for health and social care image

The Information Centre logo
HESonline logo

IN THIS SITE...

spacer
Articles and research
Quick links

Met Office health forecasts for COPD patients

This article has been written and provided by Dr William Bird of the Met Office, and is subject to Crown Copyright.

The Met Office is well known for providing the weather forecasts that we see on TV and read in the newspapers. However, that is only a small part of its work. For example, it delivers specific forecasts to:

  • electricity companies, to help predict electricity demand
  • supermarkets, to help order the right food
  • airlines, to plan the most efficient flight path for the aircraft
  • the military.

One less well known area of the Met Office is the Health Forecast Unit, which provides specific forecasts to help the NHS target care to the patients when they are most vulnerable. The unit also works with hospitals to help plan workload. One of the unit's current services is a forecast for Chronic Obstructive Pulmonary Disease (COPD).

COPD

COPD is an umbrella term that includes chronic bronchitis and emphysema. Figures from the British Thoracic Society and from HES show that:

  • in the UK, 900,000 people have been diagnosed as having COPD. Half as many again are thought to be living with COPD without the disease being diagnosed, bringing the total to 1.5 million
  • there are about 30,000 deaths each year due to COPD (five per cent of all deaths)
  • globally, COPD will become the third most frequent cause of death by 2020
  • the five-year survival rate for sufferers is about 75 per cent in mild cases and 27 per cent in severe cases
  • emergency hospital admissions due to COPD have been increasing by about five per cent per year, but many of these admissions are thought to be preventable.
COPD and the weather

There are strong associations between COPD admissions and the weather. The strongest relate to cold weather (admissions peak 12 days after a cold snap) and season (worse over the New Year). Other associations are with high humidity, snow and low indoor temperatures.

A paper from the European Journal of Epidemiology (The effect of cold temperatures on patients over 65yrs presenting with chest infections - Hajat S, Bird W, Haines A, Nov 2004) shows that there is up to a 15 per cent rise in GP respiratory consultations in the elderly for every 1�C fall in outdoor temperature below 5�C. This link with cold weather is partly due to a cold nose. The Rhinovirus (common cold) replicates faster in a cold nose whereas the immune system becomes more sluggish in colder temperatures. A common cold is therefore more likely to develop which can lead to a chest infection in COPD patients.

Health forecast model

Weather is only part of the model used for COPD forecasts. The Health Forecast Unit has used 10 years of HES data to understand seasonal and regional variations in COPD. For example, they have found that COPD admissions appear five to six days earlier in cities compared to rural areas, with the very first admissions taking place in SW London. The unit also takes into consideration virus data and real-time NHS workloads.

The unit makes twice-weekly health forecasts, based on standard deviations from the annual mean. These forecasts use five bands of potential severity:

  • average
  • above average
  • high
  • very high
  • exceptional.

However, a forecast of increased risk on its own cannot help; the forecast has to trigger an action at the right time in the right person.

Identifying patient risk

To act on the right person, the unit has used HES data (to track previous hospital admissions) and GP data to split patients into four groups:

  • Non-exacerbation group: no exacerbation recorded
  • Exacerbation group: Exacerbation in the past 12 months, but no hospital admission
  • Moderate admission group: Admitted one to two times over the past 12 months
  • Frequent admission group: Admitted more than three times over the past 12 months.

Patients in the frequent admission group have an almost 60 per cent chance of admission during the following winter, compared to just a 10 per cent chance for those with one previous admission.

However, 85 per cent of COPD admissions during the winter peaks have had no recorded admission in the previous 12 months. These account for about 90 per cent of the rise in admissions during winter. This means that to reduce total hospital admissions all COPD patients must be reached, not just those at high risk.

Prevention and patient support

At the start of each winter, every patient who has been diagnosed with COPD receives a pack containing practical advice and an indoor thermometer to help reduce their risk of an exacerbation (Cold houses increase the risk of hospital admission just as much as cold outdoor temperatures). The pack gives them advice in several areas, all of which can prevent hospital admissions, including:

  • how to look out for early signs of an exacerbation
  • when to start treatment with steroids and antibiotics
  • how to keep the house warm and how to remain physically active.

During winter 2005-06, 35 primary care trusts across eight strategic health authorities are taking part in the Met Office Health Forecasting project. When the forecasts go out, about 1,000 patients in Birmingham receive telephone calls from NHS Direct staff to talk them through the checklist in their COPD pack, and many more receive calls from practice nurses. There is also an automated text service that is being used by patients in the south west.

As part of the project, the Health Forecasting Unit is collecting data to identify how people are using the forecast and whether they find the telephone interventions, text messages and COPD packs useful.

The future of health forecasting

Early results of the COPD health forecasting are promising, with many patients staying at home over winter rather than being frequently admitted to hospital.

Looking to the future, there are plans to produce forecasts for heart failure, asthma and even strokes. The weather forecast is really changing.


Copyright � 2005-2012, The Health and Social Care Information Centre. All Rights Reserved.