Exploring Types of Study Design in Research

Study design is a crucial aspect of research that determines the validity and reliability of the findings. This presentation provides an in-depth overview of various types of study designs, their applications, advantages, and limitations. Understanding these designs helps researchers choose the most appropriate method for their studies.

Importance of Study Design Study design is essential for:

1. Ensuring the accuracy and reliability of research findings

2. Minimizing biases and errors

3. Facilitating the replication of studies

4. Enhancing the credibility of research

1. Types of Study Design Under the supervision of :- Dr. Naveen Kumar Vishwakarma (Department of Biotechnology) GGV, Bilaspur (C.G.) Presented by :- Ap arna Baya Sangita Dhruv Shweta Panday Sakshi Bhardwaj Vidhi Date Presented :- November 22 nd , 2024

2. Study Design Study design refers to the plan and structure of a research study, It involves defining the research question, population, sample size, measures, and methods used to collect and analyse data. It is a critical component of biostatistics, as it determines the validity and reliability of the study’s findings .

3. Characteristics:- Experimental The essential characteristics of a good study design are: Clear research question : The study should have a specific, well-defined research question or hypothesis. Well-defined population : The study should clearly define the population being studied. Appropriate sample size : The study should have a sufficient sample size to ensure reliable results. Valid and reliable measures : The study should use valid and reliable measures to collect data. Minimization of bias : The study should minimize bias by using objective measures and controlling for confounding variables. These characteristics are crucial for ensuring the quality and reliability of a study’s findings.

4. Types of study design:- TYPES OF STUDY DESIGN Completely randomised design Randomised block design Factorial design LSD Experimental Observational Analytical Descriptive Cross sectional Case control Cohort Longitudinal Case report Case series

5. Observational An observational study design is a type of research study where the researcher observes and records the behaviour, characteristics, or outcomes of a group of individuals or a population without intervening or manipulating the variables being studied. Observational studies are used to identify the relationship between variables and to understand the causes of a particular outcome.

6. Descriptive A descriptive study is a type of research design that aims to describe the characteristics of a population, such as demographics, behaviours, or outcomes. It provides a snapshot of the current situation and does not attempt to explain the reasons behind the observed phenomena.

7. Experimental Case report A case report is a type of study design that involves the detailed description of a single case or a small number of cases of a particular disease, condition, or phenomenon.
It is a descriptive study that aims to provide a detailed account of a unique or unusual case, often with a focus on the clinical presentation, diagnosis, treatment, and outcome.

8. One example of a case report is a study published in a medical journal that describes a patient with a rare genetic disorder. The study includes a detailed description of the patient’s symptoms, medical history, and treatment outcomes, as well as any relevant laboratory tests or imaging studies. A detailed description of the patient’s symptoms and medical history
A description of the diagnostic tests and procedures used to diagnose the condition
A description of the treatment provided to the patient
A description of the patient’s response to treatment
Any relevant laboratory tests or imaging studies

9. Experimental Case series Describes characteristics of a group of people who have same disease or the same exposure. the aim of this is to understand the demographics, clinical presentation, prognosis or other characteristics of people who have a particular disease or describe something unusual for example in the early 1980s the occurrence of an unusual pneumonia in men led to recognition and identification of HIV.

10. Experimental A case series can be thought of as similar case reports being lumped together. Instead of making four separate case reports describing a certain cardiovascular condition you can lump them into one case series if there’s enough similarities between the participants.

11. Experimental Analytical Analytical research is about critically analyzing the facts ,data & information already available. Analytical research focuses on “HOW &WHY?” How it is happening ? Or why it is happened? An analytical study design is a type of study design that aims to analyse the relationship between variables, often using existing data. The study aims to analyse the relationships between variables

12. Experimental Cross sectional A cross-sectional study takes a selected population and measures health information at a given point of time giving a snapshot of their health it usually involves asking participants a series of questions using a questionnaire Health surveys that collect health information about people in a population is an example of a cross sectional study because these studies commonly measure how many people have a disease at a particular point of time they’re also called prevalence studies
It’s important to make sure that the selected population is representative of the total Population.

13. Advantage Relatively inexpensive Easy to conduct compared other studies they can provide information about on multiple exposures and outcomes More useful for chronic diseases Show pattern of disease However because the information is collected at a single point in time it can’t be used to determine whether a particular exposure caused the disease or not. Disadvantage

14. Experimental Case-control Case control study starts off with cases , these are people with a disease it uses a comparison group called controls who are similar to cases but do not have the disease then both groups are asked about their previous exposures to different risk factors Both exposure & outcome have occurred before the start of study Study proceeds backward from effect to course Uses a control group to support an inference

15. Steps 1. Selection 1.1Selection of cases Definition of case Diagnostic criteria Eligibility criteria Sources of cases Hospitals General population 1.2 Selection of controls Controls must be as similar to the cases as possible, except for the absence of disease If study group is Small choose up to 4 control per case(1:4) & in Large studies 1:1 of cases & controls should be taken Sources of controls Hospital (often source of selection bias) Relatives Neighborhood General population

16. 2.Matching To ensure comparability between cases& controls Types Group matching- age , occupation, social class Matching in pairs 3.Measurement of exposure Obtained by interviews , by questionnaire , or by studying past records( hospital record, employment record) 4.Analysis 2×2Table

17. Odds ratio (Cross product ratio) OR= a × d ÷ b × c Indicator of increased risk of disease in predisposed population Its just on estimate relative risk not the calculation of relative risk similar to relative risk OR>1 Positive association So many time odds that cases were exposed to a risk factor is more to the odds that the controls were exposed Example- OCPS & Thromboembolic 2) OR<1 Negative association So many time odds that cases were exposed to risk factor is less than the odds that the control were exposed Example- Regular physical exercise &CHD

18. 3 . OR=1 No association Odds that cases were exposed to a risk factor is same as the odds that the controls were exposed Example- smoking & HIV/AIDS Let’s say we have a study that compares the odds of developing lung cancer in smokers versus non-smokers. The data shows that:
100 smokers developed lung cancer (cases)
500 smokers did not develop lung cancer (controls)
20 non-smokers developed lung cancer (cases)
480 non-smokers did not develop lung cancer (controls)
We can calculate the odds ratio as follows:
OR = (odds of lung cancer in smokers / odds of lung cancer in non-smokers)
= ((100/500) / (20/480))
= 4.5
This means that smokers are 4.5 times more likely to develop lung cancer than non-smokers. The odds ratio of 4.5 indicates a strong association between smoking and lung cancer. This suggests that smoking is a significant risk factor for developing lung cancer.

19. Advantage Disadvantage Quick and cheap Good for uncommon diseases Not rare exposures
control selection
recall may be a problem

20. Experimental Cohort A cohort study a group of people is followed over a period of time to see what happens to them and information about risk factors is collected we can then compare the occurrence of an outcome like disease in “those who were exposed to a particular risk factor” to “those who were not exposed to that risk factor” the main measurement used in cohort studies is called the relative risk .
A relative risk is the ratio between the risk of disease in the exposed group compared to the risk of disease in the unexposed group.
A relative risk of greater than one means that the exposure is associated with an increased risk of the disease
If it is one it indicates that the risk is the same and if it’s less than one indicates that the risk is lower.

21. A well-known cohort study is the British doctor study done in the 1950’s where a group of doctors were followed up for many years this study provided valuable scientific evidence of the harmful effects of smoking especially the link between smoking and lung cancer

22. Let’s consider an example: Suppose we are studying the relationship between smoking and lung cancer. We have a cohort of 1000 smokers and 1000 non-smokers, and we follow them for 10 years to observe the occurrence of lung cancer.

Incidence Rate (Smokers) = 50 cases of lung cancer / 1000 smokers / 10 years = 0.05 cases per person-year
Incidence Rate (Non-Smokers) = 10 cases of lung cancer / 1000 non-smokers / 10 years = 0.01 cases per person-year

Relative Risk = 0.05 / 0.01 = 5 This means that the relative risk of developing lung cancer is 5 times higher in smokers compared to non-smokers.

23. Cohort studies first divide the groups based on risk factors here we’re intersecting time at two different points Whenever you go back to a previous time as you’re conducting the study, it is called a retrospective study Whenever you follow up with participants at a later time as you’re conducting the study, it is called a prospective study. Retrospective is more time-efficient, less costly but it suffers from the fact that you can only trust what has been documented, or might be facing recall bias, which is simply to depend on a person’s own memory to collect data. For example whether a pregnant lady has had fever in the first three months of her last pregnancy (I personally wouldn’t recall exactly) In this case cohort studies are recorded using relative risk,

24. An example of a cohort study would look like this: let’s say a study was made to see whether there was any relationship between smoking and lung cancer Participants were divided into two groups based on their smoking status, and were followed for ten years to see how many from each group would develop lung cancer. Cohort and case-control are observational studies that can help us draw comparisons between a group based on a risk factor or an outcome of interest

25. Advantage Disadvantage The time sequence of events can be determined this is useful when trying to determine what caused a disease That information about several different outcomes and risk factors can be collected at same time this allows for sub analysis to be conducted on the data. High cost as they can involve a large number of people being followed over a long period of time Generally not suitable to study rare diseases Ensuring that people who started the study stay until the end of the study

26. Experimental Longitudinal In a longitudinal study, researchers repeatedly examine the same individuals to detect any changes that might occur over a period of time No set amount of time is required for a longitudinal study, so long as the participants are repeatedly observed. They can range from as short as a few weeks to as long as several decades. However, they usually last at least a year, oftentimes several. You decide to study how a particular weight-training program affects athletic performance. If you choose a longitudinal study, the impact of natural talent on performance should be eliminated, since that would not change over the study period.

27. Because cross-sectional studies are shorter and therefore cheaper to carry out, they can be used to discover correlations that can then be investigated in a longitudinal study Cross-sectional vs longitudinal example
You want to study the relationship between smoking and stomach cancer. You first conduct a cross- sectional study to see if there is a link between smoking and stomach cancer, and you discover that a link exists in men but not in women. You then decide to design a longitudinal study to further examine this relationship in men. Without the cross-sectional study first, you would not have known to focus on men in particular.

28. Advantage Disadvantage Studies participants over time
Examines whether causes are present before the disorder develops Time-consuming
Higher Research and control cost

29. Experimental Experimental An experimental study design is a type of research design that involves manipulating one or more independent variables to observe their effect on a dependent variable. In other words, it is a controlled study where the researcher intentionally alters one or more variables to measure their impact on the outcome.

30. Experimental Completely randomised design A completely randomized design (CRD) is one where the treatments are assigned completely at random so that each experimental unit has the same chance of receiving any one treatment. This is suitable only for the experiments such as laboratory experiments or greenhouse studies etc. where the experiment material is homogeneous and not for heterogeneous studies.

31. A Completely randomized design uses simple randomization to assign participants to different treatment options (in general, a treatment group and a control group).

32. In CRD each experimental unit is randomly assigned to one of the treatment levels. For e.g.: Let us take an example from optical industry where we want to study the Impact of different varnish types (coating formulations) on the final yield of our lens coating process. Here the experimental unit is the lens on which coating will be done. Here each sample will be randomly allocated to a treatment group hence in this case let’s say we have 60 samples & three types of varnishes (let, say X,Y,Z) thus the entire samples will be divided into three groups of 20 each & one group will be subjected to Varnish X, other to Y & the third to Z. This can be shown as Varnish X Varnish Y Varnish Z Group B Group A Group C We will be taking into account the variability within each unit in the overall sample (SS within) & the variability between groups subjected to the three varnish types X,Y,Z (SS between)

33. Advantage Disadvantage By replicating each treatment the same number of times in each block, we can reduce the error variance and increase the precision of our estimates. Allows us to estimate the treatment effects more accurately The design can be complex to implement Expensive The design is not flexible, as it requires a specific number of blocks and treatments.

34. Experimental Randomised block design A randomized block design groups participants who share a certain characteristic together to form blocks, and then the treatment options get randomly assigned within each block. The objective is to make the study groups comparable by eliminating an alternative explanation of the outcome (i.e. The effect of unequally distributing the blocking variable), therefore reducing bias.

35. With small sample sizes, using simple randomization alone can produce, Just by chance, unbalanced groups regarding the patients’ initial characteristics. In these cases, manually reducing variability between groups by using a randomized block design will offer a gain in statistical power and precision compared to a completely randomized design.

36. Let’s say we observed that the suppliers (let’s say Supplier A,B,C) from which the varnishes (X,Y,Z) are imported also influences the final yield of our coating process. Here the supplier factor will become the blocking variable. In this case the units are first assigned to each block & each unit within the block will be subjected to all the treatments but cannot be assigned to other blocks & other treatments. Thus let’s say we have 180 samples , first we will divide these samples into three groups of 60 I.e. One for supplier A, one for Supplier B & one for supplier C & these three groups will be further subdivided into groups of 20 & one subgroup will be subjected to Varnish X, second with Y & third with Z & likewise for supplier B & C group. Block Varnish X Varnish Y Varnish Z Group 1 Sub group A Sub group 1 Sub group 2 Sub group 3 Group 2 Sub group B Sub group 2 Sub group 3 Sub group 1 Group 3 Sub group C Sub group 3 Sub group 1 Sub group 2 Here we will be taking into account the variability within each unit in the overall sample (SS within), variability in groups amongst the blocks I.e. Supplier A & B (SS blocks) & the variability between groups basis the three varnish types X,Y,Z (SS between)

37. Advantage Disadvantage Blocks are designed to be homogeneous, which means that the units within a block are more similar to each other than units in different blocks. This reduction in variability makes it easier to detect Increased Complexity Limited Flexibility Increased Cost

38. Experimental Factorial design A factorial design allows the researcher to examine the main effects of two or more independent variables simultaneously. It also allows the researcher to determine interactions among variables. A factor is a major independent variable. The independent variable, is what the researcher controls. It doesn’t depend on any other variable in the study. A level is a subdivision of a factor. Factorial design is depicted with a numbering notation. Ex, 2 x 2 factorial design, in this notation, the number of numbers tells you how many factors are there and the number values tell you how many levels.

39. For example , suppose a botanist wants to understand the effects of sunlight (low vs. Medium vs. High) and watering frequency (daily vs. Weekly) on the growth of a certain species of plant. This is an example of a 2×3 factorial design because there are two independent variables, one having two levels and the other having three levels: Independent variable #1 : Sunlight
Levels: Low, Medium, High Independent variable #2 : Watering Frequency
Levels: Daily, Weekly
And there is one dependent variable : Plant growth . then perform a two-way ANOVA to formally test whether or not the independent variables have a statistically significant relationship with the dependent variable .

40. Advantages Disadvantages Can be complex to implement and analyse Factorial designs often require larger sample sizes to achieve . Ability to investigate multiple factors. By manipulating multiple independent variables, it can increase the statistical power of a study. These enable researchers to identify the main products of each independent variable and any interaction effects between them.

41. Experimental Latin square design A latin square is a design in which each treatment is assigned to each time period the same number of times and to each subject the same number of times. Characteristics : It is a square array with an equal number of rows and columns. Each symbol appears exactly once in each row and column. The symbols are arranged in a way that each symbol is paired with every other symbol exactly once. A B C D B C D A C D A B D A B C

42. Suppose different brands of petrol are to be compared with respect to the mileage per liter achieved in motor cars. Important factors responsible for the variations in mileage are The difference between individual cars. The difference in the driving habits of drivers We have three factors – Cars , Drivers , and Petrol brands . Suppose we have 4 types of cars denoted as 1, 2, 3, 4. 4 drivers represented as a, b, c, d. 4 brands of petrol are indicated as A, B, C, D. Now the complete replication will require 4 x 4 x 4 x = 64 numbers of experiments We choose only 16 experiments. Cars 1 2 3 4 Drivers a A B C D b B C D A c C D A B d D A B C

43. Advantage Disadvantage Latin square design ensures that each factor is tested independently of the others Allows for the testing of multiple factors with a minimum number of experimental units Easy to analyse Not flexible as RBD and CRD as the number of treatments is limited to the number of rows and columns. It is seldom used when the number of treatments is more than 12. Sensitive to errors which can affect the accuracy of the results

44. Experimental Ranganathan P. (2019). Understanding Research Study Designs. Indian journal of critical care medicine: peer-reviewed, official publication of Indian Society of Critical Care Medicine, 23( Suppl 4), S305–S307. https://doi.org/10.5005/jp-journals-10071-23314 Le, T., & Bhushan, V. (2019). First aid for the USMLE step 1 2019. New York: Mcgraw-Hill Education. https://pmc.ncbi.nlm.nih.gov/articles/PMC6996664/ [National library of medicine] Blair, R. C., & Taylor, R. A. (2008). Biostatistics for the health sciences (1 st ed.). Upper Saddle River, NJ: Pearson Prentice Hall https://youtu.be/iCgC32Cf8ac?si=Hb3TpOStKpZQczNB [ATP] https://youtu.be/Jd3gFT0-C4s?si=72529Nhr3-G168ol [let’s learn public health] REFERENCES

45. THANK YOU