Construct and Predictive Validity of a Self-Reported Measure of Preclinical Mobility Limitation

Article Outline

• Abstract
• Methods
  • Design
  • Participants
  • Measurements
    • Background characteristics
    • Maximal walking speed
    • Leg extensor muscle power
    • Preclinical and manifest mobility limitation
    • Subgroup division
    • Reliability of mobility assessment
  • Statistical Analysis
• Results
  • Validation Against Criterion Standard
  • Predictive Validity
• Discussion
  • Study Limitations
• Conclusions
• References
• Copyright

Abstract 

Mänty M, Heinonen A, Leinonen R, Törmäkangas T, Sakari-Rantala R, Hirvensalo M, von Bonsdorff MB, Rantanen T. Construct and predictive validity of a self-reported measure of preclinical mobility limitation.

Objectives

To validate self-reported preclinical mobility limitation concept and self-report assessment method against muscle power and walking speed, and to study the predictive validity of preclinical mobility limitation with respect to future risk of manifest mobility limitation.

Design

Observational prospective cohort study and cross-sectional analysis.

Setting

Research laboratory and community.

Participants

A total of 632 community-living (age range, 75−81y) women and men took part in the baseline assessments and 302 persons in the semi-annual interviews on mobility limitation over 2 years.

Interventions

Not applicable.

Main Outcome Measures

Walking speed, muscle power, and self-reported preclinical and manifest mobility limitation. Preclinical mobility limitation was defined as self-reported tiredness or modification of task performance without task difficulty. At baseline, 4 subgroups were created according to self-reported preclinical mobility limitation in any of 3 mobility tasks (walking 2km, walking 0.5km, climbing up stairs): no limitation, preclinical limitation, and minor and major manifest limitation.

Results

At baseline, participants with preclinical mobility limitation showed intermediate levels of walking speed and muscle power, compared with those with no limitation or manifest mobility limitation. Participants reporting baseline preclinical mobility limitation had 3- to 6-fold higher age- and sex-adjusted risk of progressing to major manifest mobility limitation during the 2-year follow-up compared with participants with no limitation at baseline, whereas the risk among those with minor limitation at baseline was 14- to 18-fold higher compared with those with no limitation.

Conclusions

The self-report assessment tool proved to be a valid measure to capture the early signs of disability and may serve as an inexpensive tool for identifying those nondisabled persons at high risk for future disability.

Key Words: Aging, Disabled persons, Muscles, Rehabilitation, Walking

MOBILITY DIFFICULTIES ARE common among older people, increasing the risk of further disability and development of dependency, and are an important public health concern. Difficulties in mobility are often the first noticeable signs of decline in functional ability.¹, ², ³

In the perspective of primary prevention, it is important to identify people who are as yet nondisabled but are at high risk for future functional decline by characterizing an early functional state associated with increased risk for subsequent disability.⁴ Earlier findings have shown that poor performance in objective measures of physical performance—such as lower-extremity muscle strength and walking speed—are highly predictive of subsequent disability⁵, ⁶, ⁷, ⁸ and dependency¹, ⁹ among nondisabled older adults and thus may be considered as a criterion standard in terms of impairment level assessment relevant in the disablement process. In certain circumstances, however, we also need valid self-report measures, because they do not require a specific place or equipment, and because they may be administered, for example, through phone interview or questionnaire. Most existing self-report instruments primarily assess difficulty, inability, or degree of assistance required to perform specific tasks of mobility, household management or personal care. Thus, these measures may not be sensitive enough to recognize early steps in the course of disability.¹⁰, ¹¹ In early stages of functional decline prior to the onset of task difficulty, older persons may be able to compensate for underlying disease by modifying their task performance and thereby maintain their function without the perception of difficulty. This stage of functional decline, that is, changes in method, frequency, or time used in task performance¹², ¹³, ¹⁴ or increased tiredness,¹⁵, ¹⁶ has been proposed as preclinical disability. There is some evidence that persons with preclinical disability have an increased risk for disability.¹², ¹⁴

The aim of this study was to validate the self-reported preclinical mobility limitation concept and assessment tool among Finnish older population by investigating (1) whether self-reported modification of task performance or increased tiredness without task difficulty, that is, preclinical mobility limitation, is associated with decrements in objective measures of muscle power and maximal walking speed and (2) whether preclinical mobility limitation is predictive of occurrence of major difficulties in performing the task, that is, manifest mobility limitation.

Back to Article Outline

Methods 

Design 

This observational study is based on analyses of baseline data and 2-year follow-up on the development of mobility limitation. In the cross-sectional analyses, the validation of preclinical mobility limitation assessment tool was done against the criterion standard, that is, muscle power and walking speed, using the baseline data of the 2-year randomized controlled trial Screening and Counseling for Physical Activity and Mobility in Older People (SCAMOB)¹⁷ (ISRCTN 07330512). The predictive validity of preclinical mobility limitation was studied as risk for developing future manifest mobility limitation in the SCAMOB control group, where the naturally occurring changes in function took place. Maximal walking speed and muscle power examinations and face-to-face interview on mobility were performed at baseline. In addition, telephone interviews about mobility were carried out 3 times at 6-month intervals after the baseline and face-to-face interview again at 2-year follow-up point.

Participants 

The original target population consisted of all 75- to 81-year-old registered residents of the City of Jyväskylä, Central Finland, living in the city center area in March 2003 (N=1310). Subjects’ selection procedure has been described in detail elsewhere.¹⁷ To be eligible to this study, persons had to be able to walk 500m without assistance, have Mini-Mental State Examination (MMSE)¹⁸ score greater than 21, have no medical contraindications for physical activity, and be only moderately physically active or sedentary. Persons with specific physical activities were excluded from the study. After screening we had 632 participants in the cross-sectional analysis and 314 participants in the SCAMOB control group for the follow-up of the naturally occurring changes in function. Participants without the outcome of major manifest limitation at the baseline were followed up for the development of mobility limitation in 2-km walk (n=266), 0.5-km walk (n=302), and climbing up stairs (n=295). The study was approved by the Ethics Committee of the Central Finland Health Care District. All participants gave their written informed consent before the study.

Measurements 

Information on self-reported physician-diagnosed chronic conditions lasting over 3 months and prescription medication was collected during the face-to-face interviews at the participant’s home. During the study-center visit, the study nurse checked the responses concerning chronic diseases and prescription medications filled in by the interviewers at home interviews. Depression was measured with Center for Epidemiologic Studies Depression Scale (CES-D)¹⁹ and cognitive impairment was assessed with the MMSE.¹⁸

We measured walking speed over 10m in the study-center corridor. Participants were allowed 2- to 3-m acceleration before the start line and they were encouraged to walk as fast as possible without risking their health. Timing was done using a stopwatch. Participants wore walking shoes or sneakers, and use of a walking aid was allowed if needed. Walking speed has previously been shown to be a reliable and valid measurement of functional performance among older people.⁶ The test-retest coefficient of variation (CV) for maximal walking speed in our study center was 4.6%.²⁰

At baseline, we measured leg extensor muscle power of a single leg using Nottingham Leg Extensor Power Rig.^21,a During testing, the participant was seated with arms folded, 1 foot was placed on the pedal attached to a flywheel, and the other foot rested on the floor. After 2 to 3 practice trials, each participant was asked to push the pedal as hard and fast as possible. Five to 9 maximal efforts per leg, separated by 30-second rests, were conducted.²², ²³ In this study, the best performance of dominant leg was used as the measure of maximal power. Muscle power measurement with the Nottingham power rig has been validated and found to be safe and acceptable among older people.²¹ The test-retest CV in our study center was 8%.²²

The questions on mobility were formulated according to the hypothesis that the progression of major manifest mobility limitation develops through stages of preclinical mobility limitation and minor manifest mobility limitation (fig 1). Self-reported mobility was studied using a structured interview regarding walking 2.0km, walking 0.5km, and climbing up 1 flight of stairs. The questions were formulated as follows: “Do you have difficulty in …” and 5 alternative response options were given: (1) able to manage without difficulty, (2) able to manage with some difficulty, (3) able to manage with great deal of difficulty, (4) able to manage only with help of another person, and (5) unable to manage even with help. To identify persons at an early stage of mobility limitation, that is, preclinical mobility limitation, additional questions were posed to participants who reported no task difficulty. The questions concerned the modification of task performance and the alternatives given were resting in the middle of the performance, using an aid, taking support from handrails, having reduced the frequency of performing the task, having slowed down performance of the task, experiencing tiredness when performing the task, or some other change in carrying out the task.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Subgroup division according to the level of mobility limitation and its association to the hypothesized development of mobility limitation.

At baseline 4 subgroups for each mobility task were created according to self-reported mobility difficulties and task modification: (1) no limitation (no mobility difficulties and no modification), (2) preclinical limitation (no mobility difficulty and ≥1 modification), (3) minor manifest limitation (some difficulty), and (4) major manifest limitation (great deal of difficulty, need for help of another person or not able to manage even with help) (see fig 1).

We assessed test-retest reliability for the mobility and preclinical limitation questions among a sample of 29 people of similar age by replicating the questions in an interval of 2 weeks. The agreement of self-reports between 2 different days for the question on the 2-km walk was 93%; for the 0.5-km walk, 97%; and for climbing up 1 flight of stairs, 100%. The task modification questions agreement ranged from 93% to 100% for the 2-km walk, from 97% to 100% for the 0.5-km walk, and from 72% to 100% for walking up 1 flight of stairs.

Statistical Analysis 

Means and standard deviations (SDs) were used as descriptive statistics. Analyses of covariance were used to estimate the mean differences in muscle power and maximal walking speed between the groups based on hypothesized progression of mobility limitation. Age, sex, and number of long-term diseases and prescription medication were used as covariates. The relative differences between the study groups were determined using log transformation of the variables. When the 95% confidence intervals (CIs) did not include zero, the differences were regarded as statistically significant at α equal to .05. These analyses were performed using Mplus software.^b

In the 2-year follow-up, we constructed a generalized estimating equations model²⁴ on the dichotomized (no major manifest limitation, major manifest limitation) mobility variables to test the significance of subgroup differences in the risk for major manifest limitation over time. Subjects who had major manifest limitation with mobility task at baseline were excluded from the analyses of the given task. The analyses were performed in SAS,^c using the GENMOD procedure. To study factors that may underlie on the theoretic pathway to mobility limitation, 3 separate models were constructed for each mobility task using sex, age, presence of chronic diseases known to affect mobility (osteoarthritis, rheumatoid arthritis, diabetes, chronic obstructive pulmonary disease, ischemic heart disease, myocardial insufficiency, sciatica), depressive symptoms (CES-D score), weight, height, walking speed, and muscle power as covariates. For cases with missing values in a mobility task at some point of the 2-year follow-up, data were imputed with the multiple imputation procedure implemented in SAS by using information on the other mobility tasks and baseline information such as number of long-term diseases, and MMSE and CES-D scores. Sensitivity analyses performed suggested no significant differences in effects due to imputation. The number of missing observations at different measurement points ranged from 9 to 28. Subjects who died over the course of the study were censored at the date of their death and missing values were not imputed. During the 2-year follow-up, 8 participants died.

Back to Article Outline

Results 

The mean age of our study population (N=632) was 77.6±1.9 years and 75% were women. At the baseline all the participants (N=632) completed the face-to-face interview on mobility and 629 participants performed the maximal walking speed test and 614 participants the muscle power measurement. In our follow-up sample, attrition over 2 years was under 10%.

A substantial number of the participants who reported no difficulty in the 2-km walk, 0.5-km walk, or climbing up stairs reported task modification or increased tiredness in performing the task. Depending on the task being assessed, altogether 31% to 55% of the participants were categorized into the preclinical mobility limitation subgroup. The distribution of participants according to our subgroup division is given in detail in table 1.

Table 1. Frequency of Participants According to the Level of Mobility Limitation at Baseline for the 2-km Walk, 0.5-km Walk, and Climbing Up Stairs (N=632)

Activity	No Difficulty		Difficulty
	No Limitation	Preclinical Limitation	Minor Manifest Limitation	Major Manifest Limitation
2-km walk	167(26)	257(41)	118(19)	90(14)
0.5-km walk	347(55)	194(31)	70(11)	21(3)
Climb up stairs	132(21)	346(55)	111(18)	43(7)

NOTE. Values are n (%).

Baseline characteristics according to self-reported mobility in the 2-km walk are shown in table 2. According to the number of long-term diseases and prescription medication, length of education, body mass index, depressive symptoms, muscle power, and walking speed, participants with preclinical mobility limitation were an intermediate group between the participants reporting no limitation and those reporting manifest limitation. Similar results were observed also when comparison was done based on the 0.5-km walk or stair climbing.

Table 2. Baseline Characteristics (N=632) According to the Level of Limitation in Walking 2km

NOTE. Values are mean ± SD.

Validation Against Criterion Standard 

We validated the self-report assessment tool against objectively measured maximal walking speed and muscle power. Adjusted mean percentage differences between the subgroups in maximal walking speed and muscle power compared with preclinical limitation subgroup are given in figure 2. The participants with preclinical mobility limitation had intermediate levels of maximal walking speed and muscle power between participants with no limitation and those with manifest limitation. This result was consistent in all 3 mobility tasks, being most evident in the 2-km walk and climbing up 1 flight of stairs (see fig 2). The participants reporting no limitation had significantly (P<.05) faster walking speed (11%−20%) and higher muscle power (14%−16%), compared with those with preclinical mobility limitation. In proportion, the participants reporting either minor or major manifest limitation had significantly (P<.05) slower walking speed (10%−28%), and lower muscle power (13%−26%), compared with participants with preclinical limitation.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Adjusted mean differences (in percent) and their 95% CI in (A) maximal walking speed and (B) lower-extremity muscle power compared with people with preclinical mobility limitation (0 level).

Predictive Validity 

The prevalence of major manifest limitation in the 2-km walk, 0.5-km walk, and climbing up 1 flight of stairs during the 24-month follow-up are shown in figure 3. The risk for the development of major manifest mobility limitation was intermediate among participants with preclinical mobility limitation compared with those with no limitation or minor manifest limitation. Table 3 shows the risk ratios for the onset of task-specific major manifest limitation in each mobility task compared with the subjects with no limitation. In the first model, adjusted for sex and age, preclinical limitation in the 2-km walk and 0.5-km walk increased the risk for major manifest limitation almost 3- to 6-fold compared with participants with no limitation. The trend was similar also in climbing up stairs, only with nonsignificant risk ratio. Subjects with minor manifest limitation had 14- to 18-fold risk for major manifest limitation in each mobility task compared with those with no limitation. In the second and third model, we adjusted the risk ratios with potential confounders together with lower-extremity muscle power and maximal walking speed. These procedures attenuated the risk ratios in all of the 3 tasks (table 3).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Unadjusted prevalence (in percent) of major manifest limitation for the (A) 2-km walk, (B) 0.5-km walk, and (C) climbing up 1 flight of stairs during 24-month follow-up. NOTE. At baseline, the number of participants with preclinical mobility limitation in 2km was 140; in 0.5-km walk, 100; and in climbing up stairs, 178. Respectively, the numbers for participants with no limitation was 74, 166, and 59, and for participants with minor manifest limitation 52, 36, and 58.

Table 3. Risk Ratios and Their 95% CIs for the Onset of Major Manifest Limitation for the 2-km Walk, 0.5-km Walk, and Climbing Up 1 Flight of Stairs Among Participants With Preclinical or Minor Manifest Mobility Limitation, Compared With Participants With No Limitation at Baseline

Activity	Model I⁎		Model II†		Model III‡
	RR	95% CI	RR	95% CI	RR	95% CI
2-km walk
Preclinical limitation	5.8	2.6–12.9	5.0	2.2–11.2	2.9	1.2–6.6
Minor manifest limitation	17.8	7.6–41.9	17.1	7.0–41.5	8.9	3.6–21.6
0.5-km walk
Preclinical limitation	2.5	1.2–5.1	2.3	1.1–4.6	1.4	0.7–2.9
Minor manifest limitation	13.2	6.1–28.4	10.3	4.8–22.0	5.4	2.3–12.2
Climb up 1 flight of stairs
Preclinical limitation	2.3	0.7–7.4	1.8	0.5–6.1	1.2	0.4–3.9
Minor manifest limitation	13.9	4.2–45.7	10.2	2.8–37.5	5.4	1.6–19.1

Abbreviation: RR, risk ratio.

Back to Article Outline

Discussion 

Our results indicate that self-reported preclinical mobility limitation is a valid measure to identify persons at high risk for future manifest mobility limitation. In our cross-sectional part of the study, objective measures of maximal walking speed and muscle power were associated in a stepwise relationship with the 4 categories of mobility limitation, ranging from no limitation to preclinical limitation and further to minor and major manifest limitation. Participants reporting preclinical mobility limitation had intermediate walking speed and muscle power between those with no limitation and those with minor or major manifest limitation. In addition, our observational follow-up study confirmed that the self-reported preclinical mobility limitation predicted development of major manifest limitation in the near future.

Our findings are in accordance with the previous findings of Fried,¹², ¹³ Wolinsky,¹⁴ and Avlund¹⁵, ¹⁶, ²⁵ and colleagues providing criterion and predictive validity for the preclinical mobility limitation concept. In the earlier studies of self-reported preclinical mobility limitation, the target population has consisted of 70- to 80-year-old high-functioning American women¹², ¹³ or late-middle-aged African Americans.¹⁴ Our study population represented sedentary community-living older people. In addition, to our knowledge, muscle power or self-reported ability to walk 2km has not been used as outcome in the earlier studies of preclinical disability.

In the work of Fried¹², ¹³ and Wolinsky,¹⁴ preclinical mobility limitation was assessed by self-reported changes in the method or decreased frequency of task performance without perceived task difficulty. This approach suggests that people are able to compensate for underlying disease by modifying their task performance. This is in line with the theory of selective optimization with compensation by Baltes and Baltes,²⁶ which has previously been mostly applied to studies on cognitive functioning. Baltes and Baltes²⁶ described 3 adaptational processes: selection denotes a restriction of involvement in activities in response to lost capacity, optimization refers to optimization processes like pacing activities to enable one to continue performing activities, and compensation involves efforts to meet goals by new means (eg, modifying behaviors, using assistive devices). In the recent years, this theory has been applied to physical functioning, for example among older adults with osteoarthritis,²⁷ showing variability and plasticity in older adults’ efforts to manage physical disability by selection, optimization, and compensation. This is highly relevant also in the perspective of preclinical mobility limitation concept.

In addition to self-reported changes in the method or decreased frequency of task performance, we included also increased tiredness introduced by Avlund¹⁵, ¹⁶ into the determination of preclinical mobility limitation. Self-rated tiredness in daily activities is related to subsequent functional decline, hospitalization, and use of home help.¹⁵, ¹⁶, ²⁵ Therefore it is reasonable to take tiredness into a definition of preclinical disability. More studies are needed to explore what aspects are relevant indicators of preclinical disability, and thus should be included in to the measurements of preclinical disability, and whether the same determination applies to different groups of older people.

In our follow-up analysis, adjustment for long-term diseases and depressive symptoms attenuated the increased risk for manifest limitation somewhat among participants with preclinical mobility limitation or minor manifest limitation and a further reduction was observed when walking speed, muscle power, height and weight were added into the models. This suggests that these covariates may lie on the theoretic pathway to mobility limitation. During our 2-year follow-up, participants developed major manifest limitation most often in the 2-km walk, indicating that it was the most demanding task among the 3 mobility tasks for our participants. In our cross-sectional data maximal walking speed was more strongly associated with self-reported mobility function compared with muscle power. This was expected, as self-reported mobility limitation and objectively measured walking speed represent the same category—functional limitation at the level of the organism as a whole—in the disablement process.²⁸ Respectively, muscle power represents the impairment level of organs and body systems, being more distal from the disability category.

Study Limitations 

Our participants represented a group of older people who would probably benefit the most from preventive actions, because we excluded both the physically very active and those who were not able to move independently even minimally. However, the use of this truncated distribution might cause some underestimation in our cross-sectional and follow-up results. Presumably, we would have seen stronger associations if we had included also very vigorous and more impaired older people in our study. Nevertheless, our large sample size, population-based sampling, and small dropout rate underline the reliability of our results.

The results of the present study have some implications to future prevention of disability, which are in line with earlier suggestions of Fried,¹², ¹³ Wolinsky,¹⁴ and Avlund¹⁵, ¹⁶ and colleagues. Instead of continuing the disability prevention as secondary or tertiary treatment with people already disabled, the prevention should focus on the early stages of disability. It has been suggested that in terms of maximal effectiveness, interventions for reducing or postponing disability should be targeted for people who are not yet disabled but are at high risk.²⁹, ³⁰ In our study the prevalence of manifest mobility limitation in semiannual follow-up points during the 2-year follow-up was 4% to 23% among participants with preclinical mobility limitation and 28% to 52% among participants with minor manifest limitation. This indicates that people with preclinical mobility limitation are a potential target group for preventive interventions and respectively those with minor manifest limitations might already need some rehabilitative actions. This needs to be further ascertained with intervention studies focusing on preclinical mobility limitation.

Back to Article Outline

Conclusions 

The results of this study indicated that the self-report assessment tool was a valid measure to capture the early signs of disability, that is, preclinical mobility limitation. Characterizing those at preclinical stage of mobility limitation by self-report, may serve as a simple and inexpensive tool for identifying those at high risk for future disability and would probably help to find appropriate target groups for early interventions to prevent disability.

Suppliers

Back to Article Outline

References 

1. Gill TM, Williams CS, Richardson ED, Tinetti ME. Impairments in physical performance and cognitive status as predisposing factors for functional dependence among nondisabled older persons. J Gerontol A Biol Sci Med Sci. 1996;51:M283–M288
   • View In Article
   • MEDLINE
2. Guralnik JM, Simonsick EM, Ferrucci L, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994;49:M85–M94
   • View In Article
   • MEDLINE
3. Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med. 1995;332:556–561
   • View In Article
   • MEDLINE
   • CrossRef
4. Fried LP, Guralnik JM. Disability in older adults: evidence regarding significance, etiology, and risk. J Am Geriatr Soc. 1997;45:92–100
   • View In Article
   • MEDLINE
5. Ostir GV, Markides KS, Black SA, Goodwin JS. Lower body functioning as a predictor of subsequent disability among older Mexican Americans. J Gerontol A Biol Sci Med Sci. 1998;53:M491–M495
   • View In Article
   • MEDLINE
6. Guralnik JM, Ferrucci L, Pieper CF, et al. Lower extremity function and subsequent disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery. J Gerontol A Biol Sci Med Sci. 2000;55:M221–M231
   • View In Article
   • MEDLINE
   • CrossRef
7. Bean JF, Leveille SG, Kiely DK, Bandinelli S, Guralnik JM, Ferrucci L. A comparison of leg power and leg strength within the InCHIANTI study: which influences mobility more?. J Gerontol A Biol Sci Med Sci. 2003;58:728–733
   • View In Article
   • MEDLINE
8. Rantanen T, Avlund K, Suominen H, Schroll M, Frandin K, Pertti E. Muscle strength as a predictor of onset of ADL dependence in people aged 75 years. Aging Clin Exp Res. 2002;14(Suppl 3):10–15
   • View In Article
   • MEDLINE
9. Gill TM, Williams CS, Tinetti ME. Assessing risk for the onset of functional dependence among older adults: the role of physical performance. J Am Geriatr Soc. 1995;43:603–609
   • View In Article
   • MEDLINE
10. Fried LP, Herdman SJ, Kuhn KE, Rubin G, Turano K. Preclinical disability (Hypotheses about the bottom of the iceberg). J Aging Health. 1991;3:285–300
    • View In Article
    • CrossRef
11. Fried LP, Bandeen-Roche K, Williamson JD, et al. Functional decline in older adults: expanding methods of ascertainment. J Gerontol A Biol Sci Med Sci. 1996;51:M206–M214
    • View In Article
    • MEDLINE
12. Fried LP, Bandeen-Roche K, Chaves PH, Johnson BA. Preclinical mobility disability predicts incident mobility disability in older women. J Gerontol A Biol Sci Med Sci. 2000;55:M43–M52
    • View In Article
    • MEDLINE
    • CrossRef
13. Fried LP, Young Y, Rubin G, Bandeen-Roche K WHAS II Collaborative Research Group. Self-reported preclinical disability identifies older women with early declines in performance and early disease. J Clin Epidemiol. 2001;54:889–901
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (567 KB)
    • CrossRef
14. Wolinsky FD, Miller DK, Andresen EM, Malmstrom TK, Miller JP. Further evidence for the importance of subclinical functional limitation and subclinical disability assessment in gerontology and geriatrics. J Gerontol B Psychol Sci Soc Sci. 2005;60:S146–S151
    • View In Article
    • MEDLINE
    • CrossRef
15. Avlund K, Damsgaard MT, Schroll M. Tiredness as determinant of subsequent use of health and social services among nondisabled elderly people. J Aging Health. 2001;13:267–286
    • View In Article
    • MEDLINE
    • CrossRef
16. Avlund K, Schultz-Larsen K, Davidsen M. Tiredness in daily activities at age 70 as a predictor of mortality during the next 10 years. J Clin Epidemiol. 1998;51:323–333
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (354 KB)
    • CrossRef
17. Leinonen R, Heikkinen E, Hirvensalo M, et al. Customer-oriented counseling for physical activity in older people: study protocol and selected baseline results of a randomized-controlled trial (ISRCTN 07330512). Scand J Med Sci Sports. 2007;17:156–164
    • View In Article
    • MEDLINE
18. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–198
    • View In Article
    • Full-Text PDF (635 KB)
    • CrossRef
19. Radloff LS. The CES-D scale: a self-reported depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401
    • View In Article
20. Pajala S, Era P, Koskenvuo M, et al. Contribution of genetic and environmental factors to individual differences in maximal walking speed with and without second task in older women. J Gerontol A Biol Sci Med Sci. 2005;60:1299–1303
    • View In Article
    • MEDLINE
    • CrossRef
21. Bassey EJ, Short AH. A new method for measuring power output in a single leg extension: feasibility, reliability and validity. Eur J Appl Physiol Occup Physiol. 1990;60:385–390
    • View In Article
    • MEDLINE
    • CrossRef
22. Tiainen K, Sipila S, Alen M, et al. Shared genetic and environmental effects on strength and power in older female twins. Med Sci Sports Exerc. 2005;37:72–78
    • View In Article
    • MEDLINE
    • CrossRef
23. Portegijs E, Sipila S, Alen M, et al. Leg extension power asymmetry and mobility limitation in healthy older women. Arch Phys Med Rehabil. 2005;86:1838–1842
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (149 KB)
    • CrossRef
24. Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika. 1986;73:13–22
    • View In Article
    • CrossRef
25. Avlund K, Damsgaard MT, Sakari-Rantala R, Laukkanen P, Schroll M. Tiredness in daily activities among nondisabled old people as determinant of onset of disability. J Clin Epidemiol. 2002;55:965–973
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (94 KB)
    • CrossRef
26. Baltes PB, Baltes MM. Psychological perspectives on successful aging: the model of selective optimization with compensation. In:  Baltes PB,  Baltes MM editor. Successful aging: perspectives from the behavioral sciences. Cambridge: Cambridge Univ Pr; 1990;p. 1–34
    • View In Article
27. Gignac MA, Cott C, Badley EM. Adaptation to disability: applying selective optimization with compensation to the behaviors of older adults with osteoarthritis. Psychol Aging. 2002;17:520–524
    • View In Article
    • MEDLINE
    • CrossRef
28. Nagi SZ. An epidemiology of disability among adults in the United States. Milbank Mem Fund Q Health Soc. 1976;54:439–467
    • View In Article
    • MEDLINE
    • CrossRef
29. Ferrucci L, Guralnik JM, Studenski S, Fried LP, Cutler GB, Walston JD. Interventions on Frailty Working Group (Designing randomized, controlled trials aimed at preventing or delaying functional decline and disability in frail, older persons: a consensus report). J Am Geriatr Soc. 2004;52:625–634
    • View In Article
    • MEDLINE
    • CrossRef
30. Guralnik J, Leveille S, Volpato S, Marx M, Cohen-Mansfield J. Targeting high-risk older adults into exercise programs for disability prevention. J Aging Phys Activity. 2003;11:219–228
    • View In Article

• a University of Nottingham, Medical Faculty Workshops, Queen’s Medical Centre, Nottingham, UK.
• b Version 4.2; Muthén & Muthén, 3463 Stoner Ave, Los Angeles, CA 90066.
• c Version 9.1; SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513-2414.