Topnotch Medical Board Prep Preventive Medicine and Public Health PDF April 2024

TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or https://www.facebook.com/topnotchmedicalboardprep/ This handout is only valid for April 2024 PLE batch. This will be rendered obsolete for the next batch sin...

TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or https://www.facebook.com/topnotchmedicalboardprep/ This handout is only valid for April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. SUMMARY MEASURES MEASURES OF CENTRAL TENDENCY MEAN • Most common measure of central tendency; “average” • The sum of all observed values divided by the number of observation • Not useful on a skewed data (abnormal distribution) • For numerical data and symmetric distribution MEDIAN • The value that falls in the middle position when the observations are ranked in order from the smallest to the largest. • If number of observations is odd, the median is the middle number • If it is even, the median is the average of the 2 middle numbers. • Useful on skewed data • For ordinal or numeric data if skewed ✔ GUIDE QUESTIONS In nine families surveyed, the numbers of children per family were 4, 6, 2, 2, 4, 3, 2, 1, and 7. The mean, median, and mode numbers of children per family are, respectively, A. 3.4, 2, 3 B. 3, 3.4, 2 C. 3, 3, 2 D. 3, 4, 3, 2 Answer: D Identify the mean, median, and mode per item. 1. 3, 5, 6, 9, 5 2. 11, 14, 10, 11, 21 3. 8, 12, 7, 10, 8, 9, 11 4. 18, 16, 20, 14, 12, 18 5. 44, 40, 44, 38, 40, 44 Which measure of central tendency is most affected by outliers? A. Median B. Mean C. Mode D. Standard deviation Answer: B Which measure of central tendency is least affected by outliers? A. Median B. Mean C. Mode D. Standard deviation Answer: C PRACTICE MEAN, MEDIAN & MODE • Find the sample mean for the following set of numbers: 12, 13, 14, 16, 17, 40, 43, 55, 56, 67, 78, 78, 79, 80, 81, 90, 99, 101, 102, 304, 306, 400, 401, 403, 404, 405. • Step 1: Add up all of the numbers: • Step 2: Count the numbers of items in your data set. In this particular data set, there are 26 items • Step 3: Divide the number you found in Step 1 by the number you found in Step 2. 3744/26 = 144. https://www.statisticshowto.com/probability-and-statistics/statistics-definitions/sample-mean/ RANGE • Simplest • Computed as the difference between the smallest and the largest values in a set of data. • Ungrouped data: Highest minus lowest • Grouped Data: True upper class limit of the highest class interval minus the true lower limit of the lowest interval PRACTICE MODE • The value that occurs with the greatest frequency in a set of observations • Used in public health statistics (top 10 mortality and morbidity) • For bimodal distribution MEAN, MEDIAN & MODE If you are given a data set please make sure to arrange/rank it first from the lowest observation up to highest, so at least you are ready to identify your median value! Dr. Mann Identify median? 12, 13, 14, 16, 17, 40, 43, 55, 56, 67, 78, 78, 79, 80, 81, 90, 99, 101, 102, 304, 306, 400, 401, 403, 404, 405. Since we have an even observation, we need to add 79 and 80 then divide it by 2 = 79.5 Dr. Mann Identify the mode? 12, 13, 14, 16, 17, 40, 43, 55, 56, 67, 78, 78, 79, 80, 81, 90, 99, 101, 102, 304, 306, 400, 401, 403, 404, 405 The value that occurs in greatest frequency is 78 (unimodal) Dr. Mann MEASURES OF DISPERSION • Measures of dispersion or variation locate the spread of a frequency distribution. • The measures help to describe the spread, or how far from the center the data tend to range. • Describes the variability of the observations • Homogenous o Little difference between adjacent observations • Heterogenous o Observations are scattered around the mean • Dispersion o Dispersion in statistics is a way of describing how spread out a set of data is. When a data set has a large value, the values in the set are widely scattered; when it is small the items in the set are tightly clustered. o Very basically, this set of data has a small value/small variance: § 1, 2, 2, 3, 3, 4 o …and this set has a wider one/high variance: § 0, 1, 20, 30, 40, 100 VARIANCE • Average of the squared deviation of the mean. • Each deviation should be squared first before taking the sum. • Always a positive value • Statisticians tend to consider variance a primary measure & use it extensively • Affected by outliers • Best for symmetric data • measures how far a set of numbers are spread out. LOW VARIANCE: • indicates that the data points tend to be very close to the mean (expected value) and hence to each other HIGH VARIANCE: • indicates that the data points are very spread out around the mean and from each other. • • • • STANDARD DEVIATION Square root of variance Most common and useful measure because it is the average distance of each score from the mean or how much each data value deviates from the mean Requires numeric data Highly affected by outliers • • • • COEFFICIENT OF VARIATION Expresses the standard of deviation as a % of a mean used to compare relative dispersion in one type of data with relative dispersion in another type of data Measures relative variability Used when the units of measurement of variables being compared are different Disadvantages: SD/Mean x 100 • Very sensitive to extreme S2= ∑x2- (∑x)2/ n SD= √𝑆2 observation values. n-1 Standard deviation divided by the mean multiplied by • Based only on extreme values. 100 • Least informative Variance describes the dispersion, but the squared units is harder to interpret. Therefore, the most useful measure is the standard deviation, which is of the same unit as the data being described. “Standard” means that we have an idea as to which values have one/two/three STANDARD units away from the mean. Dr. Vidal MEASURES OF LOCATION • QUARTILE o divide the observations into 4 equal parts (if observation values are 100, then every 25th value is the quartile) • DECILE o divide the observations into 10 equal parts (if the observation Answer: D values are 100, then every 10th value is the decile) • PERCENTILE o (1st percentile to 100th percentile) TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN Page 21 of 79 For inquiries visit www.topnotchboardprep.com.ph or email us at [email protected] ✔ GUIDE QUESTION Which numerical summary measure would allow you to discriminate between the two distributions? A. Median B. Mean C. Mode D. Standard deviation This handout is only valid for the April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or https://www.facebook.com/topnotchmedicalboardprep/ This handout is only valid for April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. Ito actually yung grade mo sa pa exam natin dito sa Topnotch, so ano nga ulit percentile mo? So that’s your “location”/ sa system, NMAT grade also is good example of this measures of location. Dr. Mann NORMAL DISTRIBUTION • The normal distribution is the most important probability distribution in statistics because it fits many natural phenomena. For example, heights, blood pressure, measurement error, and IQ scores follow the normal distribution • The normal distribution is a probability function that describes how the values of a variable are distributed CHARACTERISTICS: 1. Bell shaped and symmetrical about the mean 2. The mean, median, mode are all equal 3. The total area under the curve and above the x axis is equal to 1 4. It has long tapering tails extending infinitely but never touching the x axis 5. It is determined by its parameters: its mean(µ) and standard deviation(σ) 6. The standard deviation becomes a more meaningful quality than merely being a measure of dispersion SUPPLEMENT: CENTRAL LIMIT THEOREM The central limit theorem in statistics states that, given a sufficiently large sample size, the sampling distribution of the mean for a variable will approximate a normal distribution regardless of that variable’s distribution in the population. Dr. Mann INFERENTIAL STATISTICS INFERENTIAL STATISTICS • A statistical interference makes generalizations and conclusions about a target population from samples. • Summarizing figures: o Parameter – a numerical constant obtained by observing the total population (usually unknown) o Statistic – a numerical variable obtained by observing a random sample from the population. • Sampling variation – brought about by the element of chance which is inherent in random sampling • It is comprised of: o Hypothesis Testing o Estimation § Point Estimate- is a single number § Interval Estimate https://statisticsbyjim.com/basics/central-limit-theorem/ APPLICATION NORMAL DISTRIBUTION 1. Computation of proportion or percentages of values that belong to different categories of variable of interest 2. Determining the x value that bound a specified area under the normal curve. “THE 68-95-99.7% RULE”/EMPIRICAL RULE • 68% of observations fall within 1 SD of the mean • 95% of observations fall within 2 SDs of the mean • 99.7% of observations fall within 3 SDs of the mean SCHEMATIC DIAGRAM OF THE CONCEPTS OF STATISTICAL INFERENCE HYPOTHESIS TESTING • simply defined as a statement about the population, based on the probability of occurrence of the sample results if the null hypothesis were true. STEPS IN HYPOTHESIS TESTING http://statisticshelper.com/empirical-rule-calculator-mean-standard-deviation Thus, for a normal distribution, almost all values lie within 3 standard deviations of the mean SKEWED DISTRIBUTION • A distribution is said to be skewed when the data points cluster more toward one side of the scale than the other, creating a curve that is not symmetrical. In other words, the right and the left side of the distribution are shaped differently from each other. There are two types of skewed distributions. Skewed to the LEFT AKA-Negatively skewed Outlying values are small Mean is smaller than the median Mean<median<mode Skewed to the RIGHT AKA- Positively skewed Outlying values are large Mean is larger than the median Mean>median>mode © Topnotch Medical Board Prep Remember in a graph the left most part has the low value and moving to the right side of the graph the value is increasing. Follow the red arrow, tail of the arrow LOW VALUE and the arrowhead HIGH VALUE 1. State the null hypothesis and alternate hypothesis 2. State the level of significance 3. Choose the test statistic 4. Compute for the test statistic 5. Make a statistical decision o A. Rejection region approach o B. P-value approach 6. Draw conclusions about the population 1. STATE THE HYPOTHESIS TYPES OF HYPOTHESIS • Null Hypothesis (Ho) o Hypothesis of NO difference o Statement of equality o It is framed in hopes of being able to reject it so that the alternative hypothesis could be accepted. o “There is no association between the disease and the risk factor in the population” • Alternative Hypothesis (H1) o The hypothesis that the investigator believes in o Example: There is some association between the disease and the risk factor in the population Example: • Suppose we want to test whether the mean (average) attention span of topnotch online students is 10 minutes • Ho – The population mean attention span of topnotch online students is equal to 10 minutes. • H1 : o Two Tailed: The population mean attention span of topnotch online students is not equal to 10 minutes. o One-Tailed: The population mean attention span of topnotch online students is greater/less than to 10 minutes. TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or email us at [email protected] This handout is only valid for the April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. Page 22 of 79 TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or https://www.facebook.com/topnotchmedicalboardprep/ This handout is only valid for April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. 2. STATE THE LEVEL OF SIGNIFICANCE LEVEL OF SIGNIFICANCE (α) • The significance level, also denoted as alpha or, is a measure of the strength of the evidence that must be present in your sample before you will reject the null hypothesis and conclude that the effect is statistically significant. • The researcher determines the significance level before conducting the experiment • gives the probability of incorrectly rejecting the null hypothesis when it is actually true. o Traditional values for α: 0.05, 0.01, 0.001 α is typically set at ≤0.05 (but can be set where a study determines), which allows interpretation with 95% certainty that a detected association is true. Dr. Mann TYPE I (Α) ERROR Error of rejecting the null hypothesis when it is really true Declaring a difference when none exists. Similar to false positive test TYPE II (Β) ERROR Error of NOT rejecting the null hypothesis when it is actually false Failing to declare a difference that does exist. Similar to false negative test If we look at what can happen in a hypothesis test, we can construct the following contingency table, table sa baba: REALITY Dr. Mann REALITY DECISION H0 is TRUE Do not Reject H0 OK Reject H0 Type I Error α- Error H0 is FALSE Type II Error β -Error OK https://medium.com/@neeraj.kumar.iitg/statistical-performance-measures-12bad66694b7 MNEMONIC TYPE 1- ART-- ALPHA REJECTING TRUE hypothesis TYPE 2- BNF- BETA NOT REJECTING FALSE hypothesis Dr. Mann Type II errors are decreased by increasing the sample size, effect size, and the precision of the measurement. Dr. Vidal PURPOSE OF TEST Compares two independent samples Compares two sets of observations on a single sample Compares three or more sets of observations made on a single sample Compares three or more three or more means where (1) participants are measured multiple times to see changes to an intervention; or (2) participants are subjected to more than one condition/trial and the response to each of these conditions wants to be compared. SUPPLEMENT: STATISTICAL POWER • probability of a hypothesis test of finding an effect if there is an effect to be found • ↑power and ↓β by: o ↑sample size § Increasing the number of samples à increases the power of the study o ↑expected effect size § quantified magnitude of a result present in the population § Cohen’s d (between two means, expressed in SD) § Pearson’s r (standardized scale to measure correlations between variables) o ↑precision of measurement 3. CHOOSE THE TEST STATISTIC PARAMETRIC TEST NONPARAMETRIC TEST used when data follow a used when a normal normal distribution distribution cannot be assumed uses absolute data in uses ranked data in computations computations PAIRED TESTS used when data are paired Ex. Pre- and post-test scores UNPAIRED TESTS used when data are independent Ex. Score distribution TWO-TAILED TEST H1 states that there is a difference but does not specify its direction Ex: The proportion of topnotch reviewees who obtained a grade of >80 among those engaged in SGD is not equal to those who self-study should be used when an intervention could potentially lead to either an increase or decrease of the outcome. ONE-TAILED TEST H1 states that there is a difference and specifies its direction of difference Ex: The proportion of topnotch reviewees who obtained a grade >80 among those engaged in SGD is greater than those who selfstudy should be used when an intervention can have only one plausible effect on the outcome. ✔GUIDE QUESTIONS Identify what statistical test is ideal to use in the ff situations: Q: Is the mean systolic blood pressure (at baseline) for patients assigned to placebo different from the mean for patients assigned to the treatment group? Dataset observes a normal distribution. A: Two-sample t-test Q: Was there a significant change in systolic blood pressure between baseline and the six-month follow-up measurement in the treatment group?? Dataset observes a non-normal distribution. A: Wilcoxon matched pairs test Q: Is systolic blood pressure associated with the patient’s age? Sample size of the dataset is low (n = 10) A: Spearman rank correlation test For lower sample sizes, it is better to use a non-parametric test. STATISTICAL TESTS • Predictor and outcome variables • Difference between two group COMMONLY USED STATISTICAL TESTS PARAMETRIC NONPARAMETRIC EXAMPLE TEST TEST Two-sample To compare girls’ heights with boys’ (unpaired) Mann-Whitney U test heights t test One-sample (paired) Wilcoxon matched To compare weight of infants before and t test pairs test after a feeding One-way analysis of Kruskal-Wallis To determine whether plasma glucose variance (F test) using analysis of variance level is higher 1 hour, 2 hours, or 3 hours total sum of squares by ranks after a meal (ANOVA) One-way ANOVA with repeated measures Friedman test The same group of people were made to exercise on three different conditions: with classical music, with dance music, or without music. Each time, they were asked to rate how much effort they had exhibited in the different situations. TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or email us at [email protected] This handout is only valid for the April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. Page 23 of 79 TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or https://www.facebook.com/topnotchmedicalboardprep/ This handout is only valid for April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. PARAMETRIC TEST NONPARAMETRIC TEST Test the influence (and interaction) of two different variables Two-way analysis of variance (ANOVA) Two-way analysis of variance by ranks In the above example, to determine whether the results differ in male and female subjects Tests the null hypothesis that the distribution of a variable is the same in two (or more) independent samples χ2 (chi square) test Fisher exact test To assess whether male or female adolescents are more likely to smoke Assesses the strength of the straightline association between two continuous variables Product moment correlation coefficient (Pearson r) Spearman rank correlation coefficient (rσ) To assess whether and to what extent plasma HbA1c concentration is related to plasma triglyceride concentration in diabetic patients Describes the numerical relation between two quantitative variables, allowing one value to be predicted from the other Regression by least squares method Nonparametric regression (various tests) To see how peak expiratory flow rate varies with height Multiple regression by least squares method Nonparametric regression (various tests) To determine whether and to what extent a person’s age, body fat, and sodium intake determine his or her blood pressure PURPOSE OF TEST Describes the numerical relationship between a dependent variable and several predictor variables (covariates) EXAMPLE Data from Greenhalgh T. How to read a paper. Statistics for the non-statistician. I: Different types of data need different statistical tests. BMJ. 1997; 315(7104): 364-366. Table 29.1. Molloy M. (2020) Biostatistics and Evidence-based Medicine. K Kleinman et al. (Ed). The Harriet Lane Handbook (22nd ed. 29: p 655). Elsevier Inc. ✔GUIDE QUESTIONS Identify what statistical test is ideal to use in the ff situations: Q: Is there a difference between peak electrohysterogram density before and after administration of nifedipine in preterm patients? A: Q: Is there a difference between peak electrohysterogram density in preterm patients who deliver within 1 week of nifedipine treatment vs patients who deliver in 1-2 week of nifedipine treatment vs patients who deliver after 2 weeks of nifedipine treatment? A: Q: Is there an association between postpartum weight and self-reported gestational weight? A: Is there an association between postpartum weight and self-reported gestational weight, age, ethnicity, etc.? A: Is there an association between postpartum obesity (obese, non-obese) and self-reported gestational weight, age, ethnicity, etc.? A: SAMPLE QUESTION Q: Which statistical test to use given this graph? • • • • CORRELATION Relationship Variables move together X and y can be interchanged Data represented in single point REGRESSION • One affects the other • Cause and effect • X and y cannot be interchanged • Data represented by line In regression, the most common model involves the least-squares method. In this model, a regression line should fit the line to the data points in a way that minimizes the sum of the squared vertical distances between the line and the points. The distance between the expected value from a regression equation and the actual observed value is called a RESIDUAL. Dr. Vidal 4. COMPUTE FOR THE TEST STATISTIC Interpretation is much more important rather than computation sa boards, and usually ang computation ngayon ginagamitan na ng software like STATA. SO KALMA LANG TAYO SA PART NA ITO! Dr. Mann 5. MAKE A STATISTICAL DECISION A. REJECTION REGION APPROACH Critical Region • The critical region is the region of values that corresponds to the rejection of the null hypothesis at some chosen probability level • also known as the rejection region • is a set of values for the test statistic for which the null hypothesis is rejected. i.e. if the observed test statistic is in the critical region then we reject the null hypothesis and accept the alternative hypothesis. Rejecting or Accepting a Hypothesis A: Q: Which statistical test to use given this graph? A: CORRELATION AND REGRESSION • Correlation and regression describe the degree of linear association between two quantitative variables, but they do not imply causation. • Correlation measures the strength of association between two variables; expressed by the correlation coefficient r, also termed Pearson correlation coefficient. • Regression constructs an optimal straight line illustrating correlation and allows for prediction of a dependent variable based on an independent (known) variable. B. P-VALUE APPROACH • It is the probability of obtaining the result as extreme or more extreme than the one observed if the null hypothesis is true. The probability that the observed result is due to chance alone. • is the probability of a difference occurring by chance, and is judged against α, the preset level of significance. TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or email us at [email protected] This handout is only valid for the April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. Page 24 of 79 TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN For inquiries visit www.topnotchboardprep.com.ph or https://www.facebook.com/topnotchmedicalboardprep/ This handout is only valid for April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly. • If p is less than the significance level α, the detected association is unlikely to be due to chance alone. For example, if p <0.01, there is less than a 1 in 100 chance of the detected association being due to chance alone. Rejecting or Accepting a Hypothesis Low P High P p<α p>α Value of sample results are Value of sample results are far from the population close to population parameters parameters Unlikely events Likely events REJECT HO DO NOT REJECT HO Note: If there is no sufficient evidence to reject the null hypothesis, it is RETAINED or cannot be rejected but NOT ACCEPTED. The Rejection Region Approach and the P-value Approach are different means to INTERPRET a result from the hypothesis testing. Yung Rejection Region, magseset ka muna ng critical region for rejection, tapos kapag yung result mo pumasok dun sa critical region, magrereject ka ng null. Yung P-value approach naman, kapag nacompute ng software na p value ay less than alpha, then magrereject ka ng null. EVIDENCE-BASED MEDICINE • Evidence-based medicine refers to the method of integrating individual clinical expertise with the best available evidence from the literature. • Evidence based medicine (EBM) is the conscientious, explicit, judicious and reasonable use of modern, best evidence in making decisions about the care of individual patients. EBM integrates clinical experience and patient values with the best available research information THE THREE SKILLS OF EBM • Skill number 1: Acquiring the evidence • Skill number 2: Appraising the evidence • Skill number 3: Applying the evidence Essentially, any value inside the critical region will have a significance value less than the alpha. Same banana. Ang bawat daan ko ay patungo, ay pabalik sa'yo (Dancel, 2019) Dr. Vidal 6. DRAW CONCLUSIONS ABOUT THE POPULATION • Ho - The population mean attention span of topnotch online students is equal to 10 minutes. • H1 : o Two Tailed: The population mean attention span of topnotch online students is not equal to 10 minutes. The Claim is the Null Hypothesis Alternative hypothesis The Decision is Fail to Reject the Null Reject the Null Fail to Reject the Null Reject the Null The Conclusion is The evidence is not sufficient to reject the claim The evidence is sufficient to reject the claim There is insufficient evidence to support the claim There is sufficient evidence to support the claim ESTIMATION Painless Evidence-Based Medicine John Wiley & Sons, 2008 PICO/M PROCESS • P: Describe the patient or problem, deciding whether the evidence you seek is regarding therapy, diagnosis, prognosis, etiology, or cost effectiveness. • I: Describe the intervention under consideration. The intervention is the treatment that you would want to give or explore that is currently not the standard of care. • C: Compare the intervention with an alternative or current standard of care. • O: Formulate a specific outcome of interest. • M: Methodologies ✔GUIDE QUESTION Identify the Patient, Intervention, Comparison, Outcome, and Methods for this research question. Among CKD patients with concurrent septic shock, will reducing the fluid resuscitation reduce hospitalization stay? EBM LEVEL OF EVIDENCE • The process of computing for measures of population attributes based on data from a sample. © Topnotch Medical Board Prep 2 TYPES POINT ESTIMATE/ POWER ESTIMATE INTERVAL ESTIMATE/ CONFIDENCE INTERVAL • A single numerical value used to estimate the corresponding population parameter • Two numerical values defining an interval which with ranging degrees of confidence is expected to include or catch the parameter being tested. • Interval estimate – consists of two numbers, a lower limit and an upper limit, which serve as the bounding values within the parameter is expected to lie with a certain degree of confidence https://academicguides.waldenu.edu/library/healthevidence/evidencepyramid The levels of evidence pyramid provide a way to visualize both the quality of evidence and the amount of evidence available. For example, systematic reviews are at the top of the pyramid, meaning they are both the highest level of evidence and the least common. As you go down the pyramid, the amount of evidence will increase as the quality of the evidence decreases. Meron din variation with the different level of evidence if given sa choices ang meta-analysis choose that as the highest level of medicine. Since these studies are also epidemiological in nature, these will be further described in the next section, Epidemiology. CONFIDENCE INTERVAL • a range of values so defined that there is a specified probability (90%, 95%, 99%)that the value of a parameter lies within it Dr. Mann BIASES IN RESEARCH BIAS IN RECRUITING PARTICIPANTS/SAMPLE POPULATION • SELECTION BIAS - Comparisons are made between groups of patients that differ in determinants of outcome other than the You can calculate confidence intervals for many kinds of statistical one under the study estimates, including • Proportions 1. Medical Surveillance bias – refers to over-detection of the • Population means disease of interest because one of the groups goes to the doctor • Differences between population means or proportions (or has a diagnostic test) more often than does another group. • Estimates of variation among groups 2. Centripetal bias – Occurs when a major clinical center’s These are all point estimates, and don’t give any information about the reputation results in part from its particular expertise in a variation around the number. Confidence intervals are useful for specialized area of clinical medicine, it will be referred communicating the variation around a point estimate. problem cases likely to benefit from its expertise Dr. Vidal TOPNOTCH MEDICAL BOARD PREP PREVENTIVE MEDICINE AND PUBLIC HEALTH MAIN HANDOUT BY DR. MANN Page 25 of 79 For inquiries visit www.topnotchboardprep.com.ph or email us at [email protected] This handout is only valid for the April 2024 PLE batch. This will be rendered obsolete for the next batch since we update our handouts regularly.