Table 2. Characteristics of cohort studies examining the relationship between family history and risk of lung cancer.

							Standardising variables
Study	Region/country	Years of data collection	Number of cases	Data sourcesa	Type of relative	Sex of cases	Age	Smokingb (case/relative)	Family size	Other
Cannon-Albright (1994)	Utah, USA	1952–92	2477	CR	Parent/sibling	Both	Yes	No/no	No	Sex, birthplace
Goldgar (1994)	Utah, USA	1952–92	2228	CR	1st degree	Both	No	No/no	No	No
Hemminki (1998)	Sweden	1958–94	N/Ac	CR	Offspring	Both	Yes	No/no	No	Sex
Hemminki and Vaittinen (1999)	Sweden	1960–94	35 831	CR	Offspring	Both	Yes	No/no	Yes	No
Hemminki (1999)	Sweden	1958–94	N/A	CR	Parent	Both	Yes	No/no	N/A	No
Poole (1999)	USA	1959–72	877	IS	Parent/sibling	Female	Yes	Yes (qual)/no	Yes	Ethnicity, education, body mass index, hormonal factors
Hemminki (2001b)	Sweden	1958–94	N/A	CR	Parent/sibling	Both	No	No/no	No	No
Hemminki (2001a)	Sweden	1958–96	N/A	CR	Parent	Both	Yes	No/no	N/A	Sex
Dong (2001)	Sweden	1958–96	N/A	CR	Parent/sibling	Both	Yes	No/no	No	Sex
Hemminki (2002)	Sweden	1961–98	4524	CR	Parent	Both	Yes	No/no	N/A	Sex
Chene (2002)	Sweden	1958–96	N/A	CR	Parent/sibling	Both	No	No/no	No	No
Hemminki (2003)	Sweden	1961–98	N/A	CR	Parent	Both	Yes	No/no	N/A	Sex
Li (2003)	Sweden	1961–98	4524	CR	Parent	Both	Yes	No/no	N/A	Sex, region, period, socioeconomic status
Hemminki (2004)	Sweden	1991–00	5493	CR	Parent/sibling	Both	Yes	No/no	Yes	Sex, region, period, socioeconomic status
Li (2004)	Sweden	1991–00	5290	CR	Parent/sibling	Both	Yes	No/no	No	Sex, region, period, socioeconomic status
Jonsson (2004)	Iceland	1955–02	2756	CR	1st/2nd/3rd degree	Both	Yes	No/no	Yes	Sex
Li (2005)	Sweden	1961–00	55 238	CR	1st degree	Both	Yes	No/no	No	Sex, region, period, socioeconomic status

^aData sources: CR=cancer registry; IS=index subject.

^bQual=qualitative.

^cN/A=not available.