Dissemination of periodic mammography and patterns of use, by birth cohort, in Catalonia (Spain)

BMC Cancer

Table 2 Mixed effects logistic model* for estimating the percentage of women receiving periodic mammograms by age and cohort of birth

Fixed effects
Parameter	Value	Standard error	p-value
ϕ ₁(Asym)	0.968	0.065	< 0.001
ϕ ₂(xmid)	45.004	3.091	< 0.001
ϕ ₃(scal)	5.384	0.640	< 0.001
Random effects
Parameter	Standard deviation	Residual
ϕ ₁(Asym)	NS
ϕ ₂(xmid)	6.697	0.002
ϕ ₃(scal)	NS
Random effect for parameter ϕ ₂ ( *xmid* ) by cohort of birth
Cohort of birth	Random effect **	Fixed plus random ***
1938–1942	11.771	56.78
1943–1947	5.193	50.20
1948–1952	-0.244	44.76
1953–1957	-3.883	41.12
1958–1962	-5.160	39.84
1963–1967	-7.677	37.33

NS: Not significant.
* The model is:
$p = \frac{ϕ_{1}}{1 + e x p [(ϕ_{2} - a g e) / ϕ_{3}]}$ were p indicates the percentage of women receiving periodic mammograms, ϕ ₁ (asym) is the horizontal asymptote as age increases or the percentage at which the curve levels off, ϕ ₂ (xmid) indicates the age value at which approximately half of the population is receiving periodic mammograms and ϕ ₃ (scal) indicates the difference in years between the age at which 3/4 of the population is receiving periodic mammograms and the age ϕ ₂.
(**) The values indicate the departure from the fixed effect (45.0 years of age) for each cohort of birth. (***): The values can be interpreted as the age at which approximately 50% of the women will be receiving periodic mammograms.

ISSN: 1471-2407