Screening and Intervention to Prevent Falls and Fractures in Older People

“We assessed the clinical effectiveness and cost-effectiveness of a brief falls-risk screening questionnaire, sent by mail, followed by an exercise program or a multifactorial intervention targeted to persons at increased risk for falls, as compared with no screening in community-dwelling older people. All participants received advice by mail.

[..] This was a three-group, pragmatic, cluster-randomized, controlled trial with parallel economic evaluation that included 63 general practices in seven rural and urban regions in England. Drawing from their patient registries, general practices contacted community-dwelling persons 70 years of age or older who were living in their own homes. Residents of assisted living facilities (with or without nursing care) and persons with terminal illnesses or life expectancy of less than 6 months were excluded.

[..] Each participating practice randomly selected up to 400 persons from their patient registries, informed them by mail that the practice was participating in research about treatments to improve the health of older people, and recruited them to complete an 18-month series of surveys about aging. [..] The mailed recruitment invitation included a baseline survey, consent form, and an advice booklet on preventing falls, “Age U.K. Staying Steady.”10 Practices sent one reminder letter to persons who did not respond to the invitation. Once 150 to 250 participants from each practice returned the survey and a signed consent form, we closed enrollment. With computer-generated randomization administered by an independent programmer, we randomly assigned the three practices in each local health district to the three interventions, one practice to each intervention. Practices assigned to the multifactorial fall-prevention strategy and the exercise strategy sent participants an additional brief screening questionnaire about the risk of falls, with a prepaid return envelope [..]. For participants whose responses to the questionnaire indicated that they were at increased risk for falls, multifactorial fall prevention or exercise, according to the random assignment, was arranged through the participants’ usual National Health Service (NHS) provider. The falls-risk screening questionnaire was a validated algorithm that was based on guidelines of the American Geriatrics Society and British Geriatrics Society, with small adaptations to enhance sensitivity and to suit administration by mail.

[Exercise Intervention] We used the Otago Exercise Program, which includes progressive home exercises for strength and balance performed at least twice a week and a recreational walking program. We trained physical therapists to deliver a minimum of seven sessions over 6 months. At least four sessions, including the initial session, were required to be in person, and the remainder could be conducted by telephone. Therapists used behavioral interventions to encourage adherence and provided participants with ankle weights to use during exercises.

[Multifactorial Intervention] Nurses, general practitioners, and consultant geriatricians assessed falls and medical history, gait and balance, fear of falling, postural hypotension, arrhythmia, medications, visual acuity, and feet and footwear status and conducted a home environment interview. Linked treatments included a medication review, exercise (the same as that used in the exercise strategy), home modifications, and referral to opticians, medical specialists, and podiatrists. Gait and balance assessment included timed tests. All medications, including over-the-counter medications, were screened. If medications that confer a predisposition to a fall were identified, a medical practitioner conducted a face-to-face review. The multifactorial intervention was provided in general practices or hospital clinics. Trained assessors observed at least one session and provided feedback to the practitioners carrying out the intervention.

We enrolled 63 participating general practices during the period from September 2010 through June 2014, recruited 9803 participants, and obtained data on fractures from the NHS Digital Hospital Episode Statistics for 9802 of the 9803 participants. We had general practice records for 9644 participants (98%); one exercise practice refused access to records after randomization. Over the course of 18 months, 289 participants (3%) died and 1213 (12%) did not complete the surveys, with no differences between trial groups.

[..] More than 95% of intervention sessions (5996 of 6280) were provided within the usual NHS provider network, and the remainder of contacts were through university clinical staff. Acceptance of the offer of intervention was higher in the multifactorial fall-prevention group (762 of 1074 [71%]) than in the exercise group (697 of 1079 [65%]). In total, there were 3842 intervention sessions for exercise and 2530 for multifactorial fall prevention.

[..] The mean (±SD) number of exercise sessions was 5.5±1.98, with no difference between groups in attendance or strength and balance outcomes. The majority of participants had improvement or remained at the upper level of strength (391 of 454 [86%]) as measured by the Otago Exercise Program strength scale. Evaluations for balance showed that 330 of 453 (72%) of participants had improvement or remained at the top level of balance as measured by the Otago balance scale. Over the course of the trial, only prescriptions for mineral supplementation changed, from 13.8% to 15.6%, with no difference between groups.

[Outcomes] The greatest number of fractures occurred among persons assigned to the multifactorial fall-prevention strategy, and the fewest occurred in the group that received advice by mail only. There were no significant differences in fracture rates (number of fractures per 100 person-years) between the exercise group and the advice-by-mail group (adjusted rate ratio for fracture, 1.20; 95% CI, 0.91 to 1.59) or between the multifactorial fall-prevention group and the advice-by-mail group (adjusted rate ratio, 1.30; 95% CI, 0.99 to 1.71). Any differences in fall rates were not sustained over 18 months. There were no differences in the SF-12 or Strawbridge Frailty Index scores and no subgroup or adherence effects.

Among the participants in the nested analysis group, who were at increased risk for falls (4192 of 9803 participants), the fracture rate was 3.70 per 100 person-years in the exercise group (adjusted rate ratio in the comparison with advice-by-mail group, 0.94; 95% CI, 0.65 to 1.35), 5.12 per 100 person-years in the multifactorial fall-prevention group (adjusted rate ratio in the comparison with advice-by-mail group, 1.26; 95% CI, 0.89 to 1.78), and 4.28 per 100 person-years in the advice-by-mail group.

Our findings are consistent with the broader evidence base, including a recent trial of multifactorial fall prevention in women. A recent Cochrane review reported limited and variable effects of multifactorial interventions on falls and included no reliable evidence about fractures. In our trial, exercise had less effect on falls than was reported in some other published studies, but we used a longer-term follow-up than most. When applied in pragmatic settings, screening by mail followed by a targeted exercise intervention or multifactorial approach for prevention of falls did not result in a lower rate of fractures than advice by mail alone.”

Full article, Lamb SE, Bruce J, Hossain A et al. New England Journal of Medicine 2020.11.5