Question 1 – Identify the variable names in the above table as continuous, ordinal, or discrete.
Question 2 – Fill in the following table, using the appropriate summary statistics.
Discuss the possible relationship between dyslipidaemia, HbA1c or diastolic BP, patients’ age, and insulin status.
Question 3 – Graphically show the relationship between dyslipidemia (disease) and each of three variables: duration, BMI group, and occupation type.
Each of the variables should be presented on a graph. Discuss.
Question 4 – The questions that follow will demonstrate the detailed calculations.
(a.) A typical battery’s lifetime is distributed with a median life of 40 hours and an average deviation of 1.2.
Find out the probability that a random battery will last for more than 42 hour.
(b). In hospitals, the average age distribution of new employees over the past 5 years is:
95% are between 24.6-37.4 years.
Find the standard deviation and the mean age.
(c). Given that scores on a medical examination follow a normal distribution with an average score of 460, and a standard deviation 100.
Is it possible that all 41 students will take the medical test?
How will this affect the distribution of the samples mean of systolic pressure in large numbers?
Question 6 – Briefly describe the differences between
Randomized clinical trials (RCT) or prospective cohort studies
Case-control study and historical cohort study.
Prospective cohort studies and historical cohort studies
Answer to Question: MPH5041 Introductory Biostatistics
Age (in Years)DiscreteNA
1: 40 year, 2 : 40 – 49 year, 3 : 50 – 60 years, 4 : >= 60yearsGender
1: Male; 2. Female
Type of occupation
1: Sedentary Worker; 2: Moderate Worker; 3: Heavy Physical Worker
Physical Activity Level
0: Sufficient Paid;
1 Insufficient PA
Weight (in kilograms)ContinuousNA
Height in metersContinuousNA
BMI (in Kg/m2)ContinuousNABMI Group
Nominal0: Normal; 1: Overweight; 2: Obese
Diastolic Blood PressureDiscreteNASystolic BPDiscreteNAHypertension
Nominal0: Normotensive; 1: Hypertensive
Duration of DiabetesDiscreteNA
Diabetes Group – Duration
1: 5 years; 2; 5 -10 years; and 3: >=10 YearsHbA1cContinuousNA
0: 7 year; 1 = >= 7 year
0 = No; 1 = Yes
Insulin life expectancy (years).DiscreteNA
Total CholesterolDiscreteNATotal Cholesterol Group
Nominal0: <200; 1: >= 200TriglycerideDiscreteNATriglyceride Group
Nominal0: <150; 1: >= 150HDL CholesterolDiscreteNAHDL Cholesterol Group
0 is Normal; 1 is LowLDL CholesterolDiscreteNALDL Cholesterol Group
Nominal0: <100; 1: >= 100Serum CreatinineContinuousNASerum Creatinine Group
Nominal0: < 100; 1: >= 100
Nominal0: No Dyslipidemia; 1: Yes DyslipidemiaAnswer:2
Dyslipidemia (% or median and IQR, or mean &SD)
40 – 49 Years30.334.37
>= 60 Jahre7.100.82Answer:3
Clustered bar graphs are the only way to see the relationship between categorical variabilites.
This chart best shows the relationship between categorical varietals.
Figure 3.1 clearly shows how dyslipdaemia affects normal people more than it does obese people.
Figure 3.1. Relationship between BMI-group and dyslipdaemia
Figure 3.2 clearly shows that dyslipdaemia goes up with age.
Figure 3.3 illustrates the relationship among dyslipdaemia.
It can be seen that dyslipdaemia is most common in sedentary workers, and almost non-existent in heavy physically employed workers.
Figure 3.2. Relationship of diabetes group and dyslipdaemia
Figure 3.3. Relationship between occupation type & dyslipdaemiaAnswer:4
Let X represent the life expectancy for a battery.
Normally, the lifetime of a lithium-ion battery is evenly distributed.
The 40 hour average life and the 1.2 hour standard deviation are both normal.
P (X > 42 =) = The probability that a random selection bulb will last for more than 42hrs.
Let X be the age of all hospital employees who were hired within the last 5 years.
These ages should be distributed normal.
The (95% confidence interval) of the ages is given by (24.6, 37.5).
Normal distribution of data means that the difference between the confidence interval and the standard deviation can be divided into four equal parts.
This gives the standard deviation as (37.3 – 24.6)/(4) = 3.2.
It is possible to find the average by comparing two standard deviations that are located towards the left-hand side of the distribution.
This is how you get the mean: ((37.4) – (2 * 3.2) = 31
Let X represent the scores from a medical examination.
Normally, scores from a medical exam are distributed.
The standard deviation of 100 is the average score and 460 is the mean score.
P (X 589.93), is the probability that a class will receive a score above 589.93.Answer:5
Positively, there is an imbalance in the distribution of systolic pressure within the population.
The distribution of the sampling distribution will be also normal.
This can also be derived from the central limitation theorem.
The theory states that when random samples of a population are drawn, then the sample mean of each sample will equal the population standard deviation (s) and the population median u.
The sample distribution will be the same as that of the population.
If the sample becomes large, no matter how large the population, the distribution of the sample will be normal.
In this instance, it is obvious that the population distribution is normal.
It can be concluded that the population distribution for blood pressure is normal, as it includes a large number of random samples.Answer:6
1. A randomized clinical study is one in which the participants are randomly assigned to various groups that compare different treatments.
Researchers or participants cannot decide which group the participant will be in.
It will always happen randomly.
This will make the various groups more similar. The treatments received by each participant can then be compared objectively.
The best treatment cannot be determined until the trial is started.
The patient can decide whether to participate in the random trials.
This is a randomised clinical study.
Prospective cohort studies refer to a study that compares the outcomes of several individuals who share many of the same characteristics with some of those who differ in one (eg. smokers and nonsmokers) over a period of time.
2.Case Control studies are a comparison between two types patients.
One group has the disease while the other is not.
It also determines the relationship between the diseases or risks.
A retrospective cohort study is, however, a type research study in which records of medical history are compared between participants to see if there is any difference in the outcome.
In this instance, data are always available and the results are not dependent on experiments.
3.The participants of the interest are identical in all cohort studies, retrospective or prospective.
They share many of the same characteristics but have one distinct characteristic.
The difference between the types of prospective cohort studies is that they are done on the basis selected samples and monitored over a time period.
Therefore, data collection can take time.
However, retrospective studies are based upon medical history data.
In prospective study, the samples are selected first, then the data collected by monitoring them. But in retrospective study the data are obtained at first, then the data used to select the sample.