Excess winter deaths and illness and the health risks associated with cold homes

Many factors may influence the variation in death rates between winter and summer. This includes the weather, seasonal infections, air pollution, behavioural changes and micronutrient levels. Most studies on the subject are based on data analyses for large populations (often whole cities or regions) for which health outcomes are related to outdoor, rather than indoor, temperatures. These studies show an effect attributable to cold. Often, they also show a time‑lag of up to 2 or 3 weeks between exposure to the cold and death or disease. The Committee noted that, because cold affects death rates, this implies it also has an effect on any associated ill health. This is corroborated by hospital admission data where warmer and colder seasons are compared.

The UK has a relatively high rate of excess winter deaths, based on international comparisons that use this definition.

In some years, cold weather is not restricted to the period between December and March (officially designated as 'winter' by the Office for National Statistics). This was the case in 2013, when the daily death rate was higher than average from February to mid‑April. The cold‑related deaths that occurred in April would have been assigned to the 'non‑winter' period – so reducing the official number of excess winter deaths that year. Bearing this in mind, the Committee noted that it might be more accurate to use cold weather, rather than month, to calculate and examine excess winter deaths. On this basis, in the years without flu epidemics, cold is shown to be the most important factor contributing to a seasonal variation in death rates. Members also noted that such an approach may be useful when comparing differences between countries where cold winter weather may extend beyond the December to March period. Or when comparing differences with countries where winters routinely last for a relatively short period.

The Committee noted that 'excess winter deaths' is sometimes useful as a shorthand term. But members do not think it accurately describes all the health and wellbeing issues linked with cold and cold homes. They felt a focus on excess winter illnesses (as well as deaths) gives a more rounded picture of the risks associated with the cold. Members noted that cold‑related illnesses affect people of all ages.

Interventions to address the health effects of cold homes include: policy (such as providing free boiler replacements); services (such as local efforts to implement policy and changes to buildings and heating); and changes made by individuals. (The latter could include loft insulation, double glazing or installing more efficient boilers.)

The Committee noted that interventions to ensure homes are warm enough are usually funded by government, the energy and distribution companies and the community and voluntary sectors. Usually, most health sector costs come from identifying and engaging with people who are most at risk of health problems from the cold and helping to ensure they have access to, and receive, the necessary support. The Committee did hear of examples where, for example, insulation or boiler replacements were funded directly by health bodies. But this was unusual.

The Committee noted the importance of considering cold‑related illnesses (as well as deaths from the cold). There is a lack of evidence on the former. However, evidence does indicate that changes in home heating, insulation and temperature can have a beneficial effect on illnesses from a range of causes.

The Committee was aware of World Health Organization (WHO) findings from 1985 that there is no risk to healthy sedentary people living in accommodation with air temperatures of between 18°C and 24°C (World Health Organization health impact of low indoor temperatures). But this finding is rather old and does not state what air temperature is 'safe' for people who are not healthy. The Committee also noted that the 'comfort zone' for many people in the European Union appears to be around 21°C.

The Committee discussed the benefits of using a SAP assessment or other methods to determine the likelihood of a risk to health from cold housing. A home with a level B assessment would guarantee affordable warmth for any occupant and was therefore considered the ideal. It also noted that a B rating was easily achievable in new buildings. In addition, the Committee noted that, on average, social housing in Northern Ireland currently has a SAP C rating.

The Committee noted that homes need adequate ventilation to prevent the build‑up of radon, cigarette smoke and other potentially harmful pollutants in the home. Members also noted that this has to be balanced against the need to draught‑proof homes to keep in the warmth. In addition, members noted that good quality building and adequate ventilation help reduce the risk of damp and mould.

Much of the evidence relating to seasonal differences in death rates comes from time‑series studies. In these, the studied population acts as its own control and the usual 'confounders', such as smoking, age or gender, are less important.

The Committee noted that there was limited UK evidence on how to prevent cold‑related deaths (particularly relating to interventions).

The link between some minority ethnic groups and deprivation may mean that some of these groups are more likely to live in cold homes. Other groups, including recent immigrants from warmer climates, could also be particularly vulnerable during their first few years here. For example, they may be more likely to live in poor quality housing and they face an unusually complex energy market.

The Committee considered whether the problems associated with cold housing in urban areas were different to those in rural areas. The evidence did not show any significant difference in terms of the health impact. But this may be partly due to study difficulties caused by the dispersed nature of rural populations. The Committee did note that rural properties may be more likely to be 'off grid' and so reliant on more expensive forms of fuel. Members also noted that there may be more installation difficulties (due to difficulties accessing a property). But these issues were not considered to be exclusive to rural areas.

The evidence on issuing severe weather alerts did not demonstrate any health benefits. In addition, the guideline focuses on addressing issues related to cold homes all year round – and not just during periods of bad weather. So the Committee did not make any recommendations on the use of severe weather alerts.

PHAC noted the lack of health economics literature directly applicable to the UK. Thus a new health economic model was developed to assess the cost‑effectiveness of interventions associated with cold homes. Limited evidence on the relationship between indoor temperature and health meant that the economic model was based on a number of assumptions, as follows:

• People only had 1 health problem. (The benefits of interventions for those with multiple vulnerabilities or multiple health problems are not fully captured.)

• The severity of common mental health disorders was not considered and the model does not capture wellbeing or happiness.

• Utility values for different health states were not adjusted for age or severity, and the potential impact of adjusting age or severity was not explored.

The Committee acknowledged that the economic analysis under‑estimated the non‑health benefits from a societal perspective by focusing on energy cost savings. Members noted that housing energy efficiency improvements could also lead to savings on carbon and on social care costs. It could also lead to productivity gains by reducing sickness absence from work.

The Committee noted that the results of modelling and sensitivity analyses were uncertain. Overall, however, housing energy efficiency interventions (such as roof insulation, double‑glazing or boiler replacement) are cost‑effective compared with current practice. This is particularly true of interventions aimed at households with a low standard assessment procedure (SAP) rating or aimed at vulnerable people. In both cases, these target groups gained the greatest health benefits.

The Committee discussed the potential benefit of providing a short‑term fuel subsidy, combined with energy efficiency measures in the home. Members acknowledged that a short‑term fuel subsidy alone would not be an effective alternative to energy efficiency measures. Fuel subsidy alone was reported to be less cost‑effective than when combined with energy efficiency measures from a health perspective. However, the Committee noted that neither health nor non‑health benefits are fully captured.

The Committee noted that some groups are more vulnerable to the adverse effects of cold. Information about these people may be held by a variety of services involved in some aspect of their lives. But action to address problems is likely to be hindered by the lack of access to this information, or lack of understanding of the options available to address problems. (For example, people may not know how to obtain support to install insulation.)

Several groups are more likely to suffer from the effects of cold homes. This is either because they are more likely to live in cold homes, or because they are more susceptible to its effects. For example, although not a homogenous group, people with disabilities are more likely to live in materially disadvantaged circumstances than others. They are also more likely to need more heat. With this in mind, members expressed concern that some people with disabilities may need to use benefits intended to support their independence to ensure their home is warm enough.

The Committee discussed the fact that cold weather adversely affects homeless people (including those living in 'non‑typical' forms of accommodation such as mobile homes). However, to address their needs we would need to examine a different evidence base and this would be better addressed in a separate guideline. So this guideline does not include recommendations aimed at people who are homeless or not living in permanent structures that meet basic building control regulations.

There is an increase in deaths from almost all causes during cold weather. But cardiovascular and respiratory conditions are the key causes associated with cold weather and a cold home. Only a very small number of deaths are linked to hypothermia. Although the relative risk associated with respiratory conditions and the cold is greater than for cardiovascular disease, more people overall die of the latter – and therefore most excess winter deaths are attributable to cardiovascular disease.

In England, a relatively sharp increase in the risk of death occurs when outdoor temperatures fall to around 6°C (with some variation between the regions in England). This indicates that significantly more cold‑attributable deaths occur at a relatively higher mean temperature than on days of extreme cold. This is because, although the risks are relatively smaller when it is only moderately cold, there are more days of moderate than extreme cold.

Cold homes can have a significant effect on people's social activities. For example, they may not want to invite friends home because the house is cold, or only a small part is heated (to save money). The Committee noted that people living in cold homes frequently report that this has an effect on their daily life.

The Committee heard that services to ensure people are warm enough at home are generally patchy, both in terms of geographical coverage and duration. Lack of consistency makes it difficult for practitioners to know what type of service and support is available locally.

The Committee noted that local services to tackle cold homes in any given area may be provided by different local authorities. For instance, health and wellbeing boards are located in unitary, metropolitan borough and county councils, while housing departments are located in district, borough and city councils. In addition, members noted that hospitals and other health services may cover areas served by several local authorities. To add to what is a complicated picture, services provided by commercial or voluntary organisations have their own geographical distribution areas.

Many practitioners are already addressing the issue of cold homes (in particular, environmental health officers and housing officers). But the Committee felt that it would be particularly beneficial to encourage health practitioners to help target people whose health would benefit.

Often action to reduce the harm caused by living in a cold home is made more difficult because there is a lack of coordination of services, or a lack of understanding of who should take responsibility. The Committee noted that visitors to vulnerable households should not assume action is being taken by anyone to ensure the home is warm enough.

The Committee noted that some people may feel stigmatised by admitting that they cannot afford to heat their home properly and may try to hide this. (For example, they may put the central heating on only when expecting a scheduled visit from a health or other practitioner.)

The Committee noted the importance of using a 'trusted intermediary' to help negotiate with the range of contractors that can address the problems caused by living in a cold home. (For example, local authority officers or representatives from voluntary service organisations.) Members also noted that this is best achieved face to face.

Current policies (such as Public Health England's 2014 cold weather plan) already emphasise the need for year round planning. However, planning tends to focus on relatively short periods of severe weather. The Committee heard that, generally, health and wellbeing boards were not involved in planning all‑year‑round action to combat the more enduring ill effects of cold homes.

The Committee agreed that sustainable funding to maintain and coordinate local services is a key issue. It was given expert testimony that demonstrated the value of funding from clinical commissioning groups (see the expert papers for details). Many of the services in the examples given were also funded via national or utility company programmes such as the Green Deal and the Energy Companies Obligation (ECO). The testimony also described a simple referral route as important.

The Committee discussed the potential roll‑out of smart meters. This process will provide a contact with every householder (at least when smart meters are being fitted). But in addition, members discussed whether they could be used for remote data monitoring to identify homes using less energy than might be expected. Other possibilities, such as using these data in conjunction with telemedicine services, were also noted. The Committee noted data protection concerns related to data sharing.

The Committee noted that a range of people were likely to be involved with those at risk from cold homes. These include health and social care practitioners as well as others from the housing, advice, utility and energy sectors. Workers from the voluntary sector and carers and neighbours are also likely to be involved. Because of the complexity of the problem, members noted the importance of making all these groups aware of how living in a cold home can affect people's health and how to access services locally.

The Committee discussed the training needs of professionals installing heating, insulation and other heating‑related equipment (such as meters), in terms of supporting vulnerable and disabled people. Members agreed that quality of service played a big part in ensuring new equipment was accepted and used properly. They also noted that current training is largely restricted to safety issues.

There are many barriers to addressing cold homes. These include: a lack of awareness of the health issues; lack of local or national support (often linked to knowledge of what is available); and practical issues. (The latter could include not being able to insulate someone's loft because it is filled with their possessions.)

The Committee discussed the importance of using data effectively and storing it properly. It noted that data use issues are often cited as a barrier to implementing effective systems. Data handling was beyond the remit of this guideline. But the Committee noted that examples of good local practice do exist.

Stakeholders expressed concern about a possible increase in workload as a result of the recommendations. But the aim was to target people whose health might benefit the most. This, in turn, could help prevent ill health, so resulting in a reduced workload for the referring practitioner in the longer term.

Many stakeholders expressed concern about the availability and consistency of funding for interventions to improve cold homes. As previously noted, much of this comes from government or the energy companies. The Committee noted that funding regimes are complex and that health inequalities could arise because of the different criteria used. But the provision of national funding was outside the remit of this guideline.