<aside> 💡 Measuring Health 2

</aside>

  1. Calculate measures of association such as relative risk, absolute risk, and an odds ratio.

    1. Relative Risk: comparing one to another - relative to exposed or non-exposed conditions
      1. [risk of developing outcome from exposure] / [risk of developing outcome without exposure]
      2. [a/a+b] / [c/c+d]
      3. [Cle]/[Clo]
      4. Represented as a decimal
      5. Also called risk ratio
      6. Cohort Study Only
    2. Absolute Risk: actual risk of getting a disease when exposed to a condition over time; same as cumulative incidence

      1. Observed/calculated probability that an individual will develop a disease in a specified period of time, based on the population under study - another way of referring to 'incidence'

      2. Cohort Study Only

      3. Reported as a fraction out of 1000

      4. [# of events of interest] / [Total #]

        1. CIe = a/a+b
        2. CI0 = c/c+d

      5. Number needed to treat: # of people that need to be treated to prevent one additional person getting the outcome

        1. 1/[ARR]

      6. Risk difference

      7. More realistic compared to RRR; does not tend to inflate association

        1. Excess Risk:
          1. CIe - CIo
        2. Risk Reduction:
          1. Clo - Cle
    3. Odds Ratio
      1. Case control only
      2. [odds that person with the disease was exposed] / [odds that a person with disease was not exposed]
      3. [a/c] / [d/b] = [ad] / [bc]

  2. Interpret relative risk, absolute risk, and an odds ratio when given these measures, and explain what they mean

    1. Relative risk 1.

      1. Estimate magnitude of association between exposure and disease; likelihood of developing disease in exposed relative to unexposed
      2. Those who had [exposure] had [RR] times the risk of developing [the outcome] compared to those who did not have the [exposure] during the [given time frame].
      3. Needs to be associated with ARR, as RRR alone can be misleading as it can make the magnitude of relatively small ARR or ER seem large
      • RR > 1 = risk of disease may be increased as a result of the exposure
      • RR < 1 = risk of disease may be decreased as a result of the exposure
      • RR = 1 = risk of disease in the exposed and unexposed groups are equal - disease is unlikely to be related to exposure

    2. Absolute risk

      1. [With (Cle) OR Without (CI0)] [Exposure] [ARR] [more or less] people out of 1000 developed [the outcome] over [the time period].
      2. Excess Risk
        1. By having [exposure], [ARR] more people out of 1000 developed [the outcome] over [the time period].
        2. Increased relative to baseline risk
      3. Absolute Risk Reduction
        1. By having [exposure], [ARR] less people out of 1000 developed [the outcome] over [the time period].
        2. Reduced relative to baseline risk
        3. Exposure said to be protective

    3. Odds ratio 1.

      1. The odds that people with the [outcome] were exposed to the [exposure] is [OR] times the odds of those without the [outcome] being exposed
        • OR > 1 = exposure may increase the risk of disease
          • odds of exposure in cases is higher than the odds of exposure in cases is higher than the odds of exposure in the controls
        • OR < 1 = exposure may reduce risk of disease
          • Odds of exposure in the cases is lower than the odds of exposure in the controls
        • OR = 1 = odds of exposure are equal - exposure is unlikely to be related to disease development

  3. Interpret a ‘p value’ and analyse what it means in relation to the study in which it was reported

    1. p-value is the probability that the results are due to chance, that the null hypothesis is true
      1. Null hypothesis
        1. The hypothesis that there is no significant difference between specified populations, any observed difference being due to sampling or experimental error.
    2. p < 0.05 is the statistical cutoff
      1. P > 0.05
        1. Reject alternative hypothesis,
        2. Accept null hypothesis
        3. Results have a greater than 5% chance of being due to chance, and less due to any real relationship between specified populations
      2. P < 0.05 - is p-value is low, the null must go
        1. Reject null hypothesis
        2. Accept alternative hypothesis
        3. Results have less than 5% chance of being due to chance, and more due to a statistically significant relationship between specified populations

  4. Interpret a confidence interval and analyse what it means in relation to the study in which it was reported.

    1. 95% Confidence interval is the interval of values in which one is 95% sure that the true value lies within
    2. If the 95% CI crosses an OR/RR of 1, then results statistically insignificant p > 0.05
      1. This is because an OR/RR of 1 confirms the null hypothesis that disease and exposure is unrelated
      2. An OR/RR that crosses over and RR/OR of 1 means that the results could be due or not due to an event of interest
      3. Thus, statistical significance can be accessed from 95% CI without use of p-value based on if the 95% CI includes the value of 1

  5. Differentiate between qualitative and quantitative research methodologies

  6. In relation to qualitative research methods:- a. Discuss their advantages b. Recognise study questions that might best be answered by them c. Recognise studies that use them