Table 3.

Stepwise multiple linear regression to identify the significant predictors of cancer incidence risk

	All countries n = 178				Developed countries (High Quality Data)n = 65			Developing countries (Low Quality Data) n = 113
	Rank	Predictor	Beta	Adjusted R2	Predictor	Beta	Adjusted R2	Predictor	Beta	Adjusted R2
All cancers excl. Non-melanoma skin cancer (C00–97, but C44) - all ages: both sexes	1	Family Size	−0.473***	0.502	Family Size	−0.503***	0.403	Family Size	−0.396***	0.145
	2	Life Expectancy	0.330***	0.552	Life Expectancy	0.327***	0.487	^	^	^
All cancers excl. Non-melanoma skin cancer (C00–97, but C44)- all ages: female	1	Life Expectancy	0.385***	0.402	Family Size	−0.424***	0.320	Life Expectancy	0.296*	0.075
	2	Family Size	−0.351***	0.459	Life Expectancy	0.353***	0.417	^	^	^
All cancers excl. Non-melanoma skin cancer (C00–97, but C44) - all ages: male	1	Family Size	− 0.576***	0.572	Family Size	−0.563***	0.450	Family Size	−0.522***	0.262
	2	Life Expectancy	0.254***	0.601	Life Expectancy	0.268***	0.505	^	^	^
All cancers excl. Non-melanoma skin cancer (C00–97, but C44) – 0-49: both sexesa	1	Family Size	−0.754***	0.567	Family Size	−0.662***	0.432	Family Size	−0.640 ***	0.296
	2	^	^	^	^	^	^	Urbanization	−0.246*	0.340
All cancers excl. Non-melanoma skin cancer (C00–97, but C44) – 0-49: femalea	1	Family Size	− 0.755***	0.568	Family Size	−0.641***	0.404	Family Size	−0.531***	0.272
All cancers excl. Non-melanoma skin cancer (C00–97, but C44) – 0-49: malea	1	Family Size	−0.730***	0.530	Family Size	−0.685***	0.464	Family Size	−0.610***	0.247
	2	^	^	^	^	^	^	Urbanization	−0.291**	0.312

Note: Significance level: ***p < 0.001, **p < 0.01, *p < 0.05

Variables (log-transformed) entered for multiple linear regression (stepwise) analysis: Family Size, Life Expectancy (e₆₀), GDP PPP, Urbanization and Biological State Index (I_bs)

^aLife expectancy (e₅₀) was not included as it is not relevant in population segment aged 0–49 years

^ No other variable identified as the significant predictor

Data sources and variable meanings:

The International Agency for Research published cancer incidence rates (per 100,000 in 2012) of all cancers incidence rate by sex (total, male and female, 0–49 years and all ages respectively); bladder, breast, cervix uteri, colorectum, corpus uteri, ovary and stomach

The World Bank data: Total Fertility Rate (between 2009 and 2011), GDP PPP (per capita purchasing power parity in current international $ in 2010) and Urbanization (the percentage of total population living in urban areas in 2010)

The United Nations data: Life expectancy (e₆₀, 2005–2010), the total population in households and the number of households for calculating household size

United Nations published (2008) country specific fertility data and WHO published (2012) life tables were used for calculating the Biological State Index (I_bs)