front |1 |2 |3 |4
|5 |6 |7 |8 |9 |10 |11
|12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |review |
Estimated values of A, L, R0 and PC for different infections and populations.
Infection | Place | Time | A* |
L |
R0 |
PC§ |
Measles | USA | 1955-58 | 5-6 |
70 |
13 |
92% |
India | 1978 | 2-3 |
55 |
23 |
96% |
|
Senegal | 1964 | 1-2 |
40 |
28 |
96% |
|
Rubella | Sweden | 1965 | 9-10 |
75 |
8 |
88% |
Poland | 1970-77 | 6-7 |
70 |
11 |
91% |
|
Gambia | 1976 | 2-3 |
40 |
17 |
94% |
|
Measles | England and Wales | 1948-68 | 4-5 |
70 |
16 |
94% |
Mumps | England and Wales | 1970-82 | 6-7 |
73 |
11 |
91% |
Rubella | England | 1975-77 | 9-10 |
73 |
8 |
88% |
A: Average age at infection (in years)
L: Expectation of life at birth (in years)
R0: Basic reproductive rate
PC: Critical proportion of vaccination above which the infection can be eradicated
* Source of data: Anderson R, May R. Infectious diseases of humans. Oxford: OUP; 1991. p.51.
Source of data: Demographic Yearbook (various years). New York: United Nations.
Calculation from A and L: R0 = L / A for developed countries (high L), R0 = 1 + L / A for developing countries (low L).
§ Calculation from R0: PC = 1 - 1 / R0.
_____________________________________________________________________________
Observe the table and discuss possible social and viral characteristics which may relate to different values of A and L. Observe how R0 and PC depend on A and L, and therefore on the social and viral characteristics.
front |1 |2 |3 |4
|5 |6 |7 |8 |9 |10 |11
|12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |review |