Epidemiology and control of infectious diseases

32 Epidemiology and control of infectious diseases



Introduction


Epidemiology is defined as ‘The study of the distribution and determinants of health-related states or events in specified populations and the application of this study to control of health problems’ (Last, Dictionary of Epidemiology).


In epidemiology, we are concerned with populations rather than individuals. What we want to know of a disease in a population is: who, where and when. Hepatitis A outbreaks are often associated with institutions, restaurants and specific food. It is therefore important to determine who – which individuals ate potato salad, where – in a nursing home, when – 1 February 2010 – developed hepatitis A.


The field of epidemiology is divided into observational and interventional epidemiology.


Observational studies are either descriptive, describing the frequency of a disease in the population, or analytical, investigating associations between risk factors and disease. Disease surveillance describing the number of notifiable disease cases such as measles, meningitis, or cholera is an example of observational descriptive epidemiology. Studies showing an association between human papillomavirus infection and cervical cancer are examples of analytical epidemiological studies.


Interventional or experimental epidemiological studies are designed to test a hypothesis by allocating an exposure or intervention to one group of people but not the other and measuring the disease outcome. Examples of intervention studies are randomized–controlled trials investigating vaccine efficacy.


Epidemiologists talk about outcomes and exposures. The outcome is usually a disease or event such as death, infection or onset of new symptoms. Sometimes outcomes are laboratory markers, for example C-reactive protein (an acute-phase protein) or HIV viral load. These outcomes are called intermediate outcomes as they do not represent a definite endpoint. Exposures are either risk factors, for example a specific behaviour or harmful substance, or interventions such as drugs, vaccines or health education.



Outcome measurements


It is important to clearly define health-related outcomes. A definition should include the methods used to identify a case, the boundaries of a case and the unit of analysis.


Eye disease secondary to Chlamydia trachomatis (Chapter 25) is an important public health issue globally. The trachomatous inflammation is graded clinically into whether it involves follicular inflammation of the eyelid, abnormally positioned eyelashes or corneal scarring. When defining a case of trachomatous inflammation, it is important to describe (1) the methods and procedures used to determine a case: clinical examination versus direct immunofluorescence microscopy of conjunctival smear, (2) the boundaries of a case: follicular inflammation only versus all 3 grades, and (3) the unit of analysis: one or two eyes.


Disease prevalence and incidence are the two main types of measure of occurrence used in epidemiology. Prevalence is the number of existing cases in a population at a given point in time. Incidence is the number of new cases occurring in a population during a specified period of time.


Prevalence is influenced by occurrence of new cases (incidence) and the duration of each case. Prevalence of diseases with short durations such as viral gastroenteritis is mainly influenced by incidence. Prevalence of chronic diseases with relatively low mortality is likely to be high even if incidence is low. An example of the interaction between prevalence, incidence and mortality is shown in Box 32.1.



Box 32.1 imageLessons in Microbiology



The interaction between prevalence, incidence, mortality and treatment


When HIV is introduced into an HIV-negative population HIV prevalence and incidence grow exponentially (Fig 32.1). As more people become infected the proportion of individuals not infected decreases. With fewer individuals susceptible to infection the likelihood that an infectious HIV-positive individual will be in contact with an HIV-uninfected individual is reduced. This in turn reduces incidence, but prevalence continues to rise. The median time of survival in the natural course of HIV disease is 6–8    years. Thus, after a time-lag, HIV mortality grows, which reduces HIV prevalence. However, if HIV treatment becomes available, survival is prolonged and prevalence grows.


image

Figure 32.1 HIV prevalence, incidence and mortality in a hypothetical population. (Based on data from: Trends in HIV incidence and prevalence: natural course of the epidemic or results of behavioural change? unaids in collaboration with Wellcome Trust Centre for the Epidemiology of Infectious Disease, 1999.)



Types of epidemiological studies




Cross-sectional study


Cross-sectional studies measure the frequency of an outcome and/or exposure(s) in a defined population at a particular point in time (Fig. 32.2A). These studies can be either descriptive, measuring the burden of disease, or analytical, comparing the frequency of disease in people exposed and unexposed to a risk factor.


image image image image

Figure 32.2 (A) Cross-sectional study: Outcome and exposure are determined at the same time. (B) Case–control study: Cases with the outcome and controls without the outcome are identified and their exposure status determined.   (C) Cohort study: Individuals with and without the exposure are identified and followed until they develop the outcome or until study end. (D) Intervention study: Individuals are allocated an intervention (exposure) and are followed until they develop the outcome or until study end.


Examples of study questions addressed by cross-sectional studies are:



• What proportion of the population has evidence of a past infection with Lyme disease?


• Is hepatitis B associated with hepatocellular carcinoma?


Cross-sectional studies are relatively cheap and quick to do. They are particularly useful to determine the scale of a problem, to generate hypotheses for possibly causal associations and to evaluate diagnostic tests (Box 32.2). Cross-sectional studies measure disease prevalence. It is therefore difficult to differentiate between exposures causing the disease and improving the survival. With outcome and exposure determined at the same time, there remains uncertainty if the exposure preceded the outcome, which is a crucial requirement for causality. Sometimes it is difficult to exclude reverse causality (the outcome caused the exposure).



Box 32.2 imageLessons in Microbiology



Sensitivity, specificity, positive and negative predictive value


New diagnostic tests are usually evaluated using a cross-sectional study design. The new test is compared against a gold standard test and sensitivity and specificity are determined.


Sensitivity is the proportion of true positives correctly identified by the new test and specificity is the proportion of true negatives correctly identified by the new test. Both sensitivity and specificity are intrinsic to the test and do not vary according to disease prevalence. However, they can be influenced by operators and environmental conditions.


From the patient’s and physician’s point of view, the more interesting question is, ‘What are the chances for me having the disease if I have a positive test result?’ This question is answered by the positive predictive value (PPV) which is the proportion of individuals with a positive test result who actually have the disease. The negative predictive value (NPV) is the proportion of individuals with a negative test result who are free of disease. Both PPV and NPV are related to sensitivity and specificity of a test but also to the prevalence of disease in a population.


The Xpert MTB-RIF is a new automated molecular test for diagnosis of Mycobacterium tuberculosis (Chapter 19). Diagnosis of tuberculosis (TB) relies on smear microscopy in most resource-limited settings and liquid culture in resource-rich settings. Smear microscopy has a low sensitivity and detects only patients with relatively advanced disease. Liquid culture is the gold standard of TB diagnosis, but takes days to weeks to become positive. A hypothetical evaluation study in 7000    TB suspects in a high TB prevalence setting revealed a sensitivity of the Xpert MTB-RIF of 92% and a specificity of 98% (Table 32.1A). The prevalence of TB among these 7000    TB suspects was 10%. PPV was 93% and NPV 99%.


The evaluation study was repeated in a population survey with 10 000 participants, among whom TB prevalence was 1%, sensitivity and specificity remained the same, but PPV was 53% and NPV 100% (Table 32.1B).

Only gold members can continue reading. Log In or Register to continue

Related posts:

  1. Obstetric and perinatal infections
  2. Immune defences in action
  3. The bacteria
  4. Sexually transmitted infections

Stay updated, free articles. Join our Telegram channel
Tags:
Jul 9, 2017 | Posted by in MICROBIOLOGY | Comments Off on Epidemiology and control of infectious diseases

Full access? Get Clinical Tree
Get Clinical Tree app for offline access