2.1 Measuring Disease: Mortality and Morbidity

Unnumbered Figure: Link

• •

  In addition to defining disease we also need to decide how to count disease

• •

  incidence ($I$): number of new cases of the disease within a specified period of time

• •

  prevalence ($P$): number of existing cases of the disease at a particular point in time

• •

  incidence and prevalence are often very different for the same disease in the same population

• •

  prevalence depends upon incidence but also depends upon the duration of the disease

• •

  a rarely cured chronic disease will have a greater prevalence than incidence, whereas, a disease which leads to death soon after diagnosis will have a higher incidence than prevalence

• •

  if the incidence rate and average duration ($\overline{D}$) of the disease are constant over time, then

  $$P=I\cdot \overline{D}$$

• •

  example: incidence rate 9.1 cases per 100 person-years, average disease duration 3.3 years $\Rightarrow$ prevalence about 30 cases per 100 persons

• •

  both incidence and prevalence are typically measured on a relative scale: for example, to the size of population at risk at time $t$, for example, mid-year

• •

  prevalence ($P$)

  $$P=\frac{\text{number of people with the disease at a specified time }t}{\text{number in the population at risk at the specified time }t}$$

• •

  incidence rate ($I$)

  $$I=\frac{\text{number of people who develop disease in a specified time period}}{\text{sum of the length of time during which each person in the population is at risk}}$$

  The total observation time is known as the person-time of observation and is commonly expressed as person-years. Estimates of incidence rates are often given by the units: per person-years or per 1000 person-years or per 100,000 person-years

In a study in the USA the incidence rate of stroke was measured in 118,539 who were aged 30-55 years. The data and incidence rate by smoking status is given below. The rate was higher for smokers than non-smokers and ex-smokers were intermediate

• •

  prevalence measures

  • –

    better suited for descriptive as opposed to analytical studies. For example, the burden of a disease in a geographical region $\to$ make appropriate plans for allocation of resources

  • –

    whilst such measures may suggest possible causal factors they cannot provide convincing evidence

• •

  incidence

  • –

    better for studying aetiology since they can better establish the sequence of events

  • –

    they are not susceptible to survival bias

  • –

    survival bias: suppose taking vitamin C results in better survival and not taking vitamin leads to rapid death. Prevalence may then be related to vitamin C consumption

• •

  an alternative measure of disease occurrence used to estimate risk

• •

  cumulative incidence risk ($CI$)

  $$CI=\frac{\text{Number of people who get a disease during a specified period}}{\text{Number of people free of the disease at the beginning of the period}}$$

• •

  risk (probability) of developing a disease (unlike incidence)

• •

  example: 23 deaths in one year out of a population with 10,000 persons at risk at beginning of the year $\Rightarrow$ $CI=23/10,000=0.23\%$

• •

  deaths are vital registrations and information on death by cause is routinely collected in many countries

• •

  the mortality rate is a measure of the disease frequency in populations

• •

  such data provide invaluable information on temporal and graphical trends in disease

• •

  the crude mortality rate is given by the number of deaths in a specified period of time, divided by the average population at risk during that period multiplied by the length (years) of the study period (person-years)

• •

  the value is again typically multiplied by 1000 etc

• •

  typically the crude rate is not that informative due to confounders. A more useful summary is the standardised rate

• •

  in practice standardisation is typically, minimally, with respect to age and gender

• •

  we will consider standardisation later in the module