Number needed to treat can be defined as the number of patients who need to be treated to prevent one additional bad outcome.

It can be found as:

NNT=1/Absolute Risk Reduction (rounded to the next integer since number of patients can be integer only).

where ARR= (Risk factor associated with the new drug group) — (Risk factor associated with the currently available drug)

So,

ARR= (165/1050)-(132/1044)

ARR= (0.157-0.126)

ARR= 0.031

NNT= 1/0.031

NNT=32.3

Recall bias introduced when participants in a study are systematically more or less likely to recall and relate information on exposure depending on their outcome status.

In procedure bias, the researcher decides assignment of a treatment versus control and assigns particular patients to one group or the other non-randomly. This is unlikely to have occurred in this case, although it is not mentioned specifically.

Self Selection or volunteer bias occur when those subjects are selected to participate in the study who are not the representative of the entire target population. those subjects may be from high socio-economic status and practice those activities or lifestyle that improves their health.

Lead-time bias occurs when a disease is detected by a screening test at an earlier time point rather than it would have been diagnosed by its clinical appearance. In this bias, earlier detection improves the survival time in the intervention group.

The Delphi is a group facilitation technique that seeks to obtain consensus on the opinions of `experts’ through a series of structured questionnaires (commonly referred to as rounds). By using successive questionnaires, opinions are considered in a non-adversarial manner, with the current status of the groups’ collective opinion being repeatedly fed back. Studies employing the Delphi make use of individuals who have knowledge of the topic being investigated

Variance and standard deviation indicate the dispersion of the plot from mean value and thus are not really helpful in summarizing the mean.

Range is the difference between maximum and minimum value that is 45 in this case.

The mean in this case is 6.2 due to the presence of an outlier 45. In the presence of outlier mean can be misleading as it is quite sensitive to skewness in data.

Mode is the most frequent value. In this case mode has 4 values: 0,1,2,4.

In case of skewedness, median is the most apt representative of the mean as it is not affected by outliers. In this case since the data set has even values i.e. 10. Median is the average of the 5th & 6th entry after arranging the data in ascending order like that in case of the question (0,0,1,1,2,2,3,4,4,45). This turns out to be 2.

The correlation coefficient gives us the idea about relation between two parameters. i.e. to what extent change in parameter A could prompt a change in parameter B. The numerical value of correlation coefficient could not be greater than 1.

Hawthorne effect explains the change in any behavioural aspect owing to the awareness that the person is being observed.
Simpson’s Paradox explains the association developed when the data from several groups is combined to form a single larger group.

The remaining terms are made up.

The Number Needed to Treat (NNT) is the number of patients you need to treat to prevent one additional bad outcome. For example, if a drug has an NNT of 5, it means you have to treat 5 people with the drug to prevent one additional bad outcome.

To calculate the NNT, you need to know the Absolute Risk Reduction (ARR); the NNT is the inverse of the ARR:

NNT = 1/ARR

Where ARR = CER (Control Event Rate) – EER (Experimental Event Rate).

NNTs are always rounded up to the nearest whole number.

In this case, the NNT can be computed as follows:

ARR = 10% – 5% = 0.05

NNT = 1/0.05 = 20

The Kappa Statistic or Cohen’s Kappa is a statistical measure of inter-rater reliability for categorical variables. It is used when two raters both apply a criterion based on a tool to assess whether or not some condition occur. A good example can be two doctors rating individuals for diabetes occurrence on the basis of symptoms.

It gives a quantitative measure of the magnitude of agreement between observers.

Kappa can take any value between 0 and 1. 0 implies the observers are in complete disagreement and a value of 1 implies complete agreement.

Key Point: While finding out confidence intervals, standard errors are used. Standard error and Standard deviation are two distinct entities and should not be confused.

For 99.7% confidence interval, you can find the range as follows:

Multiply the standard error by 3.

Subtract the answer from mean value to get the lower limit.

Add the answer obtained in step 1 from the mean value to get the upper limit.

The range turns out to be 1-13 mmol/L.

For a confidence interval of 68%, multiply the standard error with 1 and repeat the process. The range found for this interval is 3-11 mmol/L.

For a 95% confidence interval. Standard Error is multiplied by 1.96 which gives us the limit ranging from 3.08 to 10.92 mmol/L which could be approximated to 3-11 mmol/L.

The Delphi method relies on expert consensus. This method kicks off with an open ended questionnaire and uses its responses as a survey instrument for the next round in which each of the participants is asked to rate the items that the investigators have summarized on the basis of the data collected in the first round. Any disagreement is further discussed in phases to come on the basis of information obtained from previous phases.

Observers may subconsciously measure or report data in a way that favours the expected study outcome. Therefore, by blinding the study we can eliminate expectation bias.

Recall bias is a systematic error that occurs when the study participants omit details or do not remember previous events or experiences accurately.

Verification can occur during investigations when there is a difference in testing strategy between groups of individuals, which might lead to biasness due to differing ways of verifying the disease of interest.

Nonresponse bias is the bias that occurs when the people who respond to a survey differ significantly from the people who do not respond to the survey.

A distortion that modifies an association between an exposure and an outcome because a factor is independently associated with the exposure and the outcome. Randomization is the best way to reduce the risk of confounding.

Phase 0 trials assist the scientists in studying the behaviour of drugs in humans by micro dosing patients. They are used to speed up the developmental process. They have no measurable therapeutic effect and efficiency.

Phase 1 is associated with assessing whether a drug is safe to use or not. The process is extensive and can take up to several months. It also involves healthy participants (less than 100) that are paid to take part in the study. The side effects upon increasing dosage are also addressed by the study. The effects the drug has on humans including how its absorbed, metabolized and excreted are studied. Approximately 70% of the drugs pass this phase.

Phase 2 trials involve patients that are suffering from the disease under study and are associated with determining the efficiency and the optimum dosage of the drug.

Phase 3 also assesses the efficacy but at a higher scale with larger population sample.

Phase 4 trials are involved with the long term effects and side effects of the drug.

Randomized controlled trails will be analysed by the intention-to-treat (ITT) approach. It provides unbiased comparisons among the treatment groups. ITT analyses are done to avoid the effects of dropout, which may break the random assignment to the treatment groups in a study.

ITT analysis is a comparison of the treatment groups that includes all patients as originally allocated after randomization.

In order to include such participants in an analysis, outcome data could be imputed which involves making assumptions about the outcomes in the lost participants.

The following terms are used in medical testing:

True negative – The test is negative and the patient does not have the disease.
True positive – The test is positive and the patient has the disease.
False positive – The test is positive but the patient does not have the disease.
False negative – The test is negative but the patient has the disease.

The sensitivity of a predictive test = true positives / (true positives + false negatives).

The specificity of a test = true negatives / (false positives + true negatives).

The negative predictive value of a test = true negatives / (false negatives + true negatives).

In general practice, majority of phase 3 trials and some of the trials conducted in phase 2 are randomized. Because phase 4 trials require a huge sample size, they are not randomized as much. The primal reason behind conducting phase 3 trials is to test the efficiency and safety in a significant sample population. At this point it is assumed that the drug is effective up to a certain extent.

During a case-control study, subjects that exhibit outcomes of interest are compared with those who don’t show the expected outcome. The extent of exposure to a particular risk factor is then matched between cases and controls. If the exposure among cases surpasses controls, it becomes a risk factor for the outcome that is being studied.

Pilot studies are conducted on a lower and much smaller level, to assess if a randomized controlled trial of the crucial components of a study will be plausible.

Phase 4 trials are the ones that are conducted after its established that the drug is effective and is approved by the regulating authority for use. These trials are concerned with the side effects and potential risks associated with the long term usage of the drug.

99.7% of the values within 3 standard deviations of the mean.

For 99.7% confidence interval, you can find the range as follows:

1. Multiply the standard error by 3.

2. Subtract the answer from mean value to get the lower limit.

3. Add the answer obtained in step 1 from the mean value to get the upper limit.

For a confidence interval of 68%, multiply the standard error with 1 and repeat the process. For a 95% confidence interval, Standard Error is multiplied by 1.96 to get the interval.

Funnel plot is essentially a scatterplot of the effect of treatment against a particular measure of study precision. Its primal purpose is to serve as a visual aid and help in detection of bias or systematic heterogenity.

Standard error can be calculated by the following formula:
Standard Error= (Standard Deviation)/√(Sample Size)
= (16) / √(100)
= 16 / 10
= 1.6

Number needed to treat can be defined as the number of patients who need to be treated to prevent one additional bad outcome.

It can be found as:

NNT=1/Absolute Risk Reduction (rounded to the next integer since number of patients can be integer only).

Number needed to treat is a measure of the impact of a treatment or intervention that is often used to communicate results to patients, clinicians, the public and policymakers. It states how many patients need to be treated for one additional patient to experience an adverse outcome (e.g. a death).

It is calculated as the inverse of the absolute risk reduction and is rounded to the next highest whole number.

The absolute risk reduction is 8% (16% – 8%). 100/8 = 12.5, so rounding up the next integer this gives at NNT of 13. i.e. you would need to give the new drug to 13 people to ensure that you prevented one death.

Selection bias is when randomisation is not achieved and is often a result of in efficient recruiting method.

Recall bias is specifically appropriate to the case control studies that is when ever the memories retrieved by the participants differ in accuracy.

Measurement bias can be characterized by gathering of information in a manner that is distorted.

When the participants of a research study are recruited from the hospitals rather than the general population, its called Berkson Bias.

Confounding bias is the major player here because in this case the effects of smoking can be masked behind and can be read as outcomes of lower socio economic status. This extraneous factor (Smoking), distorts the founding.

Attributable proportion is the proportion of disease that is caused due to exposure. It refers to the proportion of disease that would be eradicated from a particular population if the disease rate was diminished to match that of the unexposed group.

Risk ratio (relative risk) compares the probability of an event in an exposed (experimental) group to that of an event in the unexposed (control) group. Thus two are not the same.

The attributable risk is the rate of a disease in an exposed group to that of a group that has not been exposed to it i.e. how many deaths did the exposure cause.

The null hypothesis states that there is no significant difference between two groups.

The alternative hypothesis states that there is a significant difference between two groups.

A type I error (false positive) occurs when a null hypothesis is rejected when it should be accepted.

A type II error (false negative) occurs when the alternative hypothesis is rejected when it should be accepted.

The probability determines the rejection of a null hypothesis.

The level of significance is set at p <0.05.

The study is being conducted on a smaller level with only 200 participants and is determining the effectiveness of the drug in comparison to a placebo. These characteristics are in accordance with the second phase of trial.

Randomisation allows for performance of experimental trials in a random order. Using this method gives us control over the confounding variables that are not supposed to be held constant.

For an instance, by employing randomisation we get to control biological differences among individual human beings during experimental trials.

The question mentions a quantitative, ratio scale data set. The use of mean would be ideal under normal circumstances, however, in this situation median is preferred as it is less sensitive to the skewness of data. The median is usually preferred to other measures of central tendency when your data set is skewed (i.e., forms a skewed distribution)

Mean, also known as the average, is the most common measure of central tendency. It is the sum of all observed values divided by the number of observation. It is not useful for skewed data, which has an abnormal distribution. It is useful, instead, for numerical data that have symmetric distribution. It reflects the contributions of each data in the group, and are sensitive to outliers.

The median is the value that falls in the middle position when the observations are ranked in order from the smallest to the largest. If the number of observations is odd, the median is the middle number. If it is even, the median is the average of the two middle numbers. Unlike the mean, the median is useful on skewed data, and can be used for ordinal or numerical data if skewed.

The mode is the value that occurs with the greatest frequency in a set of observations, and is utilized for bimodal distribution.

The variance and the standard deviation are not measures of central tendency, but of dispersion.

When the association among any two classes of object is defined by the presence of a third entity it is termed as indirect association. For an instance, the age of the employee may affect the rate of pay, which would have implications on job satisfaction. So, in this example, an indirect relationship between age and job satisfaction exists due to a third party i.e. rate of pay.

Chi-square test is used to determine the statistically significant different between categorical variables. It also determines the difference between expected frequencies and the observed frequencies.

Mann Whitney U test is used to determine the statistically significant different between two independent groups.

Wilcoxon’s test is the test of dependency. it determines the statistically significant difference between two dependent groups.

Student t-test is one of the most commonly used method to test the hypothesis. It determines the significant difference between the means of two different groups.

ANOVA (analysis of variance) is similar to student’s t-test.

ANOVA is a statistical method used to determines the statistically significant difference between the mean of more than two group. In this experiment as we are dealing with three different group, ANOVA is most suitable test to determine the difference between each groups.

The incidence of a disease is the number of new cases of the disease in a population over a defined time period.

The prevalence of a disease is the number of cases of the disease in a population over a defined time period describes. It is NOT the number of new cases.

The number of new cases of a disease only, has no denominator (time period or population) from which to derive an incidence.

The number of new cases of a disease seeking medical treatment is the incidence of patients seeking medical treatment NOT the incidence of the disease in a population.

The death rate from a disease is the number of patients dying from the disease in a population.