Papers

Urban hypothermia in the west of Scotland

^a Department of Accident and Emergency Medicine, Royal Alexandra Hospital, Paisley, Renfrewshire PA2 9PL, ^b Department of Accident and Emergency Medicine, Royal Infirmary, Edinburgh EH3 9YW, ^c Department of Accident and Emergency Medicine, Western Infirmary, Glasgow G11 6NT, ^d Department of Accident and Emergency Medicine, Glasgow Royal Infirmary, Glasgow G4 0SF, ^e Department of Accident and Emergency Medicine, Monklands District General Hospital, Airdrie, Lanarkshire ML6 0JS, ^f Department of Accident and Emergency Medicine, Victoria Infirmary, Glasgow G42 9TY, ^g Department of Accident and Emergency Medicine, Southern General Hospital, Glasgow G51 4TF

Correspondence to: Mr Wyatt.

To our knowledge, the epidemiology of hypothermia in Britain is unknown: there have been no studies in a large population. Concerns about risks of hypothermia in socially isolated elderly people have resulted in calls for efforts aimed at prevention.¹ Controversy about the increased cost of fuel have added to these concerns.²

Patients, methods, and results

Patients presenting to hospital in Glasgow, Paisley, Coatbridge, or Airdrie with rectal temperatures below 35°C were studied prospectively during the winter of 1993-4 (1 December to 31 March). All eight accident and emergency departments serving a continuous, defined urban population participated. Accident and emergency staff and ambulance crews collected data. Patients' relatives, neighbours, and general practitioners and social services departments provided additional information. Patients were followed up to discharge, transfer to long stay wards, or death.

Ninety three patients presented with hypothermia. Given that the study population was 1300000,³ the rate of presentation to hospital was 1 per 14000 people per winter. Although patients presented throughout the 17 weeks of the study, the rate of presentation varied according to ambient temperatures, with 12 patients presenting during the week with the lowest temperatures.

Fifty nine patients were mildly hypothermic (core temperature > 32.2°C): 12 (20%) died. Thirty four patients presented with severe hypothermia (temperature </= 32.2°C): 17 (50%) died. Sixty nine patients were found inside and had different characteristics from those found outside (table).

Characteristics of patients admitted for hypothermia. Values are
numbers (percentages) unless stated otherwise
----------------------------------------------------------------------------
                                Found        Found
                                inside       outside        Overall
                                (n=69)       (n=24)         (n=93)
----------------------------------------------------------------------------
Men                           29 (42)       16 (67)*       45 (48)
Mean age (range) (years)      76 (40-97)    52 (13-89)+    70 (13-97)
Severe hypothermia            28 (41)        6 (25)        34 (37)
Died                          27 (39)        2 (8)*        29 (31)

Implicated causes:
  Alcohol                     20 (29)       19 (79)*       39 (42)
  Fall or immobility          18 (26)        3 (13)        21 (23)
  Confusion                   15 (22)        2 (8)         17 (18)
  Infection                    9 (13)        1 (4)         10 (11)
  Cerebrovascular accident     8 (12)        0              8 (9)
  Diabetes                     4 (6)         1 (4)          5 (5)
  Self neglect                 3 (4)         0              3 (3)
  Immersion in river Clyde     0             2 (8)          2 (2)
  No obvious cause             5 (7)         1 (4)          6 (6)
----------------------------------------------------------------------------
*Proportions significantly different between patients found inside and those
found outside, according to 2 test (P<0.05).
+Proportions significantly different between patients found inside and those
found outside according to Student's t test (P<0.05).

Sixty six of the 69 patients found inside were in their own homes. Fifty three lived alone and 27 did not have a regular daily visitor. Forty two patients had not been seen for over 24 hours. Details of heating were recorded for 48 of the 53 patients living alone: 28 had central heating, 18 had other heating, and two had no heating. Two thirds of the patients with heating (31/46) had it switched off when they were found.

Comment

The true incidence of hypothermia is difficult to define. Only 300 people are officially recorded as dying of hypothermia in England and Wales annually.⁴ Extrapolation of our results for the whole of Britain yields 4000 hospital admissions, with over 1000 deaths from hypothermia. Any extrapolation may be a considerable underestimate: many elderly people found dead at home may have died of hypothermia.

Patients found hypothermic outside differed from those found inside. Those found inside comprised a large number of socially isolated elderly people who had been found hypothermic in their own homes: mortality was high. The importance of regular daily visits by someone (family, neighbour, or someone from the social services department) to elderly people living alone has been previously noted but deserves reemphasis.¹ Isolated elderly people with associated problems (poor mobility, confusion, self neglect, and various medical conditions) are at greatest risk of hypothermia. Such patients could be identified from general practitioner records and be targeted for special attention, which could include an assessment to check whether extra social support is required.

Excess winter mortality in Britain is greatest among socially deprived people with the worst heating.⁵ In this study most of those who became hypothermic at home had heating available but were not using it. One reason why elderly people may not use heating during cold weather is the cost, or perceived cost, of fuel. The now abandoned government proposals to add full rate value added tax (VAT) to fuel were met with concern by doctors.² It remains a worry that future increases in fuel bills may influence elderly people to use heating even more sparingly, placing even more of them at risk of hypothermia.

We thank members of the Scottish Ambulance Service for collecting data for this study. We also wish to thank the following consultants in accident and emergency medicine for allowing us to study patients being treated in their departments: Mr C A Allister, Mr I W R Anderson, Dr M Brookes, Mr M H Calvert, Mr R Crawford, Mr N V Doraiswamy, Mr J Findlay, Mr M W G Gordon, Mr P T Grant, Mr D A W Ritchie, Mr R G Simpson, Mr I Swann, and Mr W M Tullett.

Funding: None.

Conflict of interest: None.

1. Otty CJ, Roland MO. Hypothermia in the elderly: scope for prevention. BMJ 1987;295:419-20.
2. Watt GCM. Health implications of putting value added tax on fuel. BMJ 1994;309:1030-1. [Free Full Text]
3. Office of Population Censuses and Surveys. Population census. London: HMSO, 1991.
4. Deaths in winter. Lancet 1985;ii:987-8.
5. Wilmhurst P. Temperature and cardiovascular mortality. BMJ 1994;309: 1029-30.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    StumbleUpon    Technorati    What's this?

Relevant Article

The management of accidental hypothermia

W G Headdon, P M Wilson, and Harry R Dalton
BMJ 2009 338: b2085. [Extract] [Full Text]

This article has been cited by other articles:

• Headdon, W G, Wilson, P M, Dalton, H. R (2009). The management of accidental hypothermia. BMJ 338: b2085-b2085 [Full text]  
• Van Mieghem, C., Sabbe, M., Knockaert, D. (2004). The Clinical Value of the ECG in Noncardiac Conditions. Chest 125: 1561-1576 [Abstract] [Full text]  
• Biem, J., Koehncke, N., Classen, D., Dosman, J. (2003). Out of the cold: management of hypothermia and frostbite. CMAJ 168: 305-311 [Full text]  
• Mallet, M.L. (2002). Pathophysiology of accidental hypothermia. QJM 95: 775-785 [Abstract] [Full text]  
• Wyatt, J, Guly, H (2002). Identifying the research question and planning the project. Emerg. Med. J. 19: 318-321 [Abstract] [Full text]  
• McInerney, J J, Breakell, A, Madira, W, Davies, T G, Evans, P A (2002). Accidental hypothermia and active rewarming: the metabolic and inflammatory changes observed above and below 32{degrees}C. Emerg. Med. J. 19: 219-223 [Abstract] [Full text]  
• Taylor, A J, McGwin, G Jr, Davis, G G, Brissie, R M, Holley, T D, Rue, L W III (2001). Hypothermia deaths in Jefferson County, Alabama. Inj. Prev. 7: 141-145 [Abstract] [Full text]  
• Morgan, R., King, D., Blair, A. (1996). Urban hypothermia. BMJ 312: 124a-124 [Full text]  
• Sloan, D S G (1996). Number of excess deaths during winter is large. BMJ 312: 124b-124 [Full text]