Considerations for recommendation 1

• The effects of exercise can be enhanced by combining it with increased protein intake and other nutritional interventions.
• Consult a physical therapist or specialist, if available, before recommending exercise for older people.
• Refer for investigations into, and treatment of, associated underlying diseases, such as arterial and pulmonary disease, frailty and sarcopenia.
• Consider tailored, simple and less structured exercise programmes for older adults with limitations in cognitive function. For older people with severely reduced capacity, advise chair- and bed-based exercise training as a starting point.
• Environmental characteristics associated with older people gaining more physical activity include providing safe spaces for walking, ensuring easy access to local facilities, goods and services, seeing people of a similar age exercising in the same neighbourhood, and regular participation in exercise with friends and family.
• The effects of multimodal exercise interventions are enhanced when prescribed in association with self-management support. Self-management support also improves adherence.
• Multimodal interventions are a combination of different modes of exercise (aerobic, resistance, flexibility, balance), with an emphasis on important muscle groups and performed in a functional manner. Older adults should be offered guidance on the physical activity recommended for their age and health conditions. WHO provides recommendations that consider different starting points and levels of capacity for physical activity to maintain health (see http://www.who.int/dietphysicalactivity/factsheet_recommendations) (27).

Supporting evidence for recommendation 1

A systematic search identified 130 reviews, 11 of which served as the basis for the primary findings summarized below.

• Further detail on the supporting evidence is in the Evidence profile: mobility loss, available at http://www.who.int/ageing/health-systems/icope.

Seven reviews from high-income countries used a multimodal exercise programme of progressive muscle strengthening or generic strength training, balance retraining exercise, aerobic training and flexibility training. Pooled data from the trials included in these reviews indicated that this intervention significantly improved critical outcomes, including muscle strength of the lower extremity (10 trials, 1259 participants), balance (16 trials, 1313 participants), gait speed (15 trials, 1543 participants), chair stand test score (9 trials, 827 participants), overall physical function (9 trials, 976 participants) and activities of daily living (7 trials, 551 cases). The overall quality of evidence was rated as moderate as the results were consistently beneficial for all critical outcomes and the GDG considered that several of the critical outcomes would individually suffice to support a recommendation for the intervention.

Eleven trials, reported in four reviews, investigated the benefit of progressive resistance training in older people with mobility impairment. Evidence suggests that progressive resistance training improves muscle strength of the lower extremity (8 trials, 655 participants) and chair stand test scores (2 trials, 38 participants). The overall methodological quality rating was moderate for the muscle strength outcome and low for the chair stand test. Progressive resistance training had no effect on other critical outcomes (balance, gait speed, Timed Up and Go score, overall physical function and activities of daily living). In addition, three trials of t’ai chi training showed a significant benefit in terms of improving balance (348 cases), but no effect on the gait speed, chair stand score, activities of daily living or the number of falls. The overall methodological quality rating was low for the balance outcome.

Rationale for recommendation 1

Moderate-quality evidence supports the use of multimodal exercise training to improve the functional outcomes in older adults with mobility impairment. The GDG recognized a greater effect on critical functional outcomes for multimodal exercise. The effects and quality of evidence for stand-alone progressive resistance training and t’ai chi were not considered sufficient for incorporation into the recommendation.

Adverse events reported in a small proportion of trials were reviewed. The most commonly reported events were muscle soreness and joint pain. Very few trials reported serious adverse events, such as fracture, hospitalization or death. No clear relationship was noted between severe events and exercise: similar events were reported in the intervention and control groups.

The GDG recognized the additional cost associated with the scaling-up of supervised exercise training for older people. The GDG felt that the programme cost could be reduced through minimal training for family members and the provision of self-help guides.

The GDG agreed that multimodal exercise was critical to maintaining physical capacity in older people, and that it would be acceptable to them, to family members and to other stakeholders. Based on the moderate quality of the evidence, the widespread acceptability of exercise and the potential opportunities to shift health care tasks, the GDG made a strong recommendation.

Considerations for recommendation 2

• Nutritional assessments should be specific to the older person and include nutritional history, records of food intake or 24-hour dietary recall, physical examination with particular attention to signs of inadequate nutrition or overconsumption, and specific laboratory tests if applicable. There are several tools available to assess nutritional status in older people (28, 29).
• Assessment of muscle mass and muscle strength must be included in the assessment of nutritional status.
• Dietary counselling to ensure a healthy diet that provides adequate amounts of energy, protein and micronutrients should be encouraged for all older people, including those who are at risk of or affected by undernutrition.
• It is important to consider specially formulated supplementary foods (in ready-to-eat or milled form), which are modified in their energy density, protein, fat or micronutrient composition, to help meet the nutritional requirements of older people.
• Mealtime interventions (including family-style meals and social dining) are important approaches for managing undernutrition in older people. Consider family-style meals or social dining for older people living alone or who are socially isolated.
• Protein absorption decreases with age. Low protein intake is associated with loss of lean body mass, and standard protein intake may not be sufficient for older people.
• Refer older people with evidence of potentially serious underlying physical illness (gross cachexia, rapid weight loss, obstruction or difficulty swallowing, vomiting, chronic diarrhoea, abdominal pain or swelling) for medical review by a physician or specialist.

Supporting evidence for recommendation 2

Our search strategy identified three systematic reviews to inform these recommendations (30–32). We conducted an independent search strategy in 2015 to update the 2009 review (31) and identified 29 additional trials.

• Further detail on the supporting evidence is in the Evidence profile: malnutrition, available at http://www.who.int/ageing/health-systems/icope.

The search strategy involved older people who were at risk of or affected by undernutrition. All but two of the trials were conducted in high-income countries. The majority of the trials were in hospital settings or long-term care facilities (nursing, retirement or residential homes). The definition of undernutrition varied in the trials. The majority applied an anthropometric measure – typically body mass index (BMI) – as a nutritional status indicator, and compared it against WHO cut-off values (where underweight is below 18.50 kg/m²). The assessed interventions were aimed at improving the intake of protein and energy using only the normal oral route. Protein was provided together with non-protein energy sources such as carbohydrate and fat, and with or without added minerals and vitamins. The types of intervention considered included supplements in the form of commercial sip feeds; milk-based supplements; fortification of normal food sources; addition of fluid milk (low-fat or fat-free) to the usual daily consumption of dairy products; commercial nourishing drink made up with either milk or water; high-protein and high-calorie feeding supplementary to the hospital diet; commercial supplements or formulated meal-replacement commercial drinks in addition to meals; and other specially formulated nutrition products. Most supplementation trials aimed to provide, per serving, 300–400 kcal, 12–20 g of protein, and additional vitamins and minerals.

The evidence indicated that the consumption of oral supplemental nutrition significantly reduced mortality compared with people on placebo or usual care. In a subgroup analysis, the pooled data from trials conducted in community settings showed no effect on mortality, whereas the treatment effect on mortality remained significant in trials performed in hospital and long-term care settings.

Weight gain, rated as a critical outcome, was reported in 70 trials. These data showed that the intervention improved weight gain for older adults affected by undernutrition. In the subgroup analysis, a significant benefit was indicated in improving weight gain in these older people in trials conducted both in the community setting and in the hospital or long-term care setting.

Rationale for recommendation 2

Moderate-quality evidence showed that administration of oral supplemental nutrition plus dietary advice could prevent mortality and improve weight gain in older people affected by undernutrition. The GDG reviewed the adverse effects associated with this recommendation. Fifteen trials reported adverse effects in both treatment and control groups, but only four of these (two hospital studies, one nursing home study and one community study) provided a systematic evaluation and comparison of adverse effects in the treatment and control conditions. Common side-effects reported in the studies were gastrointestinal symptoms, nausea and diarrhoea. A higher number of adverse effects were reported in studies conducted in hospital settings; this may reflect the baseline severity of the undernutrition, the intensity of supplementation, the presence of comorbid acute illness or, possibly, increased monitoring of adverse effects. Other trials reported a prevalence of adverse effects in both the intervention and control groups, and the majority of these studies reported no between-group differences in adverse effects. The GDG therefore concluded that the potential risks associated with nutritional intervention were small.

Adherence to the nutritional interventions was discussed in detail. The GDG suggested that oral supplemental nutrition may be acceptable to many older people, and would assist those at risk of, or affected by, undernutrition to meet their nutritional requirements. In conclusion, the GDG agreed that these recommendations would be appreciated by older adults and acceptable to key stakeholders.

The implementation of this recommendation may have major resource implications, particularly in the training of staff members. However, in many low- and middle-income countries, community health workers deliver nutritional interventions for children affected by undernutrition and for pregnant mothers. Based on this experience, the GDG concluded that training could be undertaken for existing human resources to implement these recommendations.

The GDG considered that if recourse to the provision of supplemental nutrition or specific food products was necessary to increase an individual’s dietary intake of protein, energy or vitamins and minerals, this should always be combined with dietary advice. Provision of dietary advice will aid an older person’s understanding of the need for oral supplemental nutrition and will ensure that their dignity and human rights are respected.

Based on the evidence, the GDG made a strong recommendation in favour of oral supplemental nutrition for older people affected by undernutrition.

The GDG also considered the evidence for increasing dietary intake and mealtime interventions. Although there was enough evidence about their benefits to support a recommendation, the GDG decided that, due to the generic nature of these two interventions, it was more appropriate not to issue a recommendation.

Considerations for recommendation 3

• At the primary health care level, visual screening can be performed using a Snellen chart to screen for visual acuity.
• It is important to improve public awareness and generate demand for services through regular community outreach activities.
• Promote case finding at the primary and community care settings, where health care personnel such as community health workers can be trained to screen for visual acuity.
• Establish comprehensive eye-care services, so that refraction services with the provision of suitable correction tools are available at the primary health care level.
• Specifically, provide spectacles that are new, of high quality, accessible and affordable in low-income settings.
• The most common causes of vision impairment in older people include presbyopia, cataract, glaucoma, diabetic retinopathy and age-related macular degeneration. Older people found to have a visual impairment should therefore be assessed for these medical conditions.
• Older people who have had diabetes for five years or more must be referred for an assessment with an ophthalmologist. Additionally, it is advisable that people who are at risk of glaucoma (including people of African descent and people with a positive family history), who are at risk of diabetes, or who have severe myopia undergo periodic assessment by an ophthalmologist. The WHO publication, Prevention of blindness from diabetes mellitus (33) is available at http://www.who.int/diabetes/publications/prevention_diabetes2006.
• Refer to eye-care practitioners or occupational therapists who have expertise in environmental modifications (working with colour and contrast in the environment of the person with low vision) and can teach activities of daily living and skills, such as washing clothes.

Supporting evidence for recommendation 3

Evidence was compiled from three systematic reviews: an updated systematic review that identified five trials of screening and referral, an updated systematic review that identified two trials of screening plus provision of immediate eye care, and an updated systematic review that identified three trials of expedited cataract surgery.

• Further detail on the supporting evidence is in the Evidence profile: visual impairment, available at http://www.who.int/ageing/health-systems/icope.

No new trials have been identified in a WHO update of a systematic review that was published in 2006 on screening and referral, which found five trials (35). For these guidelines, results from the three initial systematic reviews therefore comprise the evidence base. In it, pooled data from five trials of 3494 participants indicated that there was no evidence to suggest that visual screening alone could improve visual function in older people. The authors concluded that the reasons for the lack of benefit were high loss to follow-up, contamination of the intervention, similar frequencies of vision disorder detection and treatment in both groups, the use of one screening question to identify people for further testing, and a low uptake of recommended interventions.

A review that identified two trials of visual screening combined with immediate referral for correction of refractive errors revealed evidence of beneficial effects. In the first of the two trials, older people in the intervention group received prescriptions and vouchers for free eyeglasses (36), while participants in the second study were immediately provided with corrective glasses (37). The participants in the first trial were people 65 years of age and over living in the community, whereas the second trial recruited nursing home residents 55 years of age and over. In both trials, visual functioning improved in the immediate-treatment groups.

The systematic review that identified three trials examining the effectiveness of expedited cataract surgery found substantial improvements in vision for older people who had undergone expedited surgery, compared with outcomes for people in the routine surgery or waiting list groups (38–40).

Rationale for recommendation 3

The GDG acknowledged the higher prevalence of vision impairment in older people compared with younger, and the enormous individual and societal burden associated with untreated vision conditions. The group considered the limited supportive evidence for the effects on self-reported visual improvement following screening and referral (35). The GDG agreed that the use of screening as a stand-alone intervention was not warranted and that it should be combined with immediate provision of indicated eye care to improve the visual acuity of older people with visual impairment. In addition, the large beneficial effects of cataract surgery observed in three trials was noted by the GDG in support of the provision of cataract surgery, when indicated. None of the trials reviewed reported any adverse consequences associated with screening for vision plus timely provision of care. The GDG recognized the high acceptability and feasibility of this screening and care. The vision-care experts in the GDG mentioned that in many countries, including low- and middle-income ones, there were national programmes for the management of blindness in place, in which professionals trained in the early identification of avoidable blindness performed vision screening. This screening was focused largely on children, however, while many older people experienced difficulties accessing such screening and timely provision of care. The GDG felt that screening coupled with provision of indicated eye care might increase equity in this field.

Given the minor variability in values and preferences, the feasible and acceptable nature of the intervention, and the potential for benefits to greatly outweigh harms, especially in high-burden countries, the GDG made a strong recommendation despite the low quality of the evidence.

Considerations for recommendation 4

• Community awareness about hearing loss and the positive benefits of audiological rehabilitation in older people, through community case finding and outreach activities, should be promoted.
• Health care professionals should be encouraged to screen older adults for hearing loss by periodically questioning them about their hearing. Audiological examination, otoscopic examination and the whispered voice test are also recommended.
• Hearing aids are the treatment of choice for older people with hearing loss, because they minimize the reduction in hearing and improve daily functioning.
• Medications should be reviewed for potential ototoxicity.
• People with chronic otitis media or sudden hearing loss, or who fail any screening tests should be referred to an otolaryngologist.
• Additional guidance can be found in the WHO Guidelines for hearing aids and services for developing countries (41), available at http://www.who.int/pbd/deafness/en/hearing_aid_guide_en.pdf.

Supporting evidence for recommendation 4

Evidence for this recommendation was obtained by reviewing two randomized controlled trials. Both trials demonstrated the benefit of screening and immediate provision of hearing aids in older adults. The earlier of the two found that immediate provision of hearing aids was associated with statistically significant improvements in the hearing-related quality-of-life score the Hearing Handicap Inventory for the Elderly (HHIE), and in the Quantified Denver Scale of Communication Function (QDS) score (42). In the second trial both hearing aid groups experienced greater improvements in hearing-related outcomes compared with the no-treatment and assistive-listening device groups (43). The mean improvement in HHIE scores in this trial was small for control patients (2.2 points) and those who received an assistive listening device (4.4 points), larger for patients who received a conventional device (17.4 points), and considerable for patients who received a programmable device (31.1 points).

• Further detail on the supporting evidence is in the Evidence profile: hearing loss, available at http://www.who.int/ageing/health-systems/icope.

Rationale for recommendation 4

The GDG considered there was low-quality evidence supporting the use of screening and provision of hearing aids as a way to improve critical hearing-related outcomes for older people. In addition to the evidence, however, the GDG also considered issues such as the opportunity costs and usefulness of potential interventions given the very high prevalence of, and the enormous societal implications associated with, undiagnosed and untreated hearing loss; worldwide, one third of older people live with some degree of hearing loss.

The GDG members thus agreed that the benefits of the intervention outweighed the disadvantages and costs. Screening and use of hearing aids does not seem to harm individuals, high-quality hearing aids can now be fitted at an affordable cost, and most older people do not object to being assessed. Based on the acceptability, feasibility and increasing affordability of hearing aids – coupled with the potentially large beneficial effects afforded to older people living in high-burden countries if they are able to engage and communicate effectively within their communities – the GDG decided to issue a strong recommendation despite the low-quality evidence.

Considerations for recommendation 5

• Assessment for cognitive function can be performed using any locally validated tool.
• In the absence of standard assessment, the person, and also someone who knows them well, should be asked about problems with memory, orientation, speech and language, and any difficulties in performing key roles and activities. Memory, orientation and language should be assessed.
• Cognitive stimulation could be delivered in short sessions. In high-income countries, it is usually administered by psychotherapists. Some characteristics of the intervention such as duration or frequency could, however, be adapted for each setting, and it could be administered by suitably trained and supported non-specialists.
• It is important to encourage family members and caregivers to provide older people with regular orientation information (day, date, time, weather, names of people, and so on), to help them remain orientated in time and place. They can use materials such as newspapers, radio and television programmes, family albums and household items to promote communication, orientate the person to current events, stimulate memories and enable them to share and value their experiences.
• Impairment in cognitive function may be associated with memory deficits and difficulties managing instrumental activities of daily living such as finances and shopping, and with impaired social function. Cases should be referred for medical assessment.
• More guidance on dementia can be found in the WHO mhGAP intervention guide (45), available at http://www.who.int/mental_health/mhgap/mhGAP_intervention_guide_02.

Supporting evidence for recommendation 5

Evidence on the effectiveness of cognitive stimulation interventions for older adults with cognitive impairment was extracted from one systematic review (44). In this study, interventions were typically delivered in day-care or long-term care settings, and involved participants with dementia or mild cognitive impairment, or both. The review analysed pooled data from 17 trials: 12 studies (810 participants) assessed cognitive impairment using the Mini Mental State Examination (MMSE) while the other five trials assessed cognitive function using the Alzheimer’s Disease Assessment Scale – Cognitive subscale (ADAS-Cog). Evidence from all of these trials showed significant improvement in cognitive function after the intervention. The overall quality of the evidence was low. New randomized controlled trials are needed to test the efficacy of different types of cognitive-based interventions that exclusively target older adults with cognitive impairment.

• Further detail on the supporting evidence is in the Evidence profile: cognitive impairment, available at http://www.who.int/ageing/health-systems/icope.

Rationale for recommendation 5

Low-quality evidence supports the use of cognitive stimulation interventions (of any form) to improve cognitive function in older people with mild cognitive impairment and dementia. The GDG recommends that health care professionals provide these interventions to people who are eligible. The GDG identified low-quality evidence that the intervention improved important health outcomes, and concluded that the benefits outweighed the adverse effects. The intervention is non-invasive and no trial reported any harms associated with cognitive stimulation. Variability in values and preferences was noted to be minor, and the intervention was considered feasible and acceptable. Resource requirements for delivery of cognitive stimulation interventions would initially be considerable, but the GDG argued that adaptation of the intervention for specific settings, and investment in training of non-specialists, would potentially discount future costs. The strength of this recommendation is conditional, due to the low quality of the evidence – most trials involved older people who had dementia, and the effects of cognitive stimulation interventions in those with mild cognitive impairment without dementia remains unclear.

Considerations for recommendation 6

• Older people can experience psychological difficulties consistent with the symptoms of depression but without these necessarily meaning they have moderate-to-severe depression. When assessing older people, it is important to assess whether the person has depressive symptoms, but also if these are associated with social isolation, and whether the person has difficulties in day-to-day functioning due to depressive symptoms.
• Cognitive impairment and dementia may be associated with depressive symptoms and must be assessed. People with dementia often present with complaints of mood or behavioural problems, such as apathy, loss of emotional control, or difficulties carrying out usual work or domestic or social activities.
• The management and assessment of depressive symptoms is covered by the WHO mhGAP intervention guide (45) (within the module for ‘Other significant emotional or medically unexplained somatic complaints; see Box 5).
• Older people who qualify for a diagnosis of depressive disorder should be advised and treated as recommended in the mhGAP guidelines.
• Physical exercise should be considered as an adjunct to structured psychological treatments to improve intrinsic capacity in older people (see the guidance in section 3.1.1 on mobility loss).

Box 5

Summary information for treatment of depression.

Supporting evidence for recommendation 6

Evidence on the benefit of psychological intervention for managing depressive symptoms in older adults was extracted from two systematic reviews (47, 48). All the trials reviewed were conducted in high-income countries and administered by trained mental health care professionals.

Pooled data from six trials (826 older adults) that used cognitive behavioural therapy, problem-solving therapy and life-review therapy indicated that these interventions considerably reduced depressive symptoms in older adults. The overall quality of the evidence was low. Another review examined the effectiveness of behavioural activation specifically in reducing depressive symptoms in adults. However, only three of the included trials recruited older adults. Evidence from these trials (102 older adults) showed that behavioural activation significantly reduced depressive symptoms in the intervention group. The overall quality of the evidence was very low.

• Further detail on the supporting evidence is in the Evidence profile: depressive symptoms, available at http://www.who.int/ageing/health-systems/icope.

Rationale for recommendation 6

Very low-quality evidence supports the use of psychological interventions (cognitive behavioural therapy, problem-solving therapy, interpersonal counselling, behavioural activation therapy and life-review therapy) to reduce depressive symptoms in older adults. No trials reported harms associated with these interventions. In the absence of any specific harms, the GDG concluded that these interventions were likely to have limited potential for harm. The administration of behavioural activation is a relatively unsophisticated intervention that can be learned more quickly than most other evidence-based psychological treatments. The intervention has been studied mainly as a multiple-session intervention performed by specialists, however, which may not generalize to non-specialized health care and carries considerable resource implications. Nonetheless, the intervention could be modified into a brief intervention as an adjunct treatment or as part of a first step in a comprehensive care approach in primary health care. Although the evidence specifically for older people is scarce, WHO has comprehensive tools and guidelines to manage depressive symptoms in adults. Given that depression is associated with a severe decline in functional ability among older people – and that a recommendation in favour of the provision of brief psychological interventions would be consistent with the existing WHO mhGAP recommendation for depression (45) – the GDG concluded that the benefits outweighed the harms. In view of the very low quality of evidence and the possible lack of generalizability to all community settings, the GDG issued a conditional recommendation for the treatment of depressive symptoms in older adults.

Considerations for recommendations 7 and 8

• Urinary incontinence in older people is multifactorial and needs evaluation and treatment that is not focused solely on the lower urinary tract. Although an assessment of urinary incontinence can be made by non-specialized health workers, full evaluation is the responsibility of a medical professional or clinician. The full assessment is needed because of the multifactorial nature of urinary incontinence in older people. The examination should include cardiovascular, abdominal and neurological systems as well as assessment of mobility and cognition.
• An assessment of urinary incontinence includes the evaluation of fluid intake, medications, physical and cognitive capacity (including mobility), and previous urological surgeries.
• The single best question to ask when diagnosing urge incontinence is: “Do you have a strong and sudden urge to void that makes you leak before reaching the toilet?”
• A good question to ask when diagnosing stress incontinence is: “Is your incontinence caused by coughing, sneezing, lifting, walking or running?”
• The person needs to be assessed for reversible causes of urinary incontinence, such as delirium, infection, atrophic vaginitis, pharmaceutical causes such as medication-induced urinary retention, psychological disorder (depression), excessive urine output (hyperglycaemia, for example), and stool impaction.
• As a first-line treatment, provide advice on bladder training for a minimum of 6 weeks. Bladder training involves advising the older person to follow a strict schedule for bathroom visits. The schedule starts with bathroom visits every 2 hours, but the time between visits should be gradually increased to improve bladder control.
• Pelvic floor muscle training (PFMT) strengthens the muscles that support the urethra and augment its closure. Often used for stress urinary incontinence, PFMT may help with urge leakage as well. Similar to other muscle-strengthening regimens, PFMT is based on controlled repetitions of high-intensity contractions, held for as long as possible. A starting regimen could be 3 sets of 10 contractions (with adequate relaxation between each) repeated 2–3 times per week.
• Key to the success of PFMT is correct identification of the target muscles and appropriate motivation to continue the programme.

Supporting evidence for recommendations 7 and 8

Five systematic reviews were identified, of which two systematic reviews served as the basis for the primary findings on prompted voiding and PFMT interventions (52, 53).

Rationale for recommendations 7 and 8

Low-quality evidence supports the use of prompted voiding to reduce episodes of urinary incontinence among older people with cognitive impairment. Urinary incontinence is common among those with cognitive impairment and increases the need for formal and informal care. No trial has reported adverse effects associated with prompted voiding interventions. All of the included trials were conducted in high-income countries in long-term care settings and the GDG recognized that these interventions may be difficult to implement in community settings reliant on the help of family caregivers. Based on the low-quality evidence and the potential challenges to implementation in community settings, the GDG made a conditional recommendation.

Low-quality evidence supports PFMT when used on its own to reduce incontinence in older women with urinary incontinence. When combined with bladder control strategies and self-monitoring, the quality of evidence increases to moderate in support of using PFMT. Urinary incontinence has a profound impact on the older person’s quality of life and functional ability, and increases the need for care. No trial has reported adverse effects associated with this intervention, and the GDG considered that the potential for harm from PFMT was likely to be low given the non-invasive nature of the intervention. The GDG indicated that the recommendation was likely to be valued by older women with urinary incontinence, and that the intervention was highly acceptable to health care providers. Based on the moderate quality of the evidence for the combined approach, and the minimal harms, the GDG made a strong recommendation for provision of PFMT both alone and in combination with other strategies.

Considerations for recommendations 9–12

• Older people who present for medical attention because of a fall, report recurrent falls in the past year, or demonstrate abnormalities of gait and/or balance should be offered a comprehensive risk assessment.
• A comprehensive assessment may include the following items: history of falls; gait, balance, mobility and muscle weakness; osteoporosis risk; fear of falling, vision impairment, cognitive impairment, neurological examination; urinary incontinence; home hazards; cardiovascular examination; and medication review.
• Multimodal exercise, including progressive resistance training and other exercise components (balance, flexibility and aerobic training) must be included in every care plan for older people at risk of falls (see Recommendation 1 on mobility loss).
• Identification of older people with visual impairment and their referral for interventions should be implemented in any approach to prevent falls. To prevent falls for older people with cataract, for example, immediate surgery should be recommended.
• Medication review by a trained health care professional, especially to reduce psychotropic medication, has been shown to reduce falls. Older people should be encouraged to reduce their use of sleeping aids, including over-the-counter medication containing diphenhydramine or other sedating antihistamine. Benzodiazepines and antidepressants have also been associated with falls.

Supporting evidence for recommendations 9–12

One systematic review of interventions designed to reduce the incidence of falls in older people living in the community was identified (55). The review included 159 randomized controlled trials with a total of 79 193 participants. Most trials compared a falls prevention intervention with no intervention or one that was not expected to reduce falls.

Rationale for recommendations 9–12