Reciprocal associations between trajectories of physical activity and physical function among older women: the Australian Longitudinal Study on Women's Health

Key points

1. 01Activity and function form a reciprocal loop: more activity predicted better function 3 years later, and better function predicted more activity later.
2. 02Highly active women kept high function for years; they did not drop to the low-activity group's starting function level (53.7 at age 73) until age 87 (52.9).
3. 03Low-activity women had function scores below the disability threshold (67.3) throughout the whole study.
4. 04After age 80 the rate of function decline was fairly similar across groups, so banking high function earlier through activity matters most.
5. 05Three distinct trajectories (Low, Moderate, High) emerged for both activity and function across ages 73 to 90.

How it was conducted

Design

Prospective longitudinal cohort (Australian Longitudinal Study on Women's Health), 6 surveys at 3-year intervals from 1999 to 2011

Participants

8,238 women from the 1921 to 1926 birth cohort, aged 73 to 78 at baseline (1999) followed to ages 85 to 90

Measures

Physical activity by the Active Australia Survey (MET.minutes/week); physical function by the SF-36 PF subscale (0 to 100)

Analysis

Group-based trajectory modeling, mixed-effects models, and restricted cubic splines with a 3-year time lag; 20-dataset multiple imputation for missing data

What they found

• Three physical activity trajectories were identified: Low PA (26.0% of participants), Moderate PA (58.3%), and High PA (15.7%).
• Three physical function trajectories were identified: Low PF (29.2%), Moderate PF (39.3%), and High PF (31.5%).
• High PA women maintained over 1,600 MET.minutes/week until age 82, then declined to 200 MET.minutes/week by age 90; Moderate PA fell from 533 MET.minutes/week to 0 by age 89; Low PA had a median of 0 at every age.
• High PA women did not reach the Low group's starting function level (53.7 at age 73) until age 87 (52.9).
• High PF group scores stayed high until age 82, then declined to 62.5; the disability threshold was 67.3.
• Restricted cubic splines showed nonlinear dose-response: function rose markedly as activity increased from 0 to 1,000 MET.minutes/week, and activity rose markedly as function increased from 40 to 100.
• The response rate among eligible surviving participants at ages 85 to 90 was 80.9%.

Limitations

• Physical activity and physical function were both self-reported, which can overestimate true levels despite acceptable reliability and validity.
• Healthy survivor bias is likely, since participants were already age 73 at baseline and only those with at least two measurements were included.
• Missing data were substantial (32.2% for activity, 32.6% for function) and handled by imputation rather than complete observed data.
• Findings are observational and limited to older women, so causation cannot be confirmed and results may not generalize to men or younger groups.

Why it matters

For patients

Staying physically active in your 70s helps you stay independent and functioning well for years longer, so building activity into daily life early pays off.

For clinicians

Encourage and sustain physical activity well before the 80s, since function in advanced age depends heavily on the level banked earlier through habitual activity.

For readers

Activity and physical function reinforce each other across older age, so promoting one tends to support the other in a self-sustaining loop.