Podcast
Questions and Answers
What was the primary outcome measured in the study investigating the effectiveness of mirtazapine?
What was the primary outcome measured in the study investigating the effectiveness of mirtazapine?
What statistical method was used to analyze the continuous outcome data in the study?
What statistical method was used to analyze the continuous outcome data in the study?
What was the mean BDI-II score for the active treatment group at 12 weeks of follow-up?
What was the mean BDI-II score for the active treatment group at 12 weeks of follow-up?
What was the mean difference in BDI-II scores between the active treatment group and the placebo group?
What was the mean difference in BDI-II scores between the active treatment group and the placebo group?
Signup and view all the answers
How many adults were recruited for the study on mirtazapine?
How many adults were recruited for the study on mirtazapine?
Signup and view all the answers
What are the three most common types of outcome measurement seen in mental health research papers?
What are the three most common types of outcome measurement seen in mental health research papers?
Signup and view all the answers
Which statistical concept is important for understanding the reliability of study conclusions?
Which statistical concept is important for understanding the reliability of study conclusions?
Signup and view all the answers
In the PICO framework, which component refers to what researchers measure?
In the PICO framework, which component refers to what researchers measure?
Signup and view all the answers
Why is statistics particularly important in mental health research?
Why is statistics particularly important in mental health research?
Signup and view all the answers
Which of the following is NOT one of the steps in the four-step approach to research?
Which of the following is NOT one of the steps in the four-step approach to research?
Signup and view all the answers
What are binary outcomes in the context of mental health research?
What are binary outcomes in the context of mental health research?
Signup and view all the answers
What is the goal of hypothesis testing in statistical analyses?
What is the goal of hypothesis testing in statistical analyses?
Signup and view all the answers
What does a p-value indicate in statistical analysis?
What does a p-value indicate in statistical analysis?
Signup and view all the answers
What is the risk of depression for children with adverse childhood experiences (ACEs)?
What is the risk of depression for children with adverse childhood experiences (ACEs)?
Signup and view all the answers
Which of the following statements is true regarding the odds of depression in children without ACEs?
Which of the following statements is true regarding the odds of depression in children without ACEs?
Signup and view all the answers
What does an odds ratio greater than 1 indicate?
What does an odds ratio greater than 1 indicate?
Signup and view all the answers
Which variable serves as the exposure in the longitudinal study?
Which variable serves as the exposure in the longitudinal study?
Signup and view all the answers
What does a risk ratio of 1 indicate about the comparison between two groups?
What does a risk ratio of 1 indicate about the comparison between two groups?
Signup and view all the answers
Why is odds more commonly used than risk in health research?
Why is odds more commonly used than risk in health research?
Signup and view all the answers
What outcome variable is measured at age 16 in the longitudinal study?
What outcome variable is measured at age 16 in the longitudinal study?
Signup and view all the answers
What is the interpretation of an incidence rate ratio (IRR) greater than 1?
What is the interpretation of an incidence rate ratio (IRR) greater than 1?
Signup and view all the answers
How can Poisson regression models be adjusted in analyzing incidence rates?
How can Poisson regression models be adjusted in analyzing incidence rates?
Signup and view all the answers
Which is the correct null hypothesis statement for comparing outcomes between groups?
Which is the correct null hypothesis statement for comparing outcomes between groups?
Signup and view all the answers
In the context of depression rates among males and females, what were the rates for females aged 16-20?
In the context of depression rates among males and females, what were the rates for females aged 16-20?
Signup and view all the answers
What does a confidence interval (CI) generated around the IRR represent?
What does a confidence interval (CI) generated around the IRR represent?
Signup and view all the answers
How can one disprove the hypothesis that depression always begins during adolescence?
How can one disprove the hypothesis that depression always begins during adolescence?
Signup and view all the answers
What determines the outcome variable when using Poisson regression in mental health research?
What determines the outcome variable when using Poisson regression in mental health research?
Signup and view all the answers
Which rate indicates a higher level of depression between females and males in the provided data?
Which rate indicates a higher level of depression between females and males in the provided data?
Signup and view all the answers
What is the null value that would represent the null hypothesis when interpreting risk or odds ratios?
What is the null value that would represent the null hypothesis when interpreting risk or odds ratios?
Signup and view all the answers
Which statement best describes a p-value in the context of hypothesis testing?
Which statement best describes a p-value in the context of hypothesis testing?
Signup and view all the answers
How should p-values be interpreted regarding their thresholds?
How should p-values be interpreted regarding their thresholds?
Signup and view all the answers
What does a smaller p-value indicate about the null hypothesis?
What does a smaller p-value indicate about the null hypothesis?
Signup and view all the answers
Why is it considered bad practice to interpret p-values based solely on a threshold?
Why is it considered bad practice to interpret p-values based solely on a threshold?
Signup and view all the answers
What does a p-value of 0.035 suggest in comparison to a p-value of 0.051?
What does a p-value of 0.035 suggest in comparison to a p-value of 0.051?
Signup and view all the answers
What does it mean when the p-value approaches 1?
What does it mean when the p-value approaches 1?
Signup and view all the answers
What aspect of p-values is emphasized regarding their reporting?
What aspect of p-values is emphasized regarding their reporting?
Signup and view all the answers
What is the primary effect estimate produced by logistic regression?
What is the primary effect estimate produced by logistic regression?
Signup and view all the answers
Which statistical method is primarily used to compare two rates?
Which statistical method is primarily used to compare two rates?
Signup and view all the answers
How does risk differ from rate in a study context?
How does risk differ from rate in a study context?
Signup and view all the answers
What is a crucial component to include in the denominator to reduce bias?
What is a crucial component to include in the denominator to reduce bias?
Signup and view all the answers
Which of the following can logistic regression accommodate in its analysis?
Which of the following can logistic regression accommodate in its analysis?
Signup and view all the answers
In a cohort study, what is a primary characteristic of how participants are followed?
In a cohort study, what is a primary characteristic of how participants are followed?
Signup and view all the answers
What occurs when participants are in a study for longer observation periods?
What occurs when participants are in a study for longer observation periods?
Signup and view all the answers
What does the term 'immortal time bias' refer to in pharmacoepidemiology?
What does the term 'immortal time bias' refer to in pharmacoepidemiology?
Signup and view all the answers
Study Notes
Session 4: 4th Oct - pt2
-
Introduction to Applied Statistics:
- Covered three common outcome variables in mental health research: continuous, binary, and rates.
- Used practical examples to determine appropriate statistical methods.
- Provided how to interpret results from research papers.
- Reading material available from PSBS0002: Core Principles of Mental Health Research | University College London (talis.com)
- Class preparation included webpages on calculating the mean value, standard deviation, and variance.
-
Objectives:
- Describe the aims of statistical analyses and investigations.
- Comment on three primary outcome measurements in mental health research papers:
- Continuous outcomes
- Binary outcomes
- Count outcomes
- Time-to-event outcomes
-
Why use statistics in mental health?:
- Answer questions about mental health within the population.
- Example questions include:
- How common are mental health problems?
- Who is more likely to develop mental health problems?
- Causes of mental health problems?
- Outcomes of mental health problems?
- Which treatments are effective?
-
PICO Framework:
- Four-step approach to research:
- Patient/problem
- Intervention
- Comparison
- Outcome
- Simple framework for understanding research.
- Example research question: Are adverse childhood experiences (ACEs) associated with psychotic symptoms in adolescence?
- Four-step approach to research:
-
Step 2: Define the Comparison:
- Comparing two or more groups using examples like: Those exposed to risk factors versus those not exposed.
- Example: Comparing treatment effectiveness (e.g., mirtazapine + sertraline vs. sertraline alone).
-
Step 3: Collect & Summarise Data:
- Gathering mental health data on psychotic, depressive symptoms, and other relevant variables.
- Collection often takes place at multiple time points.
-
Step 4: Test for Differences Between Groups:
- Determining the magnitude of differences using effect size or association measures (e.g., mean difference, odds ratio, risk ratio).
- Estimating the results' precision using 95% confidence intervals (CI).
- Testing statistical significance using p-values.
Continuous Outcomes
-
Comparing Two Means:
- Linear Regression
- What does continuous outcome mean? (quantitative trait, e.g., depression, psychotic, eating disorder symptom score, BMI, height, age, blood pressure)
- Example: Research into the efficacy of combining mirtazapine with another antidepressant for resistant depression. Randomizing participants to active treatment (mirtazapine + usual antidepressant) or placebo (placebo + usual antidepressant), and monitoring depressive symptoms over 12, 24, and 52 weeks using the Beck Depression Inventory(BDI-II).
Binary Outcomes
-
Comparing Two Proportions:
- Probability, risks, and odds. Logistic regression.
- Frequentist Definition: Probability = proportion of times that event would occur across numerous trials.
-
Example: Exploring the link between adverse childhood experiences (ACEs) and depression in adolescence.
- Longitudinal study (4000 children aged 8-16) tracking ACEs exposure against depression diagnosis at age 16 using binary variables (exposed/unexposed, depression/no depression).
- Calculate risks, odds, and odds ratios for these groups.
- Calculation of risk and odds ratios, relating the probability to the observed values.
-
Risks:
- Risk of outcome in the exposed = a / (a + b)
- Risk of outcome in the unexposed = c / (c + d)
-
Odds:
- Odds of outcome in exposed = a/b
- Odds of outcome in unexposed = c/d
- Odds ratios = (odds in exposed)/(odds in unexposed)
- Key summary of risk and odds: Ratios close to 1 (or 0) would indicate no difference, ratios greater than 1 could indicate high risk.
Count Outcomes/Rates
-
Comparing Two Rates:
- Incidence rate ratios, Poisson regression.
- Mostly used in cohort studies.
- People with and without known exposures followed over time
- Information Collection: Exposure and outcomes tracked over the observation period. Calculating incidence rates and ratios.
-
Example: Investigating the link between family socioeconomic position in childhood and the development of depressive symptoms.
- Observing a population (e.g., specific birth cohort) to measure socioeconomic status and tracking how many develop depressive symptoms over a period of time. Calculated incidence rates for those with certain exposures against those with no exposures.
Hypothesis Testing and P-values
-
Null-hypothesis:
- There's no difference between groups.
-
P-Values:
- Probability of obtaining results at least as extreme as the observed ones if the null hypothesis were true.
- Low p-value indicates strong evidence against the null hypothesis.
- Interpretation: Smaller p-values indicate stronger evidence against the null hypothesis. Larger p-values indicate weaker evidence against the null hypothesis and may need further investigation.
-
Confidence Intervals:
- Range of likely values for an unknown population parameter (e.g., mean).
- 95% CI range means there's a 95% confidence that the range will include the true value.
Cox Regression Model
- This model explores the relationship between a time-to-event outcome.
- Example: How long until an event occurs (e.g. first episode of depression)?
- Examining how well an external factor predicts how long a person might take to experience that outcome.
Studying That Suits You
Use AI to generate personalized quizzes and flashcards to suit your learning preferences.
Related Documents
Description
Test your knowledge on the study methodologies used in mental health research, particularly focusing on mirtazapine. This quiz covers various aspects such as outcome measurement, statistical methods, and the PICO framework. Enhance your understanding of key concepts that influence study conclusions in mental health studies.