Better Lighting to Reduce Falls and Fracture? A Comment on de Boer et al. (2004): Different Aspects of Visual Impairment as Risk Factors for Falls and Fractures in Older Men and Women

Abstract

With interest we read the article of de Boer et al.1 published in the September issue of the JBMR. In this study, the relationship between several visual functions and the occurrence of falls and fractures were studied in elderly people. One of the main results was that (impaired) contrast sensitivity (a measure of someone's ability to notice small spatial differences in luminance) is an independent risk factor for falling in the population of elderly people. Unfortunately, the authors hardly attempt to explain this important finding. We believe this is an omission, because doing so may suggest directions in which preventive measures should be sought. In particular, in the study of de Boer et al., a parameter called the integrated contrast sensitivity (ICS; a measure for the total amount of visual information gathered by the visual system2 was significantly associated with recurrent falling; hazard ratio [HR] of 1.53). A related parameter, the contrast sensitivity for low spatial frequencies, showed a comparable effect (HR of 1.66). As an aside, we note that the formula for calculating the ICS as published in de Boer et al. is incorrect (the method for calculating the ICS is described in the caption of their Fig. 1B). Presumably, a typing error changed the subscripts Xi and Xi−1 to Xi and Xi-1 and the same holds for Y. The correct formula for calculating the approximate value of the ICS is ∑(Xi − Xi−1)(Yi − Yi−1)/2. We assume the authors did use the correct formula in their analyses. Why would contrast sensitivity and ICS be such important determinants of falling, whereas the ability to recognize small details was not (reading the newspaper; HR of 1.12)? We propose the answer can be found in a study on object detection and recognition under different levels of illumination.3 In this study, we assessed the ability of visually impaired subjects to recognize objects in a test room furnished with items found in a regular home, such as a table, chairs, cups, etc. Moreover, contrast sensitivity and visual acuity of the subjects were measured under different levels of illumination. Importantly, in our study, contrast sensitivity was assessed using a distance chart (VSTS 6500; Vistech Consultants, Dayton, OH, USA). Objects were inspected from a distance of ∼3 m. Therefore, the study had a high degree of ecological validity for mobility. However, what makes this study so relevant in the present context is that it clearly shows that the subjects' ICSs best explained their visual performance under these daily life conditions. The ICS (as well as the low spatial frequency contrast sensitivity) explained a large part of the variation in the ability to detect and recognize objects (for ICS, the explained variance was 69% and 78%, respectively). Visual acuity, on the other hand, had a much lower explanatory power (43% and 59% of the variance in detection and recognition, respectively). Moreover, it did not add any explanatory power when combined with the ICS (69% and 80% of explained variance, respectively). A lower ICS is thus associated with a decreased ability to detect and recognize objects. A decreased ability to do so increases the chance of bumping into obstacles, in turn, increasing the risk of falling. Therefore, the ICS/object detection relationship is quite plausibly the mediating factor that explains why a lower ICS resulted in a higher risk for falling. This line of thought suggests that improving elderly subject's contrast sensitivity could significantly reduce their risk of falling. Unlike visual acuity problems, which can often be easily corrected using optometric means, degraded contrast sensitivity of a person as such is much harder to correct by a device. However, it is relatively easy to adapt one's environment to improve contrast sensitivity by increasing light levels.4 Therefore, we suggest that one simple means through which the risk of falling for the elderly could be reduced is through adaptation of indoor and outdoor lighting. It is very likely that this will be effective. In subjects with impaired vision, the number of unintended contacts with low contrast objects can be halved by changing the lighting conditions from the mesopic (low luminance) to the photopic (high luminance) range.5 An interesting additional beneficial effect of improved lighting is an improved quality of life of the visually impaired person.6, 7 Obviously, further improvements in the visual environment could be made through an adaptive use of color and contrast. In summary, the results of the study of de Boer et al., combined with our previous work, suggest that investments in the daily living conditions and improving the visibility of the elderly visual environment will presumably reduce their risk of falling and fractures, in turn resulting in savings on medical and care expenses.