Intended for healthcare professionals


Hormone replacement therapy and risk of hip fracture: population based case-control study

BMJ 1998; 316 doi: (Published 20 June 1998) Cite this as: BMJ 1998;316:1858
  1. Karl Michaëlsson, senior registrara (Michaelsson{at},
  2. John A Baron, professorb,
  3. Bahman Y Farahmand, statisticiand,
  4. Olof Johnell, professore,
  5. Cecilia Magnusson, research fellowc,
  6. Per-Gunnar Persson, readerd,
  7. Ingemar Persson, readerc,
  8. Sverker Ljunghall, professor Swedish Hip Fracture Study Groupf
  1. aDepartment of Orthopaedics, University Hospital, S-751 85 Uppsala, Sweden,
  2. bDepartments of Medicine and Community and Family Medicine, Dartmouth Medical School, Hanover, New Hampshire, USA,
  3. cDepartment of Medical Epidemiology, Karolinska Institute Box 281, S-171 77 Stockholm, Sweden,
  4. dDepartment of Epidemiology, Stockholm County Council, 171 76 Stockholm, Sweden,
  5. dDepartment of Orthopaedics, Malmö General Hospital, S-205 02 Malmö, Sweden,
  6. dDepartment of Internal Medicine, University Hospital, S-751 85 Uppsala, Sweden
  1. Correspondence to: Dr Michaëlsson Karl
  • Accepted 12 February 1998


Objective: To determine the relative risk of hip fracture associated with postmenopausal hormone replacement therapy including the effect of duration and recency of treatment, the addition of progestins, route of administration, and dose.

Design: Population based case-control study.

Setting: Six counties in Sweden.

Subjects: 1327 women aged 50-81 years with hip fracture and 3262 randomly selected controls.

Main outcome measure: Use of hormone replacement therapy.

Results: Compared with women who had never used hormone replacement therapy, current users had an odds ratio of 0.35 (95 % confidence interval 0.24 to 0.53) for hip fracture and former users had an odds ratio of 0.76 (0.57 to 1.01). For every year of therapy, the overall risk decreased by 6% (3% to 9%): 4% (1% to 8%) for regimens without progestin and 11% (6% to 16%) for those with progestin. Last use between one and five years previously, with a duration of use more than five years, was associated with an odds ratio of 0.27 (0.08 to 0.94). After five years without hormone replacement therapy the protective effect was substantially diminished (7% to 48%). With current use, an initiation of therapy nine or more years after the menopause gave equally strong reduction in risk for hip fracture as an earlier start. Oestrogen treatment with skin patches gave similar risk estimates as oral regimens.

Conclusions: Recent use of hormone replacement therapy is required for optimum fracture protection, but therapy can be started several years after the menopause. The protective effect increases with duration of use, and an oestrogen-sparing effect is achieved when progestins are included in the regimen.

Key messages

  • Hormone replacement therapy should be continued for long periods for optimal protection of hip fracture

  • No overall substantial hip fracture protection remains after five years without hormone replacement therapy

  • Therapy can be initiated several years after menopause without loss of fracture protection

  • Oral or transdermal therapy are equally effective in reducing the risk of hip fracture

  • The addition of progestins permits lower doses of oestrogens


    • Accepted 12 February 1998
    View Full Text