Classifying study participants in a case-control study

How to label study participants in case-control study as Type-A and non-Type-A personality according to Jenkins Activity Survey ?

How do I classify my study participants of a case-control study into type-a and non-type a personality based on Jenkins activity survey? It is an online tool with no cutoff point. It only shows score. Can anyone help me?


0 / 0  ·  15 Answers  ·  467 Views

All Answers (15)

  • Bekele Belayihun · Haramaya University
    Epidemiologists benefit greatly from having case-control study designs in their research armamentarium. Case-control studies can yield important scientific findings with relatively little time, money, and effort compared with other study
    designs. This seemingly quick road to research results entices many newly trained epidemiologists. Indeed, investigators implement case-control studies more frequently than any other analytical epidemiological study.
    Unfortunately, case-control designs also tend to be more susceptible to biases than other comparative studies. Although easier to do, they are also easier to do wrong. Five main notions guide investigators who do, or readers who assess, case-control studies. First, investigators must explicitly define the criteria for diagnosis of a case and any eligibility criteria used for selection. Second, controls should come from the same population as the cases, and their selection should be independent of the exposures of interest. Third, investigators should blind the data gatherers to the case or control status of participants or, if impossible, at least blind them to the main hypothesis of the study. Fourth, data gatherers need to be thoroughly trained to elicit exposure in a similar manner from cases and controls; they should use memory aids to facilitate and balance recall between cases and controls. Finally, investigators should address confounding in case-control studies, either in the design stage or with analytical techniques. Devotion of meticulous attention to these points enhances the validity of the results and bolsters the reader’s confidence in the findings.

    Case control study
    Also known as Case-referent study
    subjects are selected on the basis of whether they do(cases) or do not (controls) have a particular disease under study
    groups are then compared with respect to the proportion having a history of an exposure or characteristics of interest.
    If all the cases of the disease have been diagnosed at the time the investigators initiates the study, it is a retrospective case control study
    If the study is begun and all new cases that are diagnosed within the next period of time are included, the study is a prospective
    Thus in the context of these definitions the use of the term retrospective to refer to all case-control studies is inappropriate
    Depending on the certainty of the diagnosis, and the amount of information available, it is often useful to perform analyses separately for cases classified as definite, probable, or possible
    Setting clear definition of cases
    Incident Vs Prevalent cases
    Representing spectrum of disease: mild, moderate and severe groups
    selection of control
    involves consideration of a number of issues including :
    -the characteristics and source of the cases
    - the need to obtain comparable information from cases and controls
    - practical and economic considerations.
    the control subjects should be selected to be comparable to the case

    how many contorl group
  • Naresh Makwana · M.P.Shah Medical College
    Thank you for your reply, Sir.
    Definitely, the answer given above is quite suitable for conducting a case-control study and it gives an immense idea about selection of cases and controls.
    But, my question (not the label) was how to classify the participants of case-control study into TYPE-A and NON-TYPE-A Personality based on Jenkins Activity Survey ??
  • Naresh Makwana · M.P.Shah Medical College
    Can anybody share something on Jenkins Activity Survey?? How to analyze the score??
  • Tatiana Andreeva · National University of Kyiv-Mohyla Academy
    Do you consider type A as the outcome or as an exposure?
  • Jagdish Khubchandani · Ball State University
    I assume you are considering type A as an outcome? If so, look at each person's score, compute population scores (mean and median), have arbitrary cut points (e.g. above 80 percentile or 75 percentile)
  • Naresh Makwana · M.P.Shah Medical College
    type A is considered as an outcome, based on data collected by oral questionnaire. I definitely calculate mean and median of the score, but how can i decide arbitrary cut-off points? Is there any reference for cut-off point??
    thanks for reply.
  • Tatiana Andreeva · National University of Kyiv-Mohyla Academy
    If your study is an exploratory one, I would really explore the real picture instead of doing anything arbitrary. If you have a promising hypothesis about certain determinants it is worth analyzing associations with a continuous outcome before switching to binary variables. Otherwise if your cut-off point is wrong you increase the type 2 error. I can go into more details if this explanation is insufficient but hope that you have got it.
  • Jagdish Khubchandani · Ball State University
    Tatiana, I guess you missed the point I was trying to make. In this scenario, I assumed that the investigator is designing a case control study. If he has to assign people into groups, there is no way he could do it without having cut points for scores (cases versus controls) or there has to be a marker for cases. The point you make about type 2 errors holds true for many studies in behavioural science (due to arbitrary cut points). However, your argument remains valuable if the investigator prefers not to conduct a case control study.

    Naresh, see what published literature suggests about cut-points. There are a few studies pointing out to score range for type A and type B.
  • Tatiana Andreeva · National University of Kyiv-Mohyla Academy
    I suggested to avoid contamination. However, it's definitely for the author to decide as we even do not know the hypotheses to comment properly and reliably.
    All the best to all.
  • Jagdish Khubchandani · Ball State University
    I agree 100% Tatiana, the investigator needs to be very careful about contamination.
    Thank you for the wonderful suggestion!
  • Naresh Makwana · M.P.Shah Medical College
    Thank you for suggesting the ways of classifying. Let me tell something more about my study. Basically the study involves the cases of hypertension(as per JNC criteria) as well as controls for the hypertension (of course, with matching variables). Data were collected regarding the various variables. One of the variable, we have to assess is the type of personality and we had used Jenkins activity survey (JAS) questionnaire format to evaluate personality. Now we have data and we have to analyze it, but we are not knowing the cut-off point of the scores obtained from JAS, to classify the cases and controls into type-A and non-type A personality. How can we say by applying statistical test that type-A is significantly associated with hypertension or not??
  • Marcelo Burattini · University of São Paulo
    You could try to perform any sort of discriminant analysis in order to identify the cut-points that better identify Typa A from non type A in your sample. Ideally you should do this in a part of the sample and test the accuracy of the model in the rest of it, provided you have a large enough number to do procced this way.
  • Tatiana Andreeva · National University of Kyiv-Mohyla Academy
    Thanks for more details about the study.
    As it was suspectable earlier, type-A is not your outcome. Obviously, hypertension is your outcome.
    However, you are not limited in your analysis to 2x2 tables anyway. You can have an independent continuous variable (type-A) and dependent dichotomous variable (hypertension). Logistic regression analysis deals with such data perfectly.
  • Naresh Makwana · M.P.Shah Medical College
    Thank you for explaining. Now I fully understood that I should not worry about type-A and non-type-A personality. Logistic regression is the right method for this type of data. Again thank you very much, Tatiana Andreeva.
  • Jagdish Khubchandani · Ball State University
    Naresh, glad you could find the answer! try different types of logistic regressions. SPSS gives you options.

Question Followers (6)