This section describes the factors and issues the Public Health Advisory Committee (PHAC) considered when developing the recommendations. Please note: this section does not contain the recommendations.

General considerations


This guideline focuses on the delay and primary prevention of dementia, disability and frailty. The PHAC acknowledged the importance of national policy levers in encouraging behaviour change, but recognised that it was outside the remit of this guideline.


The PHAC acknowledged that there are different types of dementia, with differing prevalence rates and opportunities for risk reduction. Although Alzheimer's disease is the most common type, less is known about its risk reduction than for vascular dementia, it is now thought that many people diagnosed with Alzheimer's disease have a mix of Alzheimer's disease and vascular dementia. Vascular dementia is caused by a reduced blood supply to the brain. A series of small strokes can occur that often go unnoticed or are seen as transient ischemic attacks, resulting in only temporary reductions in blood supply and brief symptoms. However, if they occur repeatedly more of the brain is damaged and dementia starts to develop. Vascular dementia therefore has the same risk factors as cardiovascular disease and stroke, and so the same preventive measures are likely to reduce risk.


A number of health conditions and environmental circumstances can contribute to dementia, disability and frailty. The behavioural risk and protective factors covered in the guideline will not be the same for all types of dementia, disability and frailty so risk reduction may be more effective for some conditions than others. The PHAC was satisfied that there is sufficient evidence to show the risk of developing them can be reduced through changing common behavioural risk factors. This includes quitting smoking, increasing physical activity, reducing alcohol consumption, having a healthy diet, and reaching and maintaining a healthy weight. However, key messages about risk reduction, particularly for dementia, are not well publicised or understood by health and other professionals or the public, unlike the link between smoking and cancer.


The guideline focused on delay and primary prevention, therefore the management of obesity was not considered to be covered by the scope. However, evidence relating to change in weight and BMI from the normal range to overweight or underweight was considered.


Sexual health was not included in the search strategy as there was a need to prioritise the areas covered because of time and resource constraints. However, sexual health matters (for example safe sex and contraceptive advice) are still issues for both men and women in mid-life. HIV is a cause of disability and frailty, and in rare cases it can cause dementia.


The PHAC was aware that people often have more than 1 behavioural risk factor. They agreed an order of importance of risk factors in the guideline based on the strength of the associations reported in expert testimony 1. However, the PHAC also agreed that the most appropriate approach will come from people working with health and social care professionals to consider their own needs along with contextual and local factors. Mid‑life is not too late for people to make meaningful changes. People often need more than 1 attempt to change, and mid‑life can be the period in which change is finally sustained. The PHAC agreed, based on the evidence, that mid‑life changes (supported by professionals and services) can help to reduce the risk of dementia, disability and frailty.


The early onset of non‑communicable chronic conditions and early mortality in disadvantaged groups was considered at the scoping stage of the guideline. To address this, the lower age limit for the inclusion of evidence in the reviews was reduced so data were included from adults aged 39 and younger from vulnerable and disadvantaged groups.


The PHAC recognised that mid‑life is not the only time to make changes to smoking, physical activity, alcohol intake and diet, and that these issues are important throughout life: the earlier healthy behaviours are adopted the more likely a person is to have more years free from illness, disability and frailty. The PHAC agreed that putting policies and services in place to encourage change in relevant behaviours could have huge benefits for individuals, families and the population as a whole. The group acknowledged the range of policies currently in place; the guideline aims to ensure these provide coherent and consistent advice so that prevention services are better value for money. The group also acknowledged that the Department of Health and Social Care and Public Health England are already undertaking many of the actions outlined in the guideline and it is important that these are maintained.


The PHAC agreed to focus on dementia in recommendation 2. The reason for this was that the non‑communicable conditions covered by current policies are significant contributors to the development of a range of disabilities and also to frailty and hence disability and frailty are to some extent included in existing policies.


The PHAC heard expert evidence that population approaches are more cost effective for reducing deaths from non‑communicable chronic diseases than individual approaches. In particular, reductions in smoking rates led to dramatic reductions in cardiovascular disease, lung cancer and chronic obstructive pulmonary disease. It considered that population approaches are therefore key to helping people reduce exposure to risk factors.


The PHAC agreed that individual behaviour change approaches are likely to be more cost effective and less likely to widen health inequalities when supported by population‑based approaches. NICE's guideline on behaviour change: individual approaches makes recommendations on effective and cost‑effective behaviour change techniques for working with individuals.


The PHAC recognised that basic needs such as housing or employment have an influence on health. Therefore, other local authority services such as housing and planning or economic development and regeneration can have an effect on risk of disease. Housing and employment problems may make it difficult for people to change behaviours, however this should not prevent them from being offered help to reduce the risk of dementia, disability and frailty.


The PHAC agreed that interventions and services need to be accessible to the whole community, particularly the more disadvantaged. Equity issues, such as language skills, information content or service location and opening hours or lack of internet access may all affect accessibility. In turn, these factors will affect the effectiveness and cost‑effectiveness of services. Providing digital services (using, for example, computers, phones and tablets) is one way of increasing accessibility. But the digital skills of people over 40 will vary widely, so this should not be the only mechanism of delivery.


Raising awareness of the links between risk factors and dementia, disability and frailty is unlikely to be enough to change people's behaviour alone, because knowledge does not always lead to action. There is also a risk of widening health inequalities because more educated people and those with greater self‑efficacy tend to access and act on new information more readily. Targeted interventions are needed to help get messages to those most at need and to help them make changes.


The PHAC noted that changes in the unhealthy behaviours covered in the guideline could also reduce the incidence of non‑communicable conditions that have impacts beyond dementia, disability and frailty in later life. Also, other household members may benefit from behavioural change (for example from less exposure to second‑hand smoke or a healthier diet).


Children and young people are influenced by what they see. By changing their own smoking, physical activity, drinking and dietary behaviours, people in mid‑life may positively influence the health of children and young people.


Social norms can affect behavioural risks. It is becoming less usual for people to smoke, and that is an important driver for change. Social norms also exist for other behaviours, and need to be challenged. Drinking alcohol daily at home has become normal for some people, and this poses a threat to health. Reducing activity, slowing down and having earned a rest, are often seen as an expected part of growing older. However, many people continue to be fully active in later life and take up and enjoy new activities, and this is good for their health and well‑being.


The PHAC was aware of the alcohol risk curve for cardiovascular disease reported in the literature that shows a small risk for non‑drinkers, a lesser risk for very low drinkers and increasing risk with increasing consumption of alcohol. They heard evidence that this apparent relationship between alcohol and risk may be compounded by problems with the evidence base, for example, the failure of some studies to disaggregate people who have never consumed alcohol from those who have stopped for health reasons. This means that risk observed for non‑drinkers may be overstated. The risk for some cancers shows a different pattern, with lowest risk for non‑drinkers and a linear increase in risk with increasing consumption.


The PHAC heard expert testimony suggesting that, in light of current evidence and issues with the evidence base, the overall message should be that there is no safe level of alcohol consumption. The PHAC were aware that the chief medical officer was asked to review the alcohol guidelines from 2012, and that draft conclusions from this review were likely to be published in 2015.


Alcohol action teams, and risk identification and brief interventions, are not always provided systematically and sufficiently scaled to address health inequalities. The PHAC was also aware that the advice given differs. It is important that all health and social care professionals are given training and information so they can take advantage of every opportunity to identify risk and provide brief advice.

The evidence


The evidence reviews developed to inform this guideline focused on the referral from the Department of Health and the scope questions. The reviews included primary studies and systematic reviews that investigated interventions and services that were explicitly aimed at mid‑life populations and at vulnerable and disadvantaged adult populations. There is a broader evidence base that includes the general population as well as people in mid‑life that could have been used when developing this guideline. However, including the general population in the reviews would have made them unmanageable in the time and resources available.


Literature searching without age restrictions gave more than a million results, so a search strategy was developed to identify evidence that was specific to people in mid‑life. Studies of people in mid‑life that did not include mid‑life terms in the title or abstract may not have been found using this strategy. However, the inclusion of recent systematic reviews is likely to mean much of this literature was covered. The review team noted that the primary study participants in many reviews were in mid‑life, although there was no mention of a focus on mid‑life.


The wider evidence base has been used across multiple topic‑specific NICE guidelines; when appropriate, the PHAC considered other relevant recommendations. The decision whether or not to cross‑refer to or include recommendations from other NICE guidelines was based on the PHAC's view of whether evidence from the general adult population could be applied to people in mid‑life. This wider evidence base was also the subject of testimonies from a range of experts, detailed in the evidence.


Any lack of evidence in the commissioned reviews should not be interpreted as suggesting that a particular behaviour does not have a role in the development of or protection against dementia, disability and frailty. Nor should it be inferred that there is no evidence of its effectiveness. Instead, it implies that there is a lack of evidence exclusive to mid‑life. When recommendations have been made the PHAC members used their judgment about the applicability and relevance of interventions.


Only a limited amount of vision and hearing loss literature was found that reported dementia, disability and frailty outcomes. This covered studies done mainly in groups of older people not in mid‑life, and there was little evaluation of the effect of mid‑life interventions on preventing the loss. This is relevant to this guideline because vision and hearing problems are risk factors for dementia, disability and frailty. They can make related issues, such as social isolation, depression and the risk of falls worse, as well as affecting people's ability to be involved in their own medical care.


Expert testimony suggested that easy access to hearing tests is important, as there is a high rate of undiagnosed hearing loss in people in mid‑life. If needed, early use of a hearing aid can result in greater benefit through additional years of use and better adaptation to use (Davis 2007). Likewise, tests of vision for early detection of eye diseases such as glaucoma are important to commence early treatment and prevent sight loss. Health and Safety Executive standards regulate noise exposure and hazards to sight in the workplace but protection in domestic and social settings is equally important.


The PHAC received expert testimony on sleep disorders and their link with mild cognitive impairment, dementia and other non‑communicable chronic conditions such as diabetes and Parkinson's disease. Particular groups of people may be at high risk of non‑communicable chronic conditions as a result of disordered sleep, for example shift workers and people with untreated sleep apnoea. Sleep apnoea adversely affects cognitive function; NICE has produced guidance on its management (see NICE technology appraisal guidance on continuous positive airway pressure for the treatment of obstructive sleep apnoea/hypopnoea syndrome and NICE interventional procedures guidance on radiofrequency ablation of the soft palate for snoring). The PHAC considered that the evidence for other sleep interventions to reduce the risk of dementia and other non‑communicable chronic diseases was insufficient to make recommendations, but has made a recommendation for research.


There is emerging evidence on the importance of psychosocial risk factors throughout life such as loneliness, isolation and depression. These factors may reduce resilience to disease onset and progression. However, there is a lack of evidence on the most effective ways to address psychosocial factors and on their effect on the development of dementia, disability and frailty. Psychosocial factors may be as important as physical factors in reducing the risk of dementia, but more evidence is needed.


There may also be some protective factors, such as high levels of education. Expert testimony suggested that people with high educational attainment appear to be more resilient to dementia. Access to education and training in mid‑life provides an opportunity to improve lifestyle, and hence reduce risk.


The PHAC was aware of the limitations of the economic analysis. This focused on dementia outcomes resulting from the effect of physical activity alone. This approach was taken because time, resources and the evidence available did not allow extensive economic analyses. Also, existing economic models for tobacco and alcohol in other NICE guidance demonstrate cost‑effectiveness. PHAC members were confident that changing multiple behaviours was likely to result in further benefits.


The model demonstrated the potential cost effectiveness of population‑level interventions using a cohort of people aged 40. Population‑level interventions would have a beneficial impact on the whole population, so a model based only on a mid‑life cohort is a conservative estimate of the total benefit.


People who increase their levels of physical activity not only reduce their probability of developing dementia, but also their probability of developing cardiovascular disease and cancer. The physical activity model showed the potential for adding healthy years to life as a result of a small change from no activity (sedentary) to low–moderate activity. Because of living longer, some people who otherwise would have died earlier will go on to develop dementia. However, the model suggests that the net overall effect of an intervention to increase physical activity is to decrease the prevalence of dementia, to increase life expectancy and to reduce the average length of disability.


The economic model links an increase in physical activity to future health benefits, and to future health cost savings (typically accruing to the NHS) and future social care cost savings (typically accruing to local government). Two specific interventions were modelled: exercise referral schemes and mass media campaigns. Although the evidence is that these interventions have only weak effects, the model suggested they could be cost effective. This was based on an assumption that exercise would be sustained over time, with relapse rates of no more than 10 to 20% per year. The model makes no assumption about who funded the interventions in the scenarios above, and no assumptions were made about where the future health benefits and cost savings accrue.


The economic model estimates that the biggest gains in reducing dementia come from interventions that raise physical activity levels from sedentary to low‑level activity, and that the gains from raising activity levels beyond a low level are considerably smaller. However, this result relies on the relative risks for dementia applied to the physical activity categories in the model. There is currently no detailed information on the dose‑dependent relationship between physical activity in mid‑life and risk of developing dementia in later life. The study by Sofi et al. (2011) used in the base case analysis provided risk ratios for the onset of cognitive decline in people with high and low‑to‑moderate levels of physical activity compared with people who were sedentary. A sensitivity analysis was done using an alternative data source that suggests that targeting the sedentary group as well as targeting the low active group will be cost saving. However, more detailed information on the association between physical activity and dementia would further improve the reliability of the model results.


The NHS Health Check programme is an opportunity to encourage people to change behavioural risk in mid‑life. The PHAC acknowledged the updating of the programme, the work to increase availability and use, as well as the aim to take the programme directly to people. The members were concerned about the focus on vascular dementia and the limited acknowledgment of risk reduction for Alzheimer's and mixed dementias. In addition the group agreed that dementia information should be given to everyone eligible for a health check, whatever their age. The existing programme includes providing information and support for people to change behavioural risk factors as well as medical support to manage disease, although the behavioural support component needs strengthening. Systematic referral to support, preventive and care services as appropriate is an essential part of the NHS Health Check programme.


The Group discussed the fact that NHS Health Check could be a suitable way to deliver hearing and vision checks. Local Authority (Public Health Functions and Entry to Premises by Local Healthwatch Representatives) Regulations 2013 set out who should receive an NHS Health Check, the assessments that should be undertaken and how the check should be done. Amending or introducing new content may need the approval of the government. The Group also noted that the Expert Scientific and Advisory Panel for NHS Health Check has developed a content review process that would have to be followed.