Background: (Papers I and II) The objectives were to first determine incidence and contributing factors to cold-related injuries in northern Sweden, both those that led to hospitalization and those that led to fatality.  (Papers III and IV) A further aim was to assess post-cooling hand-rewarming responses and effects of training in a cold environment, both on fingertip rewarming and on function of the autonomic nervous system, to evaluate if there was adaptation related to prolonged occupational cold exposure. Methods:  In a retrospective analysis, cases of accidental cold-related injury with hospital admission in northern Sweden during 2000-2007 were analyzed (Paper I).  Cases of fatal hypothermia in the same region during 1992-2008 were analyzed (Paper II).  A cohort of volunteers was studied before and after many months of occupational cold exposure. Subject hand rewarming response was measured after a cold hand immersion provocation and categorized as slow, moderate or normal in rewarming speed.  This cold provocation and rewarming assessment was performed before and after their winter training.  (Paper III).  Heart rate variability (HRV) was analyzed from the same cold provocation/recovery sequences (Paper IV). Results:  (Paper I) For the 379 cases of hospitalization for cold-related injury, annual incidences for hypothermia, frostbite, and drowning were 3.4/100,000, 1.5/100,000, and 1.0/100,000 inhabitants, respectively.  Male gender was more frequent for all categories.  Annual frequencies for hypothermia hospitalizations increased during the study period.  Hypothermia degree and distribution of cases were 20 % mild (between 32 and 35ºC), 40% moderate (31.9 to 28ºC), and 24% severe (< 28ºC), while 12% had temperatures over 35.0ºC.  (Paper II) The 207 cases of fatal hypothermia showed an annual incidence of 1.35 per 100,000 inhabitants, 72% in rural areas, 93% outdoors, 40% found within 100 meters of a building.  Paradoxical undressing was documented in 30%.  Ethanol was detected in femoral vein blood in 43%. Contributing co-morbidity was common including heart disease, previous stroke, dementia, psychiatric disease, alcoholism, and recent trauma.  (Paper III) Post-training, baseline fingertip temperatures and cold recovery variables in terms of final rewarming fingertip temperature and vasodilation time increased significantly in moderate and slow rewarmers.  Cold-related injury (frostbite) during winter training occured disproportionately more often in slow rewarmers (4 of the 5 injuries).  (Paper IV) At ‘pre- winter-training’, normal rewarmers had higher power for low frequency and high frequency heart rate variability.  After cold acclimatization (post-training), normal rewarmers showed lower resting power values for the low frequency and high frequency heart rate variability components.  Conclusions: Hypothermia and cold injury continues to cause injury and hospitalization in the northern region of Sweden.  Assessment and management is not standardized across hospitals.  With the identification of groups at high risk for fatal hypothermia, it should be possible to reduce the incidence, particularly for highest risk subjects; rural, living alone, alcohol-imbibing, and psychiatric diagnosis-carrying citizens.  Long-term cold-weather training may affect hand rewarming patters after a cold provocation, and a warmer baseline hand temperature with faster rewarming after a cold provocation may be associated with less general risk for frostbite.  Heart rate variability results support the conclusion that cold adaptation in the autonomic nervous system occurred in both groups, though the biological significance of this is not yet clear.