For example, breast cancer patients who walk briskly for three hours a week after receiving standard treatment have been reported to enjoy 32 percent better outcomes, a success rate few drugs can match, Hicks says.

Other lecture topics include exercise and diabetes, heart disease and brain health. At the beginning of the quarter, Hicks polled the class on how they spend their downtime and created a word cloud to display the results.

The No. Hicks hopes students will be less sedentary by the end of the course. One of the reasons the U. In LASA, an abbreviated version was used, consisting 12 items [ 25 ]. These items cover 3 different aspects: willingness to initiate behavior, persistence when facing adversity, and effort to complete behavior.

Respondents could answer each question on a 5-point scale ranging from 1 strongly disagree to 5 strongly agree. Sum scores ranged from 12 to 60, with higher scores reflecting more self-efficacy. Depressive symptoms were measured using the Center for Epidemiologic Studies Depression Scale CES-D [ 26 ].

The CES-D is a item self-report scale ranging from 0 to 60, with higher scores reflecting more depressive symptoms. Anxiety was measured with the Hospital Anxiety and Depression Scale-Anxiety HADS-A [ 27 ]. The HADS-A is a 7-item self-report questionnaire. The scale ranges from 0 to 21, with higher scores indicating higher levels of anxiety.

Cognitive functioning was measured using the Mini-Mental State Examination MMSE [ 28 ]. The MMSE consists of 23 items and scores range from 0 to 30, with higher scores reflecting better cognitive functioning. Grip strength was measured using a JAMAR J1 Hydraulic Hand Dynamometer.

Respondents were instructed to perform 2 maximum grip strength trials with each hand. Blood pressure was measured twice, with significant time between the measurements, using an automatic Omron device Omron HEM F.

All measurements were performed at the upper left arm. When this was not possible, the right arm was used. Respondents were not allowed to smoke, eat, or be physically active during the last hour before the measurement. Height and weight were also measured during the visit. Respondents were asked about their medication use.

To measure hearing problems, respondents were asked whether they could follow a conversation in a group of 3 or 4 persons with and without a hearing aid, and whether they could follow a conversation with 1 person with and without a hearing aid.

Response categories were 1 yes, without difficulty, 2 yes, but with some difficulty, 3 yes, but with much difficulty, and 4 no, I cannot.

Respondents were categorized as having hearing problems if they had at least some difficulty with more than one of these items. Response categories and categorization were the same as for hearing problems. The number of chronic diseases was measured by self-reports of the following 7 chronic diseases: chronic nonspecific lung disease, cardiovascular diseases, peripheral artery diseases, diabetes mellitus, stroke, arthritis, and malignancies.

Five items measuring pain were included in the self-administered questionnaire: I am in pain when I am standing, I find it painful to change position, I am in pain when I am sitting, I am in pain when I walk, and I am in constant pain [ 29 ].

Response categories were 1 no and 2 yes. Sum scores were calculated ranging from 5 no symptoms of pain to 10 5 symptoms of pain. We used the multivariate imputation by chained equations package in R statistical software [ 30 ] to multiply impute missing values in the predictor variables.

Missingness at random was assumed. We used logistic regression analyses to first examine univariable associations between the predictors and the outcomes. Because prediction models generally perform better in the sample used to develop the model than in an external sample, shrinkage factors can be used to correct for this optimism [ 33 ].

If the calibration slope of a model is lower than 1, this reflects overfitting and it can be interpreted as reflecting the need for shrinkage of the coefficients.

First, bootstrap samples were drawn from each imputed dataset before results were combined [ 34 ]. The AUC was corrected for optimism in each imputed dataset, and a pooled estimate was presented.

All analyses were performed in R version 3. In reporting our prediction models, we followed the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis TRIPOD statement [ 36 ].

Table 1 shows the characteristics of the sample. In our sample, When we stratified by age-group, we found a prevalence of In Table 2 , the estimates of the univariable regression analyses can be found, as well as the optimism adjusted estimates of the final models. The final model predicting vitality included: age, sex, alcohol use, received emotional support, depressive symptoms, hearing problems, number of chronic diseases, and SRH.

The calibration slope of this model was 0. The optimism adjusted AUC was 0. We found that age, depressive symptoms, number of chronic diseases, and SRH were also predictors in the final model predicting frailty.

In addition, cognitive functioning, polypharmacy, and pain remained in the final prediction model. The calibration slope for this model was 0. The frailty model also had a good discriminative value, with an optimism adjusted AUC of 0. To ensure baseline frailty and vitality status did not affect our models, we conducted sensitivity analyses in which we adjusted the prediction models for baseline frailty and vitality status.

These analyses yielded similar prediction models. The number of older adults aged 75 years and over will strongly increase in the coming decades. This particular age-group is not only more at risk of frailty but also seems to be at higher risk of adverse outcomes due to frailty [ 13, 14 ].

Identifying older adults at risk of frailty or vitality could guide health and social care professionals in appropriate use of health and social care resources and implementation of person-oriented preventive strategies.

It has been suggested that early intervention initiatives would be best situated within a setting where older people feel comfortable, for example, in their own homes, at their GP practice, or at facilities they visit regularly [ 37 ]. Therefore, it is important to identify a set of predictors that can be easily measured in such a setting.

The prevalence of frailty at follow-up in our sample was Frailty prevalence rates vary greatly between studies, depending on the countries where the studies were conducted, the age of the sample, and measurements used [ 38 ].

Studies using the FI, which is a multidimensional frailty measure, generally report higher prevalence compared to studies using the physical frailty phenotype Fried criteria , another widely used frailty construct. The prevalence found in our study is consistent with another Dutch study estimating the prevalence of frailty between The prevalence of frailty increases substantially with age, explaining the rather high prevalence of frailty in our sample with a mean age of The prevalence of vitality was This means that almost one-fifth of older adults aged 75 years and over did not experience a decrease in physiological reserve in multiple domains of functioning.

This finding emphasizes the heterogeneity in older adults and the need to not only focus on frailty but to consider the entire continuum from fit to frail.

Almost all of our 33 candidate predictors were univariably associated with frailty and vitality. After backward stepwise selection, prediction models for both frailty and vitality included age, depressive symptoms, number of chronic diseases, and SRH.

We also found differences between the models for vitality and frailty. Male sex, moderate alcohol use, more emotional support received, and no hearing problems were all predictors of being vital after 3 years. Lower cognitive functioning, polypharmacy, and having more pain were predictors of frailty after 3 years.

These findings are supported by earlier research examining factors associated with frailty [ 2, 9, 10 ] and vitality [ 8 ]. To our knowledge, we are the first to develop and validate prediction models for frailty and vitality in European older adults using such a broad set of predictors.

Our study has some limitations. First, we were not able to externally validate the prediction models. We did internally validate the models using bootstrapping techniques, but because bootstrap samples are derived from the same dataset, this only partially solves the problem of optimism.

Therefore, external validation is recommended in future research. Second, the low percentage of vital older adults in our sample led to a low events-per-variable EPV. While an EPV of 10 is generally recommended, a recent study concluded that the evidence for this criterion is weak [ 40 ], suggesting that a violation of this EPV recommendation may not necessarily lead to biased results.

As our prediction models were stable, we expect that exceeding the recommended EPV did not influence the performance of our models. Third, we operationalized vitality as having less than 5 out of 32 deficits, using the LASA-FI. Thus, we only considered the absence of deficits rather than the presence of positive traits.

While the use of an FI to identify both frail and vital older adults has practical advantages, further research is needed to assess whether vitality operationalized in this way is indeed associated with a higher risk of positive outcomes and a lower risk of adverse outcomes.

While for frailty the FI methodology considers each item in the FI as equal, it is possible that for vitality the absence of some deficits may be more important than others. For example, to remain vital, the absence of mobility limitations may be more important than the absence of a disease that can be easily controlled with medication.

Therefore, in future studies, other cutoffs as well as the option of weighting items should be examined when using the FI to measure vitality. Finally, we included predictors that were also included in the FI: chronic diseases, several items of the CES-D depressive symptoms , physical activity, memory complaints, and several items of the MMSE cognitive functioning.

However, previous studies have shown that this is not problematic, since the FI is largely indifferent to its underlying items [ 19 ].

It is the accumulation of deficits that matters, rather than the presence of specific combinations of deficits. Strengths of our study include the use of a representative sample of Dutch older adults, inclusion of predictors from multiple domains, and internal validation of our models.

Also, the predictors included in our study can all be measured relatively easily in different settings, and no laboratory tests are needed.

In addition, we focused not only on frailty but also on vitality, which so far has been largely neglected in research on older adults. In conclusion, it is important to not only focus on frailty in older adults but also on the other end of the spectrum, that is, vitality, to take into account the heterogeneity in older adults.

Our study shows that although predictors of vitality and frailty partially overlap, there are indeed differences in the final prediction models, suggesting conceptual differences between vitality and frailty.

We found that predictors of frailty and vitality encompassed several domains, with predictors in the physical domain being most dominantly present.

The readily accessible predictors in our models may help to easily identify older adults who are likely to be vital and who are at risk of frailty. This study received approval by the Medical Ethics Committee of the VU University medical center.

Signed informed consent was obtained from all study participants. This study was supported by a grant from the SeW Foundation. The Longitudinal Aging Study Amsterdam is largely supported by a grant from the Netherlands Ministry of Health, Welfare and Sports, Directorate of Long-Term Care. Emiel O.

Sascha de Breij and Emiel O. Hoogendijk were responsible for the design of the study. Sascha de Breij analyzed and interpreted the data and drafted the manuscript.

All authors critically revised the research design and the manuscript. All authors approved the final version of the manuscript. Sign In or Create an Account.

Search Dropdown Menu. header search search input Search input auto suggest. Many societies offer single sign-on between the society website and Oxford Academic. If you do not have a society account or have forgotten your username or password, please contact your society.

Some societies use Oxford Academic personal accounts to provide access to their members. See below. A personal account can be used to get email alerts, save searches, purchase content, and activate subscriptions. Oxford Academic is home to a wide variety of products. The institutional subscription may not cover the content that you are trying to access.

If you believe you should have access to that content, please contact your librarian. For librarians and administrators, your personal account also provides access to institutional account management. Here you will find options to view and activate subscriptions, manage institutional settings and access options, access usage statistics, and more.

To purchase short-term access, please sign in to your personal account above. Don't already have a personal account? Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide.

Sign In or Create an Account. Navbar Search Filter Journal of Gerontology This issue GSA Journals Geriatric Medicine Gerontology and Ageing Books Journals Oxford Academic Mobile Enter search term Search.

Issues Journal of Gerontology The Journals of Gerontology, Series A present The Journals of Gerontology, Series B present Editor's Choice About About Journal of Gerontology About The Gerontological Society of America Advertising and Corporate Services Journals Career Network Contact Us Journals on Oxford Academic Books on Oxford Academic.

GSA Journals. Issues Journal of Gerontology The Journals of Gerontology, Series A present The Journals of Gerontology, Series B present Editor's Choice About About Journal of Gerontology About The Gerontological Society of America Advertising and Corporate Services Journals Career Network Contact Us Close Navbar Search Filter Journal of Gerontology This issue GSA Journals Geriatric Medicine Gerontology and Ageing Books Journals Oxford Academic Enter search term Search.

Advanced Search. Search Menu. Article Navigation. Close mobile search navigation Article Navigation. Volume Journal Article. Productive Aging: Enhancing Vitality in Later Life Get access.

Select Format Select format. ris Mendeley, Papers, Zotero. enw EndNote. bibtex BibTex. txt Medlars, RefWorks Download citation.