Patterns and Predictors of Frailty Transitions in Older Men: The Osteoporotic Fractures in Men Study

Journal of the American Geriatrics Society, September 2017. doi:10.1111/jgs.15003

 Presented by: Dr R. Christie


  • Frailty is a geriatric syndrome leading to reduced physiological reserve. Frail patients are therefore more vulnerable to adverse outcomes i.e. anaesthesia and surgery.
  • 2 small scale studies have suggested that frailty is a dynamic process and prevention and reversal may be possible.

Primary Objectives:

  • To determine patterns and probability of frailty progression and improvement.

Secondary Objectives:

  • To Identify predictors of transitions in frailty status

Design and Setting

  • Prospective observational cohort study across 6 US clinical centres
  • Recruited patients from osteoporotic fractures in men study (MrOS) which used mailings, newspaper advertisements and presentations aimed at older populations to recruit individuals
  • 5,086 men received a second visit at 4.6 +/- 0.4 years after recruitment to MrOS. 908 of the 5,994 originally recruited were not included in the study.


  • Men, over 65.
  • 10% from minority groups (African-American, Asian).
  • All lived in the community and were able to walk independently at time of recruitment.


  • At recruitment the cohort completed:
    • A self-administered questionnaire which included factors based on socioeconomic status and ADLs.
    • A mini mental state examination to assess cognitive function.
  • A fifth of the men also had physiological parameters measured.
  • Men were classified as robust, pre-frail or frail based on defined criteria at the start and end of the study period and the transition between groups was compared with data from the questionnaires, MMSE and physiological parameters.


Primary Outcomes:

  • Probability of transition between frailty states between visit 1 and visit 2.

Secondary Outcomes:

  • The odds of progression in frailty status from visit 1 to visit 2 based on baseline characteristics.
  • The odds of improvement in frailty status from visit 1 to visit 2 based on baseline characteristics.


Probability of transition:

  • Robust patients:
    • Robust to Robust – 0.6
    • Robust to Prefrail – 0.32
    • Robust to Frail – 0.03
    • Robust to Death – 0.05
  • Pre frail patients:
    • Prefrail to Robust – 0.15
    • Prefrail to Prefrail – 0.55
    • Prefrail to Frail – 0.17
    • Prefrail to Death – 0.12
  • Frail Patients:
    • Frail to Robust – 0.005
    • Frail to Prefrail – 0.16
    • Frail to Frail – 0.45
    • Frail to Death – 0.28

Characteristics that correlate with progression of frailty:

  • DM
  • Instrumental activity of daily living limitations
  • Smoking
  • CCF
  • Low albumin
  • High interleukin 6

Characteristics that correlate with improvement in frailty:

  • Leg power
  • Marriage


  • Improvement in frailty status is possible and is associated with social, functional and clinical factors.
  • Future studies need to target interventions in frail and pre-frail states such as improving strength and lower limb power, improvement management of co-morbidities and social and nutritional support.


  • Large study
  • Prospective
  • Well written and presented


  • Single sex (male) and predominantly white cohort.
  • Residential/Nursing home residents and those living in assisted living not included. May not be representative of the over 65 population.
  • Only two visits over 4.6 +/- 0.4 year period.
    • Did not identify transitions in between visits nor identify ongoing factors that may have influenced transitions e.g. men who stopped smoking
  • Acute illnesses at time of visits may affect results classification of frailty.
  • Did not state cause of death in the 568 patients that died, and was this related to their frailty status.
  • Did not state why over 700 patients were left out of study.


  • Identified possible targets for intervention that could be further investigated to determine if they improve patient’s frailty scores. These include improving social support networks, smoking and increased physical activity.

Potential for impact

  • Nothing new identified
  • Confirmed what they had hypothesised from data already available
  • This paper is unlikely to drastically change what is already done in in pre-assessment clinics.