An Early Risk Index: The Original Frailty Measure?

To the Editor: In 1962, Arnold and Exton-Smith devised a risk index to predict and measure outcomes on an early geriatric inpatient unit. This index predicted length of stay and discharge home.1 In many ways, it was a precursor to the contemporary accumulation of deficits model of frailty.2 It has been demonstrated that a simple modification of this index predicts death and institutionalization in community-dwelling older adults in Manitoba.3 This risk index is now considered in a second, larger population-based cohort study. A secondary analysis was conducted of the Canadian Study of Health and Aging (CSHA),4 the methods and measures of which have been described elsewhere. Participants of the Manitoba component of the CSHA (who were included in the previous analysis) and those who had missing data were excluded. The original score1 considered mental status, functional status, and continence. Mental status was measured on a 5-point scale (normal = 4, slight impairment = 3, moderate impairment = 2, gross impairment = 1, coma = 0), incontinence was measured on a 3-point scale (not = 2, moderate = 1, severe = 0), and functional status was measured with several items, each scored on a 3-point scale (activity performed unaided = 2, with help = 1, unable to do it or never did it = 0); the items were walking out of doors, climbing stairs, walking indoors, getting in and out of a chair, getting in and out of bed, feeding, dressing, washing, shaving (men) and attention to hair (women), cooking and cleaning. To replicate the score, items from the CSHA that corresponded to the original index were used. Specifically, the Mini-Mental State Examination (MMSE) was used to measure cognitive status.5 There were no comatose participants. An MMSE score of less than 20 indicated severe impairment; a score of of 20 to 23, moderate impairment; a score of 24 to 26, mild impairment; and a score >26, no impairment. Continence was measured according to the answer to the question: “Do you ever lose control of your bladder?” and was scored yes or no. Functional status was measured using the Older Americans Resource Survey;6 mobility out of doors, mobility indoors, getting in and out of bed, feeding, dressing, grooming, cooking, and cleaning were scored as 0 (unable to do), 1 (needs assistance), or 2 (able to do independently). The total score ranged from 0 to 24. Two items from the original index were missing in the CSHA: climbing stairs and rising from a chair. Outcomes were death or nursing home (NH) placement measured 5 years later. Bivariate analyses were conducted using chi-square tests for categorical variables and Student t-tests (assuming unequal variance) for continuous variables. To adjust for age, sex, and education, multinomial logistic regression models were constructed for the outcomes of alive (reference), NH, and dead (including those who were admitted to a NH and subsequently died) 5 years later. Receiver operator characteristic (ROC) curves were also constructed, and the area under the ROC curve (AUC) was calculated using a c-statistic. Eight thousand five hundred seven participants (59.5% of whom were women) with a mean age of 75.7 ± 7.1 and a mean length of education of 10.1 ± 3.9 years were included in these analyses. The mean MMSE score was 26.4 ± 3.5. Over the 5 years, 2,057 (24.9%) participants died, and 1,010 (11.9%) were admitted to a NH. The risk index predicted both outcomes in a graded manner (Figure 1). In unadjusted multinomial logistic regression models, the odds ratio (OR) for NH placement was 0.68 (95% confidence interval (CI) = 0.65–0.70) and for death was 0.69 (95% CI = 0.67–0.71). In multinomial logistic regression models adjusted for age, sex, and education, the OR for NH placement was 0.76 (95% CI = 0.73–0.79) and for death was 0.75 (95% CI = 0.73–0.77). ROC curves were also constructed, and the AUC for death was 0.69 (95% CI=0.68–0.71) and for NH placement was 0.77 (95% CI=0.75–0.78). A risk index devised in 19621 to predict and measure rehabilitation outcomes on inpatient units also predicts adverse outcomes in community-living older adults in the modern era. This finding underscores the central importance of functional and cognitive status in older adults across settings and historical eras. Use of this score in clinical practice is not advocated—in particular, there is a ceiling effect, with most community-living older adults scoring high on the scale. Nonetheless, the cumulative effect of cognitive and functional loss remains important in the clinical care, administration, and research into the health of older adults. This early risk score also demonstrates the historical roots of some of the current frailty models. Although not defined as frailty, the notion has deep roots in geriatric medicine. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. The Manitoba Study of Health and Aging (MSHA) was funded primarily by Manitoba Health, with additional funding provided through the Canadian Study of Health and Aging by the Seniors Independence Research Program of the National Health Research and Development Program of Health Canada. No endorsement from these agencies is implied. Author Contributions: Drs. St. John and Montgomery acquired funding for the MSHA, conducted clinical interviews, and conceived the idea behind these analyses. Dr. St. John conducted the analyses and wrote the draft manuscript. Dr. Montgomery corrected errors and made comments on the manuscript. Sponsor's Role: The sponsors had no role in the design, methods, subject recruitment, data collections, analysis, or preparation of paper.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.