpretation is MORE FLEXIBLE. This allows the researcher to EVOLVE THEIR VIEW of the problem under investigation and ADAPT TO THE CHANGING VIEW. = includes: PHENOMENOLOGY, GROUNDED THEORY, ETHNOGRAPHY RESEARCH, and NARRATIVE RESEARCH.

What is NARRATIVE RESEARCH?

QuaLitative resarch that relies on the stories that the subjects tell. The order of the narrative reflects teh meaning that the subject wants to convey. Analysis of narratives can produce themes for further investigations.

What is ETHNOGRAPHY RESEARCH?

QuaLitative research based on observations and collection of data on a group of subjects within their usual setting (context) over a prolonged period of time.

What is PHENOMENOLOGY?

QuaLitative research that study the ways in which subjects conceive and interpret the world. In this view, reality is relative and subjective.

What is GROUNDED THEORY?

QuaLitative research which is a systematic inquiry into a problem w the intention to develop and overall theory based on personal experiences.

What is the definition of research?

Research is a systematic investigation of a specific question in order to establish new facts and draw new conclusions, that involves (a) the discovery of new knowledge, (b) Interpretation and Revision of Current knowledge.

You have a good research question, what do you do next?

Reviewing and evaluating the literature! - learn what has been done in this area and what related info is out there that can help you develop your research study. - devise a Database Literature Search Strategy Ysing a Research Question: Searching Databases can be done by using: A) Text Word Search - this is a LARGE, BUT LESS SPECIFIC LITERATURE SEARCH: typing a specific word or phrase in the search box to search a database directly. All the documents that contain that word or phrase will be retrieved. B) Thesaurus Search - this search AVOIDS THE PROBLEMS OF DIF. SPELLINGS OR TERMINOLOGY THUS IT IS MORE SPECIFIC: searching w terms, such as Medical Subject Headings or descriptions, that comprise the thesaurus (in an hierarchical structure) for each database.

If sample percentages are way different, there is more evidence that?

STATISTICS PRINCIPLE #1: If Samples% are way different, more evidence populations %'s are different. example: Our sample: 41% of women picked red, 49% of men picked red. -> Statistics = 41% and 49%. The more different the mens SAMPLE % is from the women's, the more evidence POPULATION %'S are different. e.g. 20% of men say "red" vs. 90% of women in one sample; and 40% of men say "red" vs. 55% of women in another sample. -> 20% vs. 90% is MORE evidence than 40% vs. 55%.

Sample vs. Population (lecture 6) - define each - how are they related?

Sample = small portion of the total population. e.g. In our sample (in the study), 41% of women picked red and 49% of men picked red. Purpose of Sample: help us draw conclusions about the population. Want the sample to be representative: want it to have similar characteristics to the population we're interested in. "the sample should be like the population, only smaller" Example: What % of inner-city teenagers have poor oral hygiene? - > want the SAMPLE to be representative of inner-city teen population.

For managing a patient with periodontal disease, will use of a Nd:Yag laser, as compared to scaling and root planning, be as effective in managing periodontal disease? is what type of pico question?

Therapy/prevention question - to determine effect of treatments/interventions that r imp. outcomes for patients (e.g. improve function, worth the effort and cost, prevent harm or avoid adverse effects).

What are things you should think about when choosing your research design - how do you choose the right one?

Think about: - what TYPE OF EVIDENCE are you looking to generate? are you looking to prove a point (quantitative) or understand something better (qualittative)/ - what research design OTEHR PEOPLE USED to answer similar questions in a dif/similar field? - If you need a QUANTITATIVE APPROACH, use the hierarchy of evidence as a guide to produce the highest posisble level of evidence. - If you need a QUALITATIVE APPROACH, consider the types of information that each method will produce to best fit your question. - Finally, think about your AVALIABLE TIME, BUDGET, SKILLS, PPL INCLUDED IN THE STUDY, AND THE HELP FROM YOUR MENTOR..

Inclusion/Exclusion Rules

To be eligible for a study, a subject: must meet all inclusion rules + must NOT meet any exclusion rules. = rules that determine whether a subject can enter the sample. INCLUSION RULES tell the types of subjects we are studying e.g. "adults who have chronic jaw pain" EXCLUSION RULES tell specific questions that would not allow a subject to participate e.g. "currently on medication for psychiatric illness." Tips for Good Inclusion/Excusion Ruels: 1) EXCLUDE subjects who COULD BE HARMED BY THE RESEARCH - e.g. studying the effect of BP medication you should exclude pregnant women bc of possible harms. 2) INCLUDE ppl who HAVE the CONDITION you're studying, and DON'T have OTHER conditions that could get in the way - e.g. compairing dif. methods for treating TMD. inclusion rule: get ppl who have been diagnosed with TMD; Exclusion rule: remove ppl on medication for psychiatric illness (could prevent themf rom accurately reporting TMD symtpoms). 3) Make rules specific, easy to "check off" - dont just say you want to include "healthy" subjects, say you'll include subjects with "no systemic chronic disease in the last 6 months" Example - Goal is to evaluate effectiveness of dif. techniques for treating temporomandibular joint disorder. Inclusion rules = age at least 18 years, pain in jaw for at leats 12 months, enoguh teeth present for fabrication of a full coverage maxillary and mandibular appliance. Exclusion rules = pregnancy, recent trauma (3 months), currently on medication for psychiatric illness.

What makes a successful researchER?

What makes a successful researchER? - Similarly to health care which is TEAM ORIENTED. Research is nearly always a TEAM EFFORT! - NEEDED CHARACTERISTICS: ability to work methodically w attention to details, Good time-planning and time management; Organizational skills; Ability to communicate clearly; A lot of patients, determination, and persistence; Thick skin to deal with setbacks and rejection! - Researchers also must... Be curious, regularly question what they do, have a desire to pursue answers n face of many difficulties. - NEEDED ENVIRONMENT: where research effort is valued; time is allocated to undertake research activities; permission to develop a network of peer support within and outside of the organization; support and guidance; promotion of team-working.

Define Statistic

a number that comes from a sample (a small portion of the total population) They are used to draw conclusions about the whole population - (bc our results don't always apply to everyone) { sample aka small portion of total population--> thru (statistical analysis) --> population info}

You are now familiar with the aims and stratifies of quantitative and qualitative reseach designs, whats the next step in the ersesarch process?

matching the research aproach and specific design to your specific research question. - > choose the right research design.

4 myths of statistics? (lecture 6)

myth #1: "if you're not a statistician, you dont need to know about statistics." myth #2: "statistics is super impossible!" (it actually can be INTUITIVE) 3. "statistics is about lying" (actually we use statistics to FIND the truth, not hide it) - but certain people will try to fool you with statistics - WARNING 1: be SUSPICIOUS if they DONT report sample size e.g. "4 out of 5 dentists recommend trident" -> if sample size is too small, results are not reliable, and if ppl dont tell you sample size they might be hiding it. - WARNING 2: be SUSPICIOUS if they dont tell you WHO was in the sample: same e.g. doesnt tell you who was surveyed. - WARNING 3: """ if they focus on numbers that support their own side - e.g. paper comparing two pain medications and the study was sponsored by company that makes one of the drugs and presented many stats about how that drug did better and few stats about how that drug did worse. - WARNING 4: misleading graph (note this can be an accident bc of excel) e.g. vertical axis doesn't go down to 0. ... or e.g. horizontal axis skips from year 2001 to 2008 and rest of it went in order. 4. "statistics is boring" (...we'll do trivia)

Medical Research Ethics (lecture 2)

Before world war two, there was limited concern in society for welfare of research subjects prior to second world war. In 1930s, " a cult of medical research" celebrated researchers who experimented... and on others. But, opposition came from animal anti-vivisectionist equating animals and humans. - Experimentation on human beings inherently immoral; but in the interest of promoting health and welfare, society is willing to make exceptions under extremely narrow circumstances. The authorization to conduct human subjects research is a privilege afford to very few individuals"

For patients with a prosthetic joint, newly placed 3 months ago, should prophylactic antibiotic coverage be prescribed prior to undergoing dental procedures to prevent infection? is what type of pico question?

Harm, etiology, and causation questions - To identify causes of a disease or condition and determine relationships btwn risk factors, potentially harmful agents, and possible causes of disease or condition.

Is evidence-based practice limited to dentistry (e.g. Evidence Based Dentistry (EBD)?

No. Started in medicine in the 1970s (EMB) = Evidence Based Practice (EBP), Evidence Based Health Care (EBHC), "" Nursing (EBN), "" Public Health (EBPH).

For an adult with dental implants, does diabetes negatively affect the longevity of the dental implant? is what type of pico question?

Prognosis question - To estimate the clinical course or progression of a disease or condition over time, anticipate likely complications and prevent them.

For adults with dental implants, will diabetes significantly decrease the longevity of the implants? is what type of pico question?

Prognosis questions - To estimate the clinical course or progression of a disease or condition over time, anticipate likely complications and prevent them.

What does a higher N (sample size) imply?

STATS PRINCIPLE #2: The higher the N (sample size), the more reliable your results.