Biostats

ANOVA assumptions

(1) The samples are random and independent of each other (2) The populations are normally distributed (3) The populations all have the same variance. 4. residuals should be normally distributed

Hill Criteria

-from Sir Austin Hill to measure causality: -strength of association, consistency, specificity of the association, temporal relationship, biological gradient, biologic plausibility, coherence, experiment, analogy

negative skew

A curve or distribution of scores that has extreme scores below the mean that are atypical of the majority of scores. More towards the right

Strength of association

Exposure and resulting outcome with greater likelihood

Specificity

Factor is consistently implicated

Type 2 error

Failing to reject the null hypothesis when it is false (false negative)

Reproducibility

Systematic error assesses

Measurement error

The difference between a measured value and the true value.

No evidence rating system

What are limitations of Pubmed, SCOPUS, OVID

evidence summaries of individual studies

What studies are within level 6 of evidence

Sampling Frame

a list of individuals from whom the sample is drawn

Type I

alpha is associated with what error

Unbiased estimator

any sample statistic obtained from a randomly selected sample that equals the value of it respective population parameter

Patient cohort explorer

epic based deidentified data base, asses patient populations for clinical trials

predictive validity

evaluates the extend to which a measurement can predict future outcomes

Temporality - must be met

exposure precedes disease

Face validity

extent to which respondents can tell what the items are measuring (surface level)

Consitency

factor repeatedly appears in many circumstances

Type II error (beta)

failing to reject a false null hypothesis (false negative)

Discovery

find something unexpectedly stupid definition check ;)

t-test independent

if the means come from two separate, unrelated groups (treatment vs control)

Dose-response

increased exposure leads to increased disease risk

Recall bias

occurs when there is systematic difference in accuracy of the recollections retrieved by study participants. Occurs often in retrospective studies

Experimental data collection

outcome in control vs intervention, controlled trial, lifestyle, educational intervention

alpha

predetermine significance level threshold used to assess the statistical significance of the results

Do no harm

primum non nocere

power

probability of detecting an effect

Kurtosis

pulled up or down

Numerical data

quantitative, numbers

Berkson's Bias

refers to selection bias created by choosing hospitalized patients as the control group

Inter-individual variability

variability due to differences among subjects due to genetics, physiology, immune status, diet, etc.

endogeneity

variable is correlated with both treatment and outcome, biased estimates

Confounding bias

variable that influences both the dependent and independent variable, spurious association

independent variable

variable that is manipulated

clinical significance

whether or not the difference was meaningful for those affected

Discrete data

whole numbers, office visits etc

How to calculate Relative Risk Reduction

1-relative risk Relative risk = (a/total)/ (c/total)

Paired t-test assumptions

1. Random Sample 2. Independent Observations 3. The differences are normally distributed

interdecile range

90%-10% (difference in first and ninth decile)

Case Study (single patient's unique condition is analyzed in great detail to gain insights into their symptoms and causes)

A 40-year-old male patient presents to the ambulatory clinic with a unique constellation of symptoms, including severe episodic headaches, unusual skin pigmentation, and fluctuating blood pressure readings. The patient's symptoms do not align with common diagnostic categories. Intrigued by this atypical presentation, a team of physicians conduct a comprehensive evaluation, including detailed medical history, extensive laboratory tests, genetic analysis, and imaging studies. The objective is to meticulously document and analyze every aspect of the patient's condition in hopes of reaching a diagnosis and understanding the underlying pathophysiology. What type of study is being conducted?

Nominal

A T-shirt company is interested in whom is buying their clothing. They gather information including age and gender of their buyers. In this case, which of the following measurement scales would gender be?

6

A case-control study is conducted to investigate the association between a high-sugar diet and type 2 diabetes. Of 100 patients with type 2 diabetes, 60 reported a high-sugar diet. Among 100 control patients without diabetes, 20 reported a high-sugar diet. What is the odds ratio for developing type 2 diabetes associated with a high-sugar diet?

Ascertainment bias

A clinical trial aims to evaluate the effectiveness of a new antihypertensive drug. The participants are selected from a list of patients who have visited a cardiology clinic in the past year. Which type of bias is most likely to affect the study's findings?

0.67

A cohort study investigates whether regular exercise reduces the risk of developing diabetes. Out of 1,000 individuals who exercise regularly, 50 developed diabetes. In a comparison group of 1,000 individuals who do not exercise regularly, 150 developed diabetes. What is the relative risk reduction for developing diabetes in the exercise group?

Systematic Review

A collaborative team of surgical researchers and clinical experts seek to evaluate the efficacy of various surgical techniques used in the treatment of knee osteoarthritis. They meticulously searched several medical databases to collect all relevant published studies, including randomized controlled trials, observational studies, and case series. The inclusion criteria were strictly defined to select studies that specifically focused on surgical interventions for knee osteoarthritis. The team then rigorously analyzed and compared the outcomes, complications, recovery times, and patient satisfaction rates reported in these studies. The objective was to provide a thorough and unbiased assessment of the most effective surgical techniques for this condition. What type of study was conducted?

Retrospective Cohort (looking back at historical data of a group of patients to study the progression and outcome of disease over time)

A group of epidemiologists at a large healthcare system conducted a study to analyze the progression of Type 2 diabetes over the last 20 years. They accessed the health records of 2,000 patients who were diagnosed with Type 2 diabetes between 2000 and 2005. These records provided comprehensive data on each patient's treatment regimen, blood sugar levels, HbA1c readings, complications, and lifestyle factors over two decades. By examining this historical data, the researchers aimed to identify patterns in disease progression, factors influencing patient outcomes, and the long-term effectiveness of various treatment approaches. What type of study was conducted?

Cohort

A group of epidemiologists at a national research institute embarked on a long-term study focusing on the progression of early-onset Alzheimer's disease. They recruited 300 patients who had recently been diagnosed with Alzheimer's and began tracking their health status from the initial stages of the disease. The participants underwent regular cognitive assessments, neurological examinations, and brain imaging at set intervals over a period of ten years. By following these patients from the onset of the disease, the researchers aimed to document the natural history of early-onset Alzheimer's, including cognitive decline patterns, quality of life changes, and the effectiveness of various treatments over time. What type of study was conducted?

Case Control

A group of patients with type 2 diabetes mellitus is matched with patients without the disease. Their consumption of foods with a large amount of high fructose corn syrup over the next five years is compared. On the basis of this information, researchers compare the odds of developing type 2 diabetes mellitus between the 2 groups. What kind of study does this best represent?

Cohort

A group of people with exposure to asbestos are followed for 20 years with a standard set of measures to determine the potential increased risk of development of lung diseases and mesothelioma. Which of the following best describes this study design?

Kappa statistic

A measure of the degree of nonrandom agreement between observers of the same categorical variable. higher number/higher agreement

unblinded non-controlled trial

A medical research center launched a pilot study to explore the potential benefits of an innovative therapy for patients with severe, treatment-resistant depression. The study enrolled 50 patients who had not responded to traditional treatments. Each participant received the new therapy, which combined pharmacological treatment with a novel psychotherapy technique. As this was an exploratory study to primarily assess the feasibility and initial efficacy of the approach, there was no comparison group. All participants and researchers were aware of the treatment being administered, with the main focus being on monitoring the changes in depression symptoms and overall mental health over six months. What type of study was conducted?

test-retest reliabilty

A method of determining the reliability of a test by comparing a test takers score on the same test taken on separate occasions (same participant)

Linear Regression

A method of finding the best model for a linear relationship between the explanatory and response variable. relationship between continuous outcome and one or more predictor variables Residuals are independent, variance constant Highly impacted by outliers

Phase 1

A new drug is created to combat daytime sleepiness. The purpose of the study is to determine side effects and provide pharmokinetic data. Based on this information, in which phase of clinical trials is this new drug?

Clinical Significance

A new study reports that aspirin marginally lowers the risk of heart attacks with a p-value of 0.20. As a physician, which aspect should be of more concern?

Double blind randomized control study

A pharmaceutical company conducted a study to test the efficacy of a new antihypertensive drug. In this study, 800 participants with hypertension were recruited from several medical centers. They were randomly assigned to receive either the new drug or a placebo. Both the medication and the placebo were designed to look identical, and neither the participants nor the research team had access to the information about who received which treatment. This setup was maintained throughout the trial period, during which various health parameters of the participants were regularly monitored. The identity of the treatment groups was revealed only after the completion of the study and the collection of all necessary data. What type of study was being conducted?

Information bias

A prejudice in the data that results when either the respondent or the interviewer has an agenda and is not presenting impartial questions or responding with truly honest responses, respectively. Loss of records

2

A public health study assesses the risk of hypertension in relation to obesity. Among 600 obese individuals, 180 are diagnosed with hypertension. In a comparison group of 400 non-obese individuals, 60 have hypertension. What is the relative risk of hypertension in obese individuals compared to non-obese individuals?

Sampling Variability

A public health survey uses systematic sampling to select participants from a city's population. The starting point is chosen randomly, and every tenth person on the city's resident registry is selected. This method of sampling helps to reduce which of the following?

Double blind randomized controlled study

A randomized controlled trial where both the participants and the researchers are blinded to the treatment allocation until the end of the study

Retrospective Cohort

A research team at a cardiovascular institute conducted a study to investigate the long-term outcomes of patients who had undergone coronary artery bypass graft (CABG) surgery. They reviewed the medical records of 500 patients who received CABG surgery at their hospital from 1995 to 2000. The study focused on tracking the post-surgical health outcomes of these patients, including rates of recurrent cardiac events, subsequent surgical interventions, and overall mortality rates over the prior 20-year period. The aim was to assess the long-term effectiveness and complications associated with CABG surgery in these patients. What type of study was conducted?

Convenience

A research team conducting a study on smoking habits decides to survey individuals entering a shopping mall over a weekend. They approach individuals who appear willing to participate. Which type of sampling is being used?

Case Control

A research team conducts a study at a large urban hospital where they meticulously review the medical records of 300 patients diagnosed with lung cancer over the past five years. These patients are matched based on age, gender, and socioeconomic status with 300 patients who visited the same hospital but did not have lung cancer. The researchers analyze various factors such as smoking history, occupational exposures, and family history of cancer in both groups to identify potential risk factors associated with lung cancer development. What type of study is being conducted?

Simple Random

A research team is designing a study to assess the prevalence of diabetes in a large urban population. As a sampling frame, they use an electoral register to select participants randomly. This approach is an example of which of the following sampling methods?

Type I errror

A researcher includes multiple statistical tests in the same experimental project. This can increase the chance of making a...

High Sensitivity

A researcher is interested in creating a new screening test for lymphoma. Which of the following is the most important characteristic of the new test?

increase the sample size

A researcher wants to determine the effects of a vegan diet on mean blood pressure control. The power of the study is 65% and the researcher wants to increase the power to 90%. Which of the following strategies will accomplish the researcher's goal?

Clinical trials offer intervention while observational studies do not

A researcher wants to evaluate the effect of nicotine use on the severity of Major Depressive Disorder. She debates between a clinical trial and an observational study. Which of the following best describes the primary difference between the two types of studies?

Meta analysis

A researcher wishes to establish the connection between hypertension and peripheral arterial disease. He performs a literature review and identifies 200 relevant studies. Of the 200 studies, 38 meet the stringent quality and design criteria he has established for inclusion in his assessment. Which of the following best describes the type of study that is being performed?

Relative Risk

A retrospective cohort study is conducted to explore the link between exposure to air pollution and the incidence of asthma in children. The study includes two cohorts: children living in high pollution areas and those in low pollution areas. Researchers retrospectively collect data on asthma development over the past 10 years. Which measure is most suitable for analyzing the association between air pollution and asthma in this study?

5

A retrospective study investigates the relationship between sun exposure and melanoma risk. In a group of 200 individuals with high sun exposure, 40 developed melanomas. In a control group of 800 individuals with low sun exposure, 40 developed melanomas. What is the odds ratio for developing melanoma associated with high sun exposure?

Null Hypothesis (H0)

A statement of "no difference."

Survival analysis

A statistical analysis that models time to event data - should address competing risks

Chi Square

A statistical test is used to determine whether the difference in the prevalence of depression between men and women is statistically significant. Which test is most likely to be utilized?

5

A study aims to evaluate the association between long-term use of mobile phones and the development of brain tumors. Of 500 participants who have used mobile phones for over 10 years, 10 developed brain tumors. Among 500 participants with less than 5 years of mobile phone use, 2 developed brain tumors. What is the relative risk for developing brain tumors in long-term mobile phone users?

confounding bias and adjust for sunscreen use

A study aims to identify the correlation between sun exposure and the risk of developing melanoma. However, it does not account for the use of sunscreen by participants. What type of bias could this introduce, and how can it be corrected?

Sample

A study aims to investigate the prevalence of obesity in a small town but only includes volunteers from a local fitness center. Which term best describes the group of volunteers?

3

A study examines the relationship between high cholesterol and the development of coronary artery disease (CAD). Of 200 participants with high cholesterol, 60 developed CAD. In a control group of 300 participants with normal cholesterol levels, 30 developed CAD. What is the relative risk of developing CAD for those with high cholesterol compared to those with normal cholesterol?

Statistical significance but not necessarily clinical significance

A study finds that the p-value for the association between caffeine consumption and reduced risk of Alzheimer's disease is 0.04. How would you interpret this finding?

Cohort

A study follows a group of non-smokers for five years to determine how many develop lung cancer. What type of study is this?

Relative Risk

A study is being designed to assess the relationship between high dietary salt intake and the development of hypertension. Researchers plan to follow two groups of participants over 5 years: one with high salt intake and one with low salt intake. Which measure of association is most appropriate for analyzing the risk of developing hypertension in this study?

6

A study is conducted to assess the association between exposure to a certain pesticide and the development of a rare skin cancer. Out of 100 individuals with skin cancer, 40 had exposure to the pesticide. Among 100 control individuals without skin cancer, 10 had exposure to the pesticide. What is the odds ratio for developing skin cancer associated with pesticide exposure?

IT shows no more than a 5% chance that the results occured by chance

A study is conducted to assess the relationship between childhood trauma and subsequent developement of unexplained pain symptoms in adulthood. The alpha value is set at 0.05. The results of the study show a significant difference. Which of the following is most correct about the probable p-value of this study?

0.5

A study is conducted to determine the effect of a new drug on reducing the incidence of stroke. Out of 500 patients taking the drug, 25 experienced a stroke. In a control group of 500 patients not taking the drug, 50 experienced a stroke. What is the odds ratio for having a stroke in the group taking the drug compared to the control group?

correlation

A study reports a strong relationship between sugar intake and dental cavities but doesn't prove that sugar causes cavities. What term describes this relationship?

Case Control

A study that compares two groups (case and control). This identifies risk factors. Compares patients with an outcome to those without to look for prior risk factors. Best for rare diseases, identifies odds ratios. Cannot determine risk of disease

Systematic Review

A team of medical researchers embarked on an ambitious project to synthesize existing knowledge about multiple sclerosis (MS). They searched through various medical databases and journals to gather all available research studies, clinical trials, and review articles related to MS. The criteria for selecting the studies included relevance to MS pathophysiology, diagnosis, treatment, and patient outcomes. The team meticulously evaluated the quality of each study, extracted key data, and then synthesized the findings to provide a detailed, comprehensive overview of the current state of medical knowledge on multiple sclerosis, including advancements in treatment and emerging therapeutic strategies. What type of study was conducted?

Cross sectional

A team of researchers wants to investigate the prevalence of diabetes in a certain community. They take a snapshot of the current health status of adults in the community. What type of study is this?

Paired t-test

A test designed to determine the statistical difference between two groups' means where the participants in each group are either the same or matched pairs. (how one persons blood pressure changes)

Triple blind randomized control

A trial was conducted to assess the efficacy of a novel cholesterol-lowering drug. In this study, 1,000 participants with diagnosed hypercholesterolemia were enrolled. They were randomly assigned to either receive the new drug or a placebo. The pills were indistinguishable in appearance to ensure blinding. Not only were the participants unaware of their group assignment, but the clinical staff administering the medication and the team conducting the periodic health assessments were also unaware. Furthermore, the data collected during the trial was anonymized and analyzed by a separate team of statisticians who were also unaware of group assignments. This setup was maintained throughout the study to eliminate bias and ensure the integrity of the data. What type of study was conducted?

instrumental variable

A variable that is strongly associated with the treatment assignment but not related to outcome except with its relationship to treatment. Should be strongly related to treatment assignment. should only be related to outcome through treatment not related to risk factors of outcome or have direct effect on outcome Example coin toss

In meta analysis we do statistical analysis of outcomes

How does meta analysis differ from systematic review

Odds Ratio

Ideal for estimating association between exposure and outcome at a single point in time. Useful in cross-sectional studies, likelihood of exposure, case-control study

no

If alpha equal 0.05 and the p-value equals 0.32, is there a stastically significant difference between the 2 values being tested?

Detection bias

If an investigator includes blinding in her clinical trial, which of the following does she seek to reduce?

Berkson's

In a case-control study investigating factors associated with lung cancer, researchers use patients from a hospital as controls. This approach is most susceptible to which of the following biases?

Odds Ratio

In a cross-sectional study aimed at assessing the relationship between high-fat diets and the presence of coronary artery disease (CAD), participants are surveyed for their dietary habits and tested for CAD at a single point in time. To determine the association between high-fat diets and CAD in this study, which statistical measure is most appropriate?

There was not a control group

In a group of 100 cigarette smokers, 2 developed oropharyngeal cancer over a 10-year follow up period. In this study, which of the following factors most limits the ability to conclude that cigarette smoking is associated with an increased risk for oropharyngeal cancer?

Case Study

In a leading medical center, a 45-year-old female patient presented with a perplexing array of symptoms, including intermittent episodes of extreme fatigue, unexplained muscle weakness, and a pattern of skin rashes not corresponding to any known dermatological conditions. Intrigued by this atypical presentation, a team of multidisciplinary specialists conducted an exhaustive investigation. This included comprehensive diagnostic tests, in-depth genetic analysis, and extensive consultations with experts in rare diseases. The goal was to document this unique case in detail, both to provide appropriate treatment for the patient and to contribute new insights to medical knowledge, given that these symptoms had not been previously reported in medical literature. What type of study was conducted?

Single-blind controlled trial

In a major hospital, a study was initiated to compare the effectiveness of a new drug for treating advanced heart failure against the current standard treatment. The trial involved 600 patients diagnosed with advanced heart failure. These patients were divided into two groups: one receiving the new experimental drug and the other receiving the standard treatment for heart failure. While the patients did not know which treatment they were receiving, the researchers and healthcare providers were fully aware of the treatment assignments. This information was necessary for closely monitoring potential side effects and managing the patients' overall care. The study's primary goal was to evaluate the efficacy and safety of the new drug compared to the standard treatment. What type of study was conducted?

Case - Control

In a major urban hospital, a research team initiated a study to explore potential risk factors for hypertension. They selected 200 patients diagnosed with hypertension and matched them with 200 patients from the same demographic background but without hypertension. The researchers conducted detailed interviews and reviewed medical records to gather information about each participant's lifestyle, diet, family medical history, and environmental exposures. The aim was to compare the two groups to identify any significant differences in these factors that might contribute to the development of hypertension. What type of study was conducted?

68% if it says on one side divided in half

In a normally distributed curve, one standard deviation from mean on both sides would account for what percent of data

99.7% if it says on one side divide by half

In a normally distributed curve, three standard deviations from mean on both sides would account for what percent of data

95% if it says on one side divide by half

In a normally distributed curve, two standard deviations from mean on both sides would account for what percent of data

The probability is greater than 5% that this difference could occur by chance alone

In a randomized, double-blind trial comparing losartan and lisinopril to prevent end stage renal disease (ESRD), 45% of patients receiving losartan developed ESRD in the following 10 years, while 55% of patients receiving lisinopril developed the disease. The authors of the study reported that P>0.05. Which of the following most correctly described the significance of this P value?

Case Series (group of patients with a specific condition are examined in detail to gather comprehensive information about the disease progression)

In a specialized research center for genetic disorders, a team of geneticists conducted an in-depth investigation of 15 patients diagnosed with a rare, autosomal recessive genetic disorder not widely documented in medical literature. The study involved a thorough review of the patients' medical histories, including the onset of symptoms, progression of the disease, response to treatments, and any genetic testing results. The researchers' objective was to compile a comprehensive profile of the disorder's clinical course and identify any common patterns or anomalies in its progression. What type of study was conducted?

Non-probability

In a study aimed at understanding the dietary habits of urban dwellers, researchers decided to use social media platforms to recruit participants. They post an online survey on various city-focused Facebook groups and Twitter feeds, inviting members to share their eating patterns and preferences. The survey respondents are self-selected from those who see the post and choose to participate. What type of sampling is being utilized?

4

In a study examining the link between alcohol consumption and liver cirrhosis, 200 out of 1,000 participants who consume alcohol daily developed cirrhosis. Among 1,000 participants who do not consume alcohol, 50 developed cirrhosis. What is the relative risk of developing cirrhosis in alcohol consumers compared to non-consumers?

cluster

In a study examining the link between diet and heart disease, researchers randomly select three community centers from a list of neighborhood community centers. All persons attending the selected community centers were then invited to participate. This approach exemplifies which type of sampling?

Selection Bias

In a study investigating the association between exercise frequency and hypertension, a researcher uses a list of gym members as the sampling frame. This approach is most likely to introduce which of the following types of bias?

Generalizability error

In a study on mental health, the participants are all college students. The researchers then generalize the findings to the entire population. What type of error is this? Correct!

3

In a study on the effect of air pollution on asthma incidence, 300 out of 1,000 residents in a highly polluted area developed asthma, while in a less polluted area, 100 out of 1,000 residents developed asthma. What is the relative risk of developing asthma in the highly polluted area?

Selection bias move to random sampling

In a study on the effectiveness of a new anti-hypertensive medication, only patients with severe hypertension are included. What type of bias could this introduce, and how can it be corrected?

Single blind randomized controlled study

In a study, 500 participants suffering from chronic migraines were enrolled to evaluate the effectiveness of a newly developed medication. Using a computer-generated random number system, the participants were assigned to either receive the new medication or a placebo. The assignment was concealed from the participants to ensure that their knowledge did not influence the outcome of the study. However, for safety monitoring purposes, the researchers kept a record of who received the medication and who received the placebo, enabling them to intervene in case of adverse reactions or emergencies. What type of study was being conducted?

Case Series (no control or comparison group)

In an advanced research facility, a group of medical researchers initiated a long-term observational study focusing on a rare genetic disorder, currently affecting fewer than 100 known individuals worldwide. They recruited 30 of these patients and planned to closely monitor their health status, treatment responses, and overall progression of the disorder over a decade. This included regular medical check-ups, laboratory tests, and psychological assessments. The study was designed to gather comprehensive data on the natural history of the disease, its varying manifestations in different patients, and the effectiveness of various treatment approaches used over time. What type of study was conducted?

Cross-Sectional

In an urban community, a team of cardiovascular researchers conducted a large-scale study involving 10,000 participants, aged 30-70, randomly selected from various neighborhoods. The study was focused on understanding the relationship between dietary habits and the prevalence of heart disease. Participants completed comprehensive questionnaires about their diet, lifestyle, and medical history. Additionally, they underwent basic health screenings for blood pressure, cholesterol levels, and body mass index (BMI). This data collection occurred at a single community health fair event, aiming to capture a snapshot of the community's health and dietary patterns at that specific point in time. What type of study was conducted?

Probability that an actual positive test is positive

In statistics, sensitivity is...

Up to Date

Includes clinical summaries, recommendations, continuously updated, GRADE system. paragraph format

non-probabilty sampling

Includes several types of sampling methods where the probability of each member of the population being chosen is not known.

Evidence Based medicine

Integrating individual clinical expertise with the best available external clinical evidence from systematic research

Unblinded non-controlled

Involves administering a treatment to a group of patients without a control group for comparison, both the researchers and participants are aware of the treatment being given

triple-blind randomized control trial

Involves blinding at three levels, participants, researchers, and data analysts to prevent bias

Snowball Sampling

Involves hard to reach populations, involves existing study subjects recruiting future subjects from among their acquaintances ex. houseless persons, undocumented, rare diseases

Retrospective Cohort

Involves looking back at a group of individuals to study outcomes, focusing on differences in exposures or characteristics within the cohort itself - not a separate control group

Random Sampling

Involves selecting participants from a population where each has an equal chance of being chosen

Quota Sampling

Involves selecting participants so that the sample has the same proportions of individuals in the target population across certain characteristics (age, sex, etc)

Convenience Sampling

Involves selecting participants who are easily accessible to the researcher without randomization

Case Series

Involves studying multiple patients with a similar diagnosis or treatment. There is NO control group.

Berkson's bias

Is a type of selection bias in hospital-based studies where hospitalized patients are chosen as the study sample.

Population

Large collection of all the actual and potential patients

One proper RCT

Level 7 of evidence

Nonrandomized control, cohort, case studies, case series, case reports

Level 8 of evidence

studies that report positive effects tend to be published more frequently that those that do not.

Limitations of meta analysis

Normal distributions of data and interval and ratio scale data

Mean is reported for...

Continuous data

Measured on a continuum, height, weight, blood pressure, age

Skewed data and ordinal scale data

Median is reported for...

Tuskegee Syphilis Study (1932-1972)

Medical treatment was deliberately withheld to study the course of the untreated disease

unimodal, bimodal, non-modal distributions of data, nominal data (most often, typical, common)

Mode is reported for..

Logistic regression

Models the relationship between a binary outcome and one or more predictor variables. (success/failure) A statistical analysis which determines an individual's risk of the outcome as a function of a risk factor. The outcome of interest has two categories. Larger sample better performance

Positive skewed distribution

Most of the scores are bunched towards the left. The mode is to the left of the mean because the mean is affected by the extreme scores tailing off to the right.

negative skewed distribution

Most of the scores are bunched towards the right. The mode is to the right of the mean because the mean is affected by the extreme scores tailing off to the left.

number needed to treat

NNT = 1/absolute risk reduction

Median

NOT INFLUENCED BY OUTLIERS the middle score in a distribution; half the scores are above it and half are below it

Odds Ratio Equation

OR = ad/bc Cross multiply

Congruence

Observations make logical sense, fit within existing knowledge

Ascertainment Bias

Occurs when certain groups are more likely to be included in the study due to how data is collected or recorded. Example.) Choosing participants from a list of patients who have visited a cardiologist for a new antihypertensive drug

Volunteer Bias

Occurs when individuals who choose to participate in a study are systematically different from those who do not. Related to willingness or availability

Volunteer Bias

Occurs when patients enroll themselves in a study, which results in a non-representative sample

Observer Bias

Occurs when researcher's expectations or knowledge influence the outcome assessment. Not related to how participants are chosen for the study

Survivorship bias

Occurs when the analysis is skewed because it only considers survivors

Selection Bias

Occurs when the method of selecting participants for a study leads to a sample not representative of the population intended to be analyzed. Example.) Gym members being surveyed about exercise frequency and hypertension

Ordinal

Ordered, pain scores, likert chart

Analogy

Other causal associations can be applied if similar

Non-respondent bias

People who fail to respond to an invitation to participate may be systematically different from the rest of the population

sufficent cause

Precedes disease and disease will occur (Tay-sachs)

necessary cause

Precedes disease must be present for disease to occur but does not always result in disease (mycobacterium tuberculosis )

Causal Inference

Process of determining a cause and effect relationship between variables, focuses on association of exposure and outcome

interquartile range

Q3-Q1

Categorical data

Qualitative form of data, nominal or ordinal

Relative risk equation

RR = [a/(a + b)] / [c/(c + d)]

causality

Random assignment can be used to determine

repeatability

Random error assesses

generalizabilty

Random selection can be used to determine

Sampling Variability

Refers to the natural differences between samples drawn from the same population

Association Association is not causation

Relationship between exposure and outcome

Case series - efficient for rare diseases

Researchers are investigating a rare autoimmune disorder and decide to include all known cases within a five-state area. What type of study is this, and why is it appropriate for rare diseases?

Convenience sampling always move to random!

Researchers are studying the effect of diet on blood sugar levels and decide to only include participants who frequently visit a nutritionist. What sampling method is this, and how can it be improved?

Selection Bias randomize your patients (always)

Researchers are studying the effects of a new drug on blood pressure. They notice that individuals who volunteered for the study are generally healthier than the average population. What type of bias is present, and how can it be corrected?

Cohort

Researchers at a public health institute enroll 500 smokers and 500 non-smokers, all of whom are free of respiratory diseases. Participants are aged between 35 and 55 years at the start of the study. Over a period of ten years, they periodically assess the participants to track the development of respiratory diseases, including chronic bronchitis, emphysema, and lung cancer. What type of study is being conducted?

Retrospective

Researchers collect data from hospital records to analyze the outcomes of patients who had a specific surgical procedure five years ago. What type of study is this?

Interval

Researchers collect temperature and humidity data to study their effect on the incidence of heat stroke in a city. What type of data is the temperature?

Cohort

Researchers follow a group of medical students over four years to see if sleep patterns change as they progress through medical school. What type of study is this?

Retrospective

Researchers review various medical records of patients who have developed kidney stones, looking for common medications that might be associated with the condition. What type of study is this?

Ratio

Researchers use a survey to collect data on the number of servings of fruits and vegetables consumed daily by participants. What type of data is this?

Information Bias

Results from inaccuracies in the measurement of exposure or outcome variables. Can be inaccuracies in data collection, measurement, or interpretation

Systematic Sampling

Selecting every nth individual from a list. Example.) selecting every 10th person on the electoral register

simple random sampling

Selection of participants from a larger population in a random manner. EACH individual has an equal chance of being chosen. Example.) Electoral register to select participants, name out of hat type of sample

Standard deviation

Spread of data distribution, raw data from one sample

always positive, describe quantitative variables

Standard deviation characteristics

Intention to treat analysis

Statistical analysis of data from subjects according to the group to which they were assigned despite noncompliance with the study protocol - maintains the benefits of randomization and provides a conservative estimate of treatment effects, reflecting real world scenarioes where not all participants adhere to the assigned treatment

instrumental variable analysis

Statistical technique to address issues of and confounding when estimating the causal effect of treatment or exposure of an outcome.

Target population sampling frame sampling unit sampling method

Steps in conducting a human population study

1. PICO 2. inclusion and exclusion 3. search for articles and assess GRADe 4. statistical analysis (heterogenicity, forest plot, bias) 5. clinical applicability

Steps of reading research

Increases sample size and power to study the effects of interest, statistical rules (objective), answers a question that was not there

Strengths of Meta analysis

Cross Sectional Study

Study that assess data from a population at a specific time to analyze the relationship Retrospective, observational population study. survey based, assess trends and effectiveness most common study design, point in time Can give an odds ratio not relative risk

Systematic review

Study that involves a thorough and methodical compilation and synthesis of existing research on a specific medical condition

Single-Blind randomized controlled study

Study that is randomized controlled trial where the participants are blinded (do not know if they have placebo or medication) but the researchers are aware of the assignment for safety reasons

Recall bias

Systematic error due to differences in accuracy or completeness of recall to memory of past events or experiences.

Ascertainment bias

Systematic failure to represent equally all classes of cases or persons supposed to be represented in a sample.

Cohort

Takes a group without the outcome of interest and follows over time. Can determine which factors produce an outcome of interest. Can determine relative risk. Good for common diseases, determine risk, difficult to conduct, LOSS to Follow up Follows a group of people over time to see how certain factors affect outcomes

Sampling distribution

The distribution of means from a sample, Theoretically this will result in the distribution being around the true population mean

Construct validity

The extent to which there is evidence that a test measures a particular hypothetical construct.

confidence level

The level of certainty that a population parameter exists in the calculated confidence interval.

The mean of the sampling distribution is an unbiased estimate of the true mean in the population

The mean of the sampling distribution is an _________ estimate of the true mean in the ________

dependent variable

The measurable effect, outcome, or response in which the research is interested.

p value

The probability of results of the experiment being attributed to chance.

statistical inference

The process of using data obtained from a sample to make estimates or test hypotheses about the characteristics of a population.

Central Limit Theorem

The theory that, as sample size increases, the distribution of sample means of size n, randomly selected, approaches a normal distribution.

Meta-analysis

This study involves pooling data from multiple studies to assess the overall effectiveness of new treatment. (multiple clinical trials)

Case Control

To study the effects of a new smoking cessation program, researchers compare the success rates of participants in the new program to those in an established program. What type of study is this?

inversely related

Type I and Type II errors are _________ related

Area Sampling

Type of sample that involves dividing a geographical area into smaller sections, then selecting areas to include all individuals in the study

association

Values of one variable tend to occur with certain values of another variable; detected when the conditional distributions differ from the marginal distribution and from each other.

Levenes test

Varience of two groups being compared is approximately equal

P value less than or equal to 0.5

What P value do we reject the null hypothesis

age, gender, socioeconomic factors, comorbitites, race, season, lifestyle

What are common confounding variables

patient characteristics, publication status, language used, research period

What are examples of how to define inclusion and exclusion criteria

Cause precedes the effect cause related to effect can find no other explanation for the effect

What are the conditions needed for causality

Generalizabilty, best practice, patient-centered values, efficacy and effectivness

What are the limitations of evidence

Not always accurate due to data distortion, study bias, lack of hindsight, requires analysis of quality, not rapidly consumed

What are the limitations of primary data

must rely on authors to ensure quality and accuracy of underlying information

What are the limitations of secondary data

Rely on other's analysis of appropriateness

What are the limitations of tertiary data

autonomy, protecting the volunteerism of research subjects, (be honest bruh), accurate definition of risk, potentials for coercion

What are the principles of ethics

Continuous and discerete

What are the two examples of quantitative data

keeping from zero, least squares, can be corrected via square root

What are three reasons why we square each variance

Confidence intervals - measure of precision hypothesis tests - measure of likelihood, p value

What are two ways that capture the uncertainty rising from a study on a sample of participants

low risk of bias, accurate and reliable results, appropriate study design, methods are rigorous and well-described

What distinguishes high quality evidence

some risk of bias, some concerns about reliability and accuracy, study design is limited or marginally well-described

What distinguishes intermediate quality evidence

high risk bias, significant concerns about reliability and accuracy, flawed study design, methods are poor

What distinguishes low quality evidence

patients interventions control (placebo or non-treatment group) outcomes (primary or secondary)

What is PICO

consistent and good quality patient oriented evidence

What is an example of A SORT

inconsistent, limited quality patient oriented evidence

What is an example of B SORT

Consensus, usual practice, expert opinion, disease oriented, case series for studies of diagnosis, treatment, prevention, or screening

What is an example of C SORT

Used for data that is positive or right skewed, will give the geometric mean as the peak

What is logarithmic transformation used for

95%

What is the confidence interval most used for a mean and a proportion

INVERSE

What is the relationship with alpha and beta

Random sampling will allow for true population parameters but if sampling bias exists your sampling distribution may not give true parameters

What is the sampling caution

NONE only percentages

What measure of center and spread is used with categorical nominal

median

What measure of center is used with categorical ordinal

median

What measure of center is used with continuous numerical but skewed data

mean

What measure of center is used with continuous numerical, normal distribution

interquartile range

What measure of spread is used with categorical ordinal

interquartile range

What measure of spread is used with continuous numerical but skewed data

standard deviation

What measure of spread is used with continuous numerical, skewed distribution

RANDOM SAMPLING Simple random stratified systematic multistage or clustering

What methods of sampling produce probability samples

P value greater than 0.05

What p value do we FAIL to reject the null hypothesis

Non-Random Sampling Judgment quota convenience snowball

What sampling methods do not produce probability samples

meta analyses, randomized controlled trials

What studies have the least potential for bias

expert opinon, cross section, cohort

What studies have the most room for bias

T-test

What test is used when comparing difference of means in 2 groups.

Continuous

What type of data is age, blood pressure, height, weight, blood glucose, interval

Nominal/Categorical

What type of data is blood type

discrete

What type of data is number of patients, number of cases, number of deaths. number of visits, ratio, score

Type 1 error (alpha)

What type of error rejects the null hypothesis when it is true (false positive)

animal research, in vivo

What type of study is at the bottom of levels of evidence

Meta-analyes

What type of study is at the top of most levels of evidence

systematic Reviews of Randomized control trials

What types of studies are within level 1 of evidence

Meta-analysis of RCT

What types of studies are within level 2 of evidence

Evidence summaries developed from systematic review

What types of studies are within level 3 of evidence

Guidelines developed from systematic reviews

What types of studies are within level 4 evidence

meta syntheses, qualitative study

What types of studies are within level 5

systematic error

When error occurs between observer variation

Random error

When error occurs within observer variation

With continuous data

When is a bland altman plot used

categorical data (high/low, positive/negative)

When is a kappa statistic used

Phase IV

Which of the following clinical trials includes post-marketing surveillance and provides data on adverse events?

Odds ratio

Which of the following is an association between an exposure and an outcome?

t-test

You are reviewing a study that compares the mean blood sugar levels between patients on two different medications. What statistical test should have been used?

Cross sectional but cannot establish causation

You encounter a study that reports current smoking rates among adolescents in a specific city. What type of study is this, and what is its primary limitation?

Causation

a change in one variable makes another variable change

central tendency

a measure that represents the typical response or the behavior of a group as a whole

Propensity score matching

a method of pairing individuals for assignment to a treatment and control condition based upon a combination of scores on participant variables. pseudo-randomization

ordinal variable

a qualitative variable that incorporates an ordered position, or ranking, a pain scale, likert scale

statistical significance

a statistical statement of how likely it is that an obtained result occurred by chance (p-value)

t-test

a statistical test used to evaluate the size and significance of the difference between two means

Sample

a subset of the population, allows to make generalization possible

point estimate

a summary statistic from a sample that is just one number used as an estimate of the population parameter

Confounding variable

a third variable associated with both the explanatory variable and response variable

Type III error

a type of error possible with one-tailed tests in which a decision would have been to reject the null hypothesis, but the researcher decides to retain the null hypothesis because the rejection region was located in the wrong tail

Ratio Variable

a variable that meets the criteria for an interval variable but also has a meaningful zero point

stratified sampling

a variation of random sampling; the population is divided into subgroups and weighted based on demographic characteristics of the national population

margin of error

an amount (usually small) that is allowed for in case of miscalculation or change of circumstances.

criterion related validity

assesses the degree to which a measurement predicts or correlates with an external criterion

concurrent validity

assesses the relationship between a new measure and an existing measure

variance

average of standard deviation

treatment effect

cause of differences between treatments and control groups - primary goal of an RCT is to measure the .....

Dynamed

clinical summaries, drug info and calculator, uses THREE level system to rate articles, bullet points

Institutional Review Board (IRB)

committee of administrators, scientists, and community members that reviews proposals for research involving human participants

Tertiary data

compilations of primary and secondary sources that do not offer original interpretation Database, dictionaries, encyclopedia Pubmed, ovid, SCOPUS Rapid consumption

Human Research protection program

comprehensive and organized system of human participants in research, to share responsibility to protect research participants and do the right thing

Intra-rater reliability

consistence of measurements when performed by SAME rater or observer

parallel forms reliability

consistency between/among alternate versions of the same instrument; e.g. creating 2 parallel forms of a questionnaire (with difficult questions) and both tests show correlation

inter-rater reliability

consistency of measurements or assessments when performed by different observers

Secondary data

description, interpretation of primary resources textbooks, review articles, quick to read

range

difference between largest and smallest value

Statistical heterogeneity

diversity of effects across primary studies included in a meta-analysis

CONSORT diagram

documents flow of people into and out of a study sample

interviewer bias

effects of interviewers on respondents that lead to biased answers

Interval estimate

interval or range of values within which the true population value is estimated to fall

Interval variable

is a variable that has numbers as its values; the distance (or interval) between pairs of consecutive numbers is assumed to be equal temperature

Observer bias

knowledge of the hypothesis, disease status, exposure status, can influence data recording

opinion of authorities, reports of expert committee

level 9 of evidence

platykurtic

means flattened down, less variability

Leptokurtic

means lifted up, less variability

Variability

measure for the dispersion or spread of data in a distribution, 0 to infinity, never negative

Correlation

measure of mutual correspondence between two variables, influence one variable has on the other.

Positive skew

mode is to the left of the mean, left curved

Wilcoxen signed-rank test

nonparametric version of a paired t test - ordinal data

Mann whitney U test

nonparametric version of independent t test - two independent samples of ordinal or skewed data. Each observation should belong to one and only one sample

Non-sampling error

occurs when the sample data are incorrectly collected, recorded, or analyzed - can produce biased study results selection bias, information bias, confounding

Type 1 error

rejecting the null hypothesis when it is true (false positive)

Counterfactual

relating to or expressing what has not happened or is not the case. If A had not occurred B would have not occurred

Translational research

research that uses knowledge derived from basic research to develop and test solutions to real-world problems

Observational data collection

studies on samples or populations not controlled by a researcher, case series, reports, cross sectional

Research

systematic investigation into and study of materials and sources in order to establish facts and reach new conclusions

internal consistency reliability

the assessment of reliability using responses at only one point in time

internal validity

the degree to which changes in the dependent variable are due to the manipulation of the independent variable. Chance, bias, cofounding (truth in the study)

sampling distribution

the distribution of values taken by the statistic in all possible samples of the same size from the same population

validity

the extent to which a test measures or predicts what it is supposed to

constant validity

the extent to which a test samples the behavior that is of interest - can the study measure the entire domain of interest

statistical validity

the extent to which statistical conclusions derived from a study are accurate and reasonable

External validity

the extent to which the results of a study can be generalized to other situations and to other people (truth in real life, or population)

construct validity

the extent to which variables measure what they are supposed to measure

weighted mean

the mean of a data set whose entries have varying weights, two or more groups of scores, Mean(samples)Xn/n

Mode

the most frequently occurring score(s) in a distribution

confidence interval

the range of values within which a population parameter is estimated to lie

Standard error

the standard deviation of a sampling distribution

Standard Error

the standard deviation of a sampling distribution (sample means)

Clinical Research

the study of individuals or small groups of people who seek help from mental health professionals or other social scientists

Alternate Hypothesis (Ha)

there IS a relationship among variables or a difference

Scholarship

Academic study, learning of high level

Relative Risk

An appropriate measure in a retrospective cohort study, cohort studies. Exposed v unexposed

Relative Risk

An observational study is conducted to evaluate the association between regular physical exercise and the incidence of type 2 diabetes. The study includes individuals who exercise regularly and those who do not, tracking the development of diabetes over a 10-year period. Given the design of this study, which statistical measure should be used to quantify the association?

Meta-analysis

An oncological research group at a leading university conducted a study to evaluate the efficacy of a new chemotherapy drug designed for lung cancer treatment. They systematically searched multiple academic databases to gather data from all available clinical trials that tested this drug. The criteria for inclusion were randomized controlled trials involving lung cancer patients receiving the new drug, with outcomes measured in terms of tumor response, survival rates, and side effects. The research team then used statistical techniques to aggregate and analyze data from these diverse studies, which varied in size, duration, and patient demographics. The aim was to provide a more precise estimate of the drug's effectiveness by synthesizing the results from these individual trials. What type of study was conducted?

per protocol analysis

Analysis of all subjects in a study who completed the study as per the protocol, remove data of participants who did not adhere to protocol.

Sampling Error

Any sample may not behave quite the same as the larger population from which it was drawn. Group of people in the study is just one of the many possible samples from the population of interest. Random Error

Risk Factor

Attribute, behavior, or exposure that increases the probability of disease when present and active

Type II

Beta is associated with what error

Ethics

Branch of philosophy dealing with values relating to human conduct, with respect to the rightness and wrongness of certain actions and to the goodness and badness of motives.

Case Study

Brief descriptive report, describes treatment outcome, disease symptoms, retrospective, no comparison group Typically focuses on a detailed examination of a single patient or a very small group of patients

Case Report

Brief, objective report of a clinical characteristic or outcome from a single clinical subject or event. Retrospective, no statistical analysis n = 1 For example: 23-year-old man with treatment resistant TB; no control group

NO

Can statistical significance infer clinical significance

Mean, median mode at 50th percentile, tails of curve are asymptotic, area under curve equals one

Characteristics of normally distributed data

ANOVA

Compares mean values of a contributes variable for multiple categories/groups (3 groups)

Survivorship bias

Concentrating on the people or things that "survived" some process and inadvertently overlooking those that didn't because of their lack of visibility.

biological plausibility

Consistent with current understanding of the biological processes

Experiment

Controlled research supports causal association

Derived Data

Data calculated from other data. Raw data are measured or collected. Odds, probability, scores/indices

Nominal data

Data which consists of names, labels, or categories.

Causation

Direct effect, rarely demonstrated

Stratified Sampling

Divide population into subgroups based on certain characteristics (age, gender) and randomly select from subgroup

Cluster Sampling

Divides a population into clusters and randomly select entire clusters for a study. Example.) Selection of a couple community centers and studying all of the people from that center

nonparametric tests

Do not make any assumptions about the population and are used when parametric tests cannot be. Use for skewed population, small samples, nominal or ordinal data

YES

Does Scopus have a citation metric

ePocrates

Drug information, disease info, continuous updates NO formal evidence rating

Micromedex

Drug information, toxicology data, alternative medicine, FIVE level evidence rating sysem

Rubin causal model

Emphasizes that individuals selected into either treatment or non-treatment groups have potential outcomes in both states. emphasizes testing that groups are SYSTEMATICALLY similar

meta-analysis

Endocrinologists specializing in diabetes treatment undertake a comprehensive project to evaluate the effectiveness of a novel oral diabetes medication. They gather and analyze data from thirty different randomized controlled trials conducted globally over the past decade. These studies vary in size, ranging from 50 to 500 participants, and compare the new medication's ability to control blood glucose levels against existing treatments and placebos. The researcher's goal is to aggregate the results to provide a clear, quantitative assessment of the medication's efficacy. What type of study is being conducted?

Chi Square

Examines the association between categorical variables, used when the outcome variable is categorical or to assess the distribution of categorical variables between groups. Expected cell counts of at least 5 each - categorical variable

baseline characteristics

Factors that describe study participants at the beginning of the study (e.g., Age, sex, disease severity). In comparison studies, it is important that these characteristics be initially similar between groups; if not balanced or if the imbalance is not statistically adjusted, these characteristics can bias the study results.

Primary data

Firsthand data collection, research studies and diaries

Nominal

Has a name, blood type, race