Epidemiology studies

HideShow resource information
View mindmap
  • Epidemiology
    • Epidemiology is the study of the distribution and determinants of health-related states or events (including disease), and the application of this study to the control of diseases and other health problems (WHO, 2016)
      • History: linked to John Snow using the river to determine the effects of a disease, however it is now outdated to do so. Approach was continued towards outbreaks of infectious diseases.
    • Exposure: Putative casual factor: demographical, behavioural, genetic and environmental factor. Maybe be a rick factor or protective factor
    • Outcome: state of health or disease; symptom or measure associated with disease risk death or survival
    • Nutritional epidemiology: the study of nutritional determinants of disease in human population
      • Exposures of interest include: diet, nutrient intake and nutritional status
      • Relatively a new branch
      • Focuses on chronic diseases
      • Early investigations into deficiency diseases
        • Study: Scurvy (Lind, 1722)-observational study based on populations deprived from fresh fruit and veg. Linked to Vitamin C.
        • Study: Beriberi (Takaki, 1906). Lack of Thiamin in diet. Common in Japanese sailors (rice only diet)
        • Rickets (Chick,1919): mainly observed in cities with poorer children. Shown to be Vitamin D deficiency
      • Limitations: Complex nature of diets, metabolism of food, different requirements of different stages of life; culturual/socio-economical/psychological factors associated with choosing food; multifactorial aetiology of chronic diseases
      • Study designs
        • Purpose: described patterns (distribution & frequency); formulates hypotheses (determinants); test hypotheses and look for associations.
        • Validity
          • Internal validity: Do the results reflect the true situation?
            • Bias: selection 7 information/ observational bias. Affects study population (self selection bias, referral bias and diagnostic bias). It effects measurement of exposure and outcome. Measurement of dietary intake can lead to information bias
          • External validity: can the results be generalised throughout the population?
        • Confounders: associated with exposures, in chance of risk factor for disease independent of it's association with exposure, but needs to be measured to make sure experiments are controlled.
        • Observational studies
          • Descriptive studies
            • Groups selected by exposure or disease and investigating distribution or determinants
            • Describe patterns of disease in a population
            • Triad: When did the exposure occur?, Who (ethnic groups, special groups, family groups) and where (geographical difference)
            • Interactions: 2 or more factors may correlate with disease incidence and with each other.
            • Types
              • Case reports: examines experiences in individuals and gives first identification of new diseases.
              • Ecological studies: describes patterns of diseases in populations.
              • Cross-sectional studies: exposure and disease assessed simultaneously in individuals
          • Cohort studies
            • Can be split into observation and analytic
              • Groups can often be defined in common characteristics
              • Case control studies
                • Groups are defined on outcomes: cases have outcome or disease and controls do not have outcome or disease
                  • Addresses problems of rare diseases and long latent periods, ideally suited to the study of chronic diseases
                    • Oftern first lined of enquiry for rare or new diseases with unknown aetiology, or when a rapid response is required
                • Odds ratio is calculated of exposure
                • Prone to selection and recall bias and risk of confounding factors
                  • Considerations for selection process: all those who have the disease must have an equal opportunity of being identified, and criteria for diagnosis must be clearly specified and there should be enough diversity of exposure.
                  • Selection of cases: hospital based or population based. Selection of controls: rep of population that cases are from. They must match case in more than one way to eliminate confounding factors and reduce bias.
                • Studies
                  • Parish et al (1995): cigarette smoking & myocardinal infarction
                  • Doll (1947) lung cancer and smoking
            • Subjects are disease-free/not developed outcome
            • Measured at baseline and followed up over time
            • Types
              • Retrospective: all events have occurred within the study, quicker and cheaper, efficient for diseases with long latency periods and information for confounding issues might be limited and exposure incomplete
              • Prospective: outcomes not yet occurred when study begun, more recent records can be used, direct questioning on confounding factors, all data has already been done and efficient for diseases which have a longer period of time to develop or appear on the radar.
            • Strengths: valuable when the exposure is rare, can examine multiple effects on one exposure, can elucidate temporary relationship between exposure and disease and allows direct measurement of incidence of disease in the exposed and non-exposed groups.
            • Limitations: inefficient for rare diseases, expensive and time consuming and validity of results are lost due to follow up
            • Studies
              • Nurse health study (1976)
              • EPIC study (1993)
          • Case control studies
            • Groups are defined on outcomes: cases have outcome or disease and controls do not have outcome or disease
              • Addresses problems of rare diseases and long latent periods, ideally suited to the study of chronic diseases
                • Oftern first lined of enquiry for rare or new diseases with unknown aetiology, or when a rapid response is required
            • Odds ratio is calculated of exposure
            • Prone to selection and recall bias and risk of confounding factors
              • Considerations for selection process: all those who have the disease must have an equal opportunity of being identified, and criteria for diagnosis must be clearly specified and there should be enough diversity of exposure.
              • Selection of cases: hospital based or population based. Selection of controls: rep of population that cases are from. They must match case in more than one way to eliminate confounding factors and reduce bias.
            • Studies
              • Parish et al (1995): cigarette smoking & myocardinal infarction
              • Doll (1947) lung cancer and smoking
        • Experimental studies
          • Intervention studies
            • Investigator controls exposure between two groups.
            • Types of groups: intervention or control groups
            • Used to investigate aetiology or evaluate the effectiveness of a treatment/prevention in healthy or at risk individuals
            • Intervention is independent variable and outcome is dependent variable
            • Crossover design
              • All participants receive treatment and control
              • Order of treatment allocation randomly assigned: which allows less variation between subjects
              • Only suitable for short-term interventions
            • Limitations: randomised treatment allocation, blinding, sample size large enough for statistical power, follow up period long enough to accumulate sufficient end points, monitoring of compliance, recruitment and cost.
              • Ethical issues must be confirmed prior by all participants
            • Studies
              • Coppel et al (2010) lifestyle over an above drugs in diabetes: improved glycaemic control
    • Limitations: Less effective within studies with weaker links; potential for many biases to occur; can not justify casual relationships and there is a high risk of over interpreting data/evidence.
    • Pros:Identify's health problems within a population; monitoring/evaluations/generate hypothesis; inform disease prevention and help improve public health
    • Types of diseases
      • Chronic: Long latent period, and exposure often unknown, low frequency of disease despite high lifetime risk, not readily reversible, multiple causes (not just diet) and might be due to excess or deficiency
      • Deficiency disease: short latent period, common in those with low intakes otherwise can be rare effects usually reversible and causation/association readily tested by experiment

Comments

No comments have yet been made

Similar Sports Science resources:

See all Sports Science resources »See all Study designs resources »