Results
During the 2-year study period, a total of 18 768 hypothermia- and hyperthermiarelated health care visits were filed to Medicare. More than half (53%) of the total visits were for hyperthermia, and 47% were for hypothermia. Of the total visits, 73% were by White beneficiaries, and 68% were men.
Most (73%) of the visits were in the outpatient care setting (n = 13 717), and of these, 64% were for hyperthermia. The remaining 27% of the visits were inpatient (n = 5051), of which 63% were for hypothermia.
Hypothermia
We identified 8761 visits for hypothermia, including 359 (4%) that resulted in fatalities. The mortality rate was 0.50 per 100 000 per year. The morbidity rates for inpatient and outpatient visits were 5.29 and 6.93 per 100 000 per year, respectively (Table 1).
Age-group differences were seen among the visit types. Compared with the referent age group (65–74 years), the oldest beneficiaries (≥ 85 years) had significantly higher rates of inpatient visits (14.92 per 100 000; RR = 7.11; 95% confidence interval [CI] = 6.46, 7.82; Table 1). By contrast, outpatient visit rates were significantly higher (14.82 per 100 000) for the youngest age group (< 65 years) compared with all age groups. For both visit types, men had significantly higher rates than women, although their outpatient rate was nearly double their inpatient rate (9.31 per 100 000 vs 5.82 per 100 000; Table 1). Significant racial differences also were observed among Black and Native American beneficiaries compared with White beneficiaries, with RRs greater than 2 for both inpatient and outpatient visits (Table 1).
Geographically, the highest rates and RRs were seen in the Northeast and Midwest for inpatient visits (RR = 1.80; 95% CI = 1.66, 1.95 and RR = 1.26; 95% CI = 1.16, 1.37, respectively) as well as outpatient visits (RR = 2.68; 95% CI = 2.49, 2.89 and RR = 1.69; 95% CI = 1.56, 1.82, respectively). The states with the highest annual rates of hypothermia-related visits per 100 000 Medicare beneficiaries were Alaska (36.1), Maine (31.3), and Montana (30.4).
More than half (56%) of the hypothermiarelated visits occurred during the winter months (December-February) with January having the highest frequency of visits recorded (Figure 1). Reports of hypothermia-related visits, however, continued year-round, and approximately 100 visits occurred per month, with 14% of the total hypothermia-related visits occurring during May through September.
(Enlarge Image)
Figure 1.
Frequency of hypothermia-related inpatient and outpatient visits recorded among Medicare enrollees, 2004 and 2005.
The most common diagnosis for hypothermia- related visits was "hypothermia (accidental), excluding hypothermia not associated with low environmental temperature," ICD-9-CM code 991.6. The second most common diagnosis was frostbite of an extremity, ICD-9-CM codes 991.0 to 991.2.
Hyperthermia
A total of 10 007 visits were for hyperthermia, and of these, 42 (0.42%) resulted in death, corresponding to a mortality rate of 0.06 per 100000 per year. The hyperthermia outpatient rate was approximately 6 times the rate of inpatient visits (12.20 per 100 000 vs 1.76 per 100 000; Table 2). Rates of visit types also varied by age group, with a significantly higher inpatient rate in the oldest beneficiaries (≥ 85 years) compared with the referent age group (Table 2). An opposite trend was observed for outpatient visits; beneficiaries younger than 65 years had an outpatient visit rate of 21.29 per 100 000 per year, nearly double that for all other age categories (Table 2). The inpatient and outpatient visit rate among men was about 2 times that among women (Table 2). Black beneficiaries had significantly higher RRs for inpatient visits (RR = 1.80; 95% CI = 1.55, 2.09), and Black and Native American patients had significantly higher RRs for outpatient visits (RR = 1.59; 95% CI = 1.50, 1.69 and RR = 2.15; 95% CI = 1.73, 2.67, respectively; Table 2).
Geographically, the Southern region represented nearly 60% of the total number of hyperthermia-related visits and had the highest rates for both inpatient and outpatient visits (Table 2). The Northeast region rates were elevated, especially for outpatient visits (12.56 per 100 000) compared with the West and Midwest rates (8.34 per 100 000 and 6.64 per 100 000, respectively). The states with the highest annual rates of hyperthermia-related visits per 100 000 Medicare beneficiaries were Mississippi (38.2), Louisiana (29.5), and Oklahoma (29.3). Hyperthermia-related visits were highest during the summer months (June-August) for both years, with the number of visits doubling in 2005 (n = 5025) compared with 2004 (n = 2607). During the study period, the peak month for hyperthermia-related visits was July (Figure 2).
(Enlarge Image)
Figure 2.
Frequency of hyperthermia-related inpatient and outpatient visits recorded among Medicare enrollees, 2004 and 2005.
Most of the hyperthermia-related visits were diagnosed as "heat exhaustion, unspecified," ICD-9-CM code 992.5. Nearly half of the inpatient admissions were diagnosed as "heat exhaustion, unspecified," and 38% were diagnosed as "heatstroke and sun stroke," ICD-9-CM code 992.0.
Emergency Department Hypothermia and Hyperthermia Visits
Of the 18 768 visits in our study, 78% (n =14572) involved care in an ED. During 2004 to 2005, 6000 visits were made to the ED for hypothermia, representing a rate of 8.37 per 100 000 per year. Of the hypothermia-related visits to the ED, 55% resulted in inpatient admission. Of these hypothermiarelated admissions, 31% were among those older than 85 years, whereas almost half (42%) of the discharged patients were appreciably younger (< 65 years). During 2004 to 2005, 8572 ED visits were made for hyperthermia, representing a rate of 11.96 per 100 000 per year. Of the hyperthermia-related visits to the ED, 11% resulted in inpatient care, whereas 89% were discharged.
Medicare Costs for Hypothermia- and Hyperthermia-related Visits
Charges submitted to Medicare for direct costs and resource use associated with hypothermia- and hyperthermia-related visits for the study period totaled $134 million. Total annual costs were higher for hypothermiarelated visits (nearly $98 million) than for hyperthermia-related visits ($36 million). However, total costs for hyperthermia-related visits more than doubled from $11 million in 2004 to $25 million in 2005.
ICU costs represented 13% to 14% of the total hypothermia cost. The median length of hypothermia hospital stays was nearly 4 days (range = 1–91 days); however, the median length of skilled nursing facility stays was approximately 3 weeks (range = 1–212 days). For hyperthermia visits, the 2005 ICU costs were higher (9% of total costs) than the 2004 ICU costs (6% of total costs). The median length of hyperthermia hospital stays increased from 2 days in 2004 to 3 days in 2005. The median length of hyperthermia-related skilled nursing facility stays also increased in 2005 by 4.5 days (13 days to 17.5 days).