Table 3 Main results of included studies

Gill et al. (1995) [35]

Each performance test (chair stand, rapid gait, 360° turn, bending over, foot taps, and hand signature) is significantly associated with the onset of functional dependence in ADL disability. Adjusted Risk Ratios (RR) vary from 1.2 (.7-2.0) for foot taps to 2.4 (1.4-4.2) for rapid gait.

Guralnik et al. (1995) [20]

Elderly people with lowest lower extremity function have a higher risk of ADL disability compared to elderly people in higher lower extremity function groups. RR 4.2 (2.3-7.7). Elderly people in the moderate group have a higher risk of ADL disability compared to elderly people in the high group. RR 1.6 (1.0-2.6).

Sonn et al. (1995) [37]

Walking speed and grip strength at age 70 are significantly associated with incident ADL disability at age 76.

Tinetti et al. (1995) [44]

Elderly people with lower usual gait speed, lower rapid gait speed, or lower balance have a higher risk of functional dependence in ADL. OR 2.0 (1.5-2.7), 2.3 (1.7-3.2), and 2.0 (1.5-2.7) respectively.

Gill et al. (1996) [36]

Elderly people in the lowest quartile of physical function (measured by walking, turning, chair stands) have a higher risk of functional dependence in ADL. RR 2.1 (1.4-3.0).

Ostir et al. (1998) [18]

Elderly people in the lowest quartile of walking speed, balance, and chair stands have a higher risk of ADL disability after a 2-year follow-up compared to elderly people in the highest quartile. OR 5.4 (1.2-23.6), OR 2.4 (1.0-5.4), and OR 2.8 (1.2-6.4) respectively.

Giampaoli et al. (1999) [33]

Elderly men with higher hand grip strength have a lower risk of disability compared to men with lower hand grip strength. OR .97 (.96-.99).

Wu et al. (1999) [40]

Elderly people who participated regularly in exercise had a lower risk of becoming chronically ADL disabled after a 3-year follow-up. RR .52 (.39-.68).

Guralnik et al. (2000) [19]

Elderly people with low lower extremity function have a higher risk of ADL disability compared to elderly people with high lower extremity function. RR ranging from 3.4 (1.7-7.1) to 7.4 (1.8-30.5). Elderly people with moderate lower extremity function have a higher risk of ADL disability compared to elderly people with high lower extremity function. RR ranging from 1.2 (.7-2.2) to 2.0 (.7-5.3). Gait speed alone performed almost as well as total lower extremity function in predicting incident disability.

Ishizaki et al. (2000) [30]

Elderly people with higher hand grip strength (1kg) have a lower risk of developing disability in basic ADL within the next 3 years. OR .91 (.84-.97).

Lee (2000) [38]

Elderly people who think that they are less active than other people their age have a higher risk of ADL disability compared to people who think that they are a lot more active than other people their age. OR 1.65 (1.14-2.39).

Sarkisian et al. (2000) [39]

Elderly people in the lowest quintile of gait speed have a higher risk of decline in basic ADL. OR 2.29 (1.66-3.17). Elderly people in the lowest quintile of exercise level also have a higher risk of basic ADL decline. OR 1.47 (1.06-2.05).

Shinkai et al. (2000) [29]

Maximum walking speed, usual walking speed, balance, and grip strength are significant predictors of the onset of functional ADL dependence after a 6-year follow-up in elderly people who are aged 65-74 and 75 or older. For elderly people in the lowest quartile the HR ranged from 2.21 (1.23-3.97) to 6.18 (3.16-12.1).

Stessman et al. (2002) [27]

Elderly people who are not physically active or who do not exercise at least four days a week at age 70 have a higher risk of ADL disability after a 7-year follow- up compared to elderly people who are physically active at age 70. OR for men 4.3 (1.1-17.1), OR for women 8.5 (2.0-36.2).

Wang et al. (2002) [45]

Elderly persons who exercise regularly have a decreased age-adjusted risk of functional decline in ADL.

Shinkai et al. (2003) [28]

Elderly people in the lowest quartile of hand grip strength, balance, usual walking speed or maximal walking speed have a higher risk of disability in basic ADL. HR 1.22 (1.07-1.39), 1.41 (1.22-1.62), 1.31 (1.14-1.50), and 1.40 (1.22-1.61) respectively.

Al Snih et al. (2004) [22]

Men and women in the lowest quartile of hand grip strength have a higher risk of ADL limitations in the next 7 years. HR for men 1.9 (1.14-3.17) and HR for women 2.28 (1.59-3.27).

Gill et al. (2004) [25]

Slow gait speed is associated significantly with the development of insidious disability. OR 2.4 (1.4-4.1).

Al Snih et al. (2005) [21]

Elderly people with weight loss of 5% or more within a 2-year follow-up after baseline have a higher risk of lower body ADL disability compared to elderly people with stable weight. Adjusted OR 1.43 (1.06-1.95).

van den Brink et al. (2005) [34]

Compared to the lowest tertile of total physical activity men from the middle and highest tertile have a lower risk of disability. OR .56 (.32-.99) and OR .50 (.29- .88) respectively.

Onder et al. (2005) [42]

Balance, chair stands, and walking speed were significant predictors of progressive incident ADL disability. Walking speed was also a significant predictor of catastrophic incident disability.

Jacobs et al. (2008) [26]

Elderly people who go out less then daily at age 70 have a higher risk of incident dependence in ADL compared to elderly people who go out daily at age 70. RR 6.9 (1.4-34.0).

Ritchie et al. (2008) [43]

A history of unintentional weight loss at baseline predicts more rapid decline in ADL.

Rosano et al. (2008) [31]

Gait speed is a significant predictor of disability. HR .88 (.80-.96). This HR remains when controlling for age, sex, race, education, and possible confounders.

Rothman et al. (2008) [23]

Slow gait speed, low physical activity and weight loss are significant predictors of chronic incident disability. HR 3.0 (2.3-3.8), HR 2.1 (1.7-2.6), and HR 1.7 (1.4-2.1) respectively. Exhaustion and grip strength do not predict chronic incident disability

Gill et al. (2009) [24]

Poor grip strength was associated with 3 subtypes of disability. OR ranging from 1.42 (1.03-1.95) to 1.80 (1.04-3.12). Lower extremity performance score was significantly associated with 5 subtypes of ADL disability. OR ranging from 1.10 (1.04-1.17) to 1.35 (1.24-1.47).

Arnold et al. (2010) [32]

Elderly people with weight loss of 5% or more between consecutive annual visits have a higher risk of incident ADL disability compared to elderly people with stable weight. Adjusted OR 1.27 (1.10-1.46).

Balzi et al. (2010) [41]

High level of physical activity compared to sedentary state is associated with a lower incidence of ADL disability after a 3-year follow-up. OR .30 (.12-.76). Lower extremity performance score is a significant predictor of disability.