Get Printable Version

The table below summarizes the available evidence on the risks and benefits of hormone therapy. Included are all randomized trials and meta-analyses looking at disease incidence or mortality. Inclusion of observational trials was dependent on size: case-control studies with 500 or more cases of the disease and cohort studies with at least 2,000 women are included here.

The year noted is when the study results were published. The name following the title of the study is the last name of the first author on the publication.

E = estrogen; P = progestin

Year	Study	Study Type	Participants	Hormones	Results	Other Notes
1984	Kaufman	Case-control	1,610 cases, 1,606 controls	E	No significant difference in risk
1986	Brinton	Case-control	1,960 cases, 2,258 controls	E	No difference in risk associated with ever-use; 20 or more years of use increased risk 50%
1989	Adami	Cohort	23,244	E	Increases risk 10%	Average follow-up 6.7 years
1989	Bergkvist	Cohort	23,244	E, E + P	E alone increases risk 70% with 9 years of use; larger but not significant increase in risk with E + P	Average follow-up 5.7 years; majority of participants used E alone
1989	Mills	Cohort	20,341	E, E + P	Current use increases risk 69%	No distinction between E and E + P; no strong increase in risk with increasing duration of use
1991	Kaufman	Case-control	1,686 cases, 2077 controls	E	No significant difference in risk
1991	Toronto Breast Cancer Study, Palmer	Case-control	607 cases, 1,214 controls	E, E + P	No significant difference in risk	93% of participants used E alone
1992	Sillero-Arenas	Meta-analysis of 23 studies	23 case-control, 13 cohort, 1 randomized	E, E + P	E alone increases risk by 8%; No increased risk with E + P
1994	Risch	Cohort	32,790	E, E + P	E alone increases risk 7%/year of use; no significant difference in risk with E + P
1994	Schairer	Cohort	49,017	E, E + P	No significant difference in overall breast cancer risk with either E alone or E + P
1995	Nurses' Health Study, Colditz	Cohort	69,586	E, E + P	E alone increases risk 32%; E + P increases risk 41%	Average follow-up 10.4 years
1995	La Vecchia	Case-control	2,569 cases, 2,588 controls	E, E + P	No change in risk associated with ever-use of either E alone or E + P; increases risk 100% among former users who discontinued use within last 10 years; no increase in risk among those who discontinued use >10 years ago	No distinction made between E and E + P for time-dependent analyses
1995	Newcomb	Case-control	3,130 cases, 3,698 controls	E, E + P	No significant difference in risk for either E alone or E + P
1995	Stanford	Case-control	537 cases, 492 controls	E + P	No significant difference in risk
1996	Persson	Cohort	22,597	E, E + P	E + P increased risk 40% after 10 years follow-up	No effect seen with estrogen alone; 13 years of follow-up
1996	Willis	Cohort	422,373	E, E + P	Decreases breast cancer mortality 16%	9 years of follow-up; No distinction between E vs. E + P
1997	Collaborative Group on Hormonal Factors in Breast Cancer	Meta-analysis	52,705 cases 108,411controls	E, E + P	Incidence of breast cancer increases with increasing duration of hormone use	Where data was available for type of hormone use (39% of women), on significant difference in risk between E and E + P; effect nearly disappears 5 years after stopping
1997	Sellers	Cohort	41,837	E, E + P	No increased risk among women with family history of breast cancer	No distinction between E vs. E + P
1997	Tavani	Case-control	5,984 cases, 5,504 controls	E, E + P	Increased risk 20%	No distinction between E vs. E + P
1998	Sourander	Cohort	7,944	E	No significant difference in risk	8 years of follow-up
1999	Lando	Cohort	5,761	E, E + P	No significant difference in risk	No distinction between E vs. E + P; 22 years of follow-up
1999	Magnusson	Case-control	3,345 cases, 3,454 controls	E, E + P	10 or more years of use E alone increases risk 170%; 10 or more years of use E + P increases risk 195%	Similar results for former and current users
1999	Schairer	Cohort	2,614	E	Reduces mortality when continued after cancer diagnosis	Mortality results determined 14-22 years post-diagnosis
2000	Ross	Case-control	3,534	E, E + P	Increases risk approximately 10% for each 5 years of use	Greater increase in risk among E + P users than E alone
2000	Pike	Case-control	3,794 cases, 3,343 controls	E, E + P	No increased risk with E alone; E + P increases risk 24% per 5 years of use
2000	Schairer	Cohort	46,355	E, E + P	Increases risk 1% to 8% for each year of use	Greater increase in risk among E + P users than E alone
2000	Nurses' Health Study, Colditz	Cohort	58,520	E, E + P	Five or more years of use increases risk 40%	Greater increase in risk among E + P users than E alone; increased risk disappears after 2-5 years discontinuation of use
2001	O'Meara	Cohort	2,755	E, E + P	Hormone reduces recurrence and mortality when continued after cancer diagnosis	79% of hormone users used E alone; results determined 2-19 years after diagnosis
2002	Women's Health Initiative	Randomized	16,608	E + P	Increases risk 26%	5.2 years of follow-up