OBJECTIVE—To characterize the epidemiology of hypoglycemia in U.S. emergency departments.

RESEARCH DESIGN AND METHODS—We analyzed data from the 1993–2005 National Hospital Ambulatory Medical Care Survey and evaluated trends and disparities over time.

RESULTS—There were ∼5 million emergency department visits for hypoglycemia from 1993–2005, and 25% resulted in hospital admission. The visit rate per 1,000 of the diabetic population was 34 (95% CI 30–37) and did not change significantly during the study period (P = 0.70 for trend). These visit rates were higher in patients aged <45 years (n = 62) and ≥75 years (n = 54) versus those aged 45–74 years (n = 21), in female (n = 37) versus male (n = 30) patients, in black (n = 40) vs. white (n = 25) patients, and in Hispanic (n = 21) versus non-Hispanic (n = 12) patients (all P < 0.001).

CONCLUSIONS—Greater emphasis on intensive glycemic control has not resulted in increased emergency department visit rates for hypoglycemia. We identified demographic disparities, however, that merit further evaluation. The emergency department provides an important opportunity for epidemiologic study and intervention for severe hypoglycemia.

Hypoglycemia is a serious complication among patients with diabetes that has significant morbidity. While tight glycemic control remains the hallmark of decreased complications in diabetes, the barrier of hypoglycemia is a major limiting factor in the maintenance of euglycemia (1–4). Severe hypoglycemia, or episodes requiring external assistance, may occur at least once a year (5,6). Although some episodes are treated at home by administration of glucose or glucagon by family members, the most severe episodes may require treatment by paramedics or the emergency department. Emergency department visits for severe hypoglycemia represent a small percentage of the total episodes of hypoglycemia but serve as a good epidemiological marker of the complication. Additionally, severe hypoglycemia causes significant cost from ambulance and hospital expenses (7). Several population-based datasets have provided important information to characterize hypoglycemia (8–12); however, these were based on single communities and may not be generalizable. In this study, we characterized the epidemiology and national trends in emergency department visits for hypoglycemia.

We analyzed the emergency department component of the 1993–2005 National Hospital Ambulatory Medical Care Survey (NHAMCS). Our institutional review board waived review of this analysis. Methodological details are described elsewhere (13,14). Briefly, the NHAMCS is conducted annually in U.S. noninstitutional general and short-stay hospitals, which provide a representative sample of all U.S. hospitals, excluding military and long-term hospitals. Trained personnel collected data annually during randomly assigned 4-week periods in selected emergency departments. Based on the ICD-9 coding manual (15) and our chart validation (16), we identified cases based on the presence of ICD-9 codes 251.0 (hypoglycemic coma), 251.1 (other specified hypoglycemia), 251.2 (hypoglycemia, unspecified), 270.3 (leucine-induced hypoglycemia), 775.0 (hypoglycemia in an infant born to a diabetic mother), and 775.6 (neonatal hypoglycemia) in any of the three diagnosis fields. We included ICD-9 code 250.8 (diabetes with other specified manifestations), except when any of the following co-diagnoses were specifically coded: 259.8, 272.7, 681.xx, 682.xx, 686.9x, 707.xx, 709.3, 730.0–730.2, and 731.8.

We used Stata 9.0 software (StataCorp, College Station, TX) to determine nationally representative estimates with 95% CIs using assigned weights, which adjust for selection probability. We calculated rates based on Centers for Disease Control and NHAMCS data (17) and made comparisons using a two-sample test of proportions. We analyzed trends over time using the least-squares method of linear regression and a nonparametric Wilcoxon-based test for trends, where appropriate. All P values were two tailed, with P < 0.05 considered statistically significant.

Between 1993 and 2005, hypoglycemia accounted for ∼5.0 million emergency department visits in the U.S., or an average 380,000 visits/year. The most common ICD-9 codes for identification of hypoglycemia were 251.2 (83%), 250.8 (11%), and 251.0 (5%), and hypoglycemia was the primary (first-listed) diagnosis for 72% of visits.

Table 1 shows patient and hospital characteristics for emergency department visits for hypoglycemia. While the total number of visits for hypoglycemia increased during the study period (P < 0.001 for trend), the rate per diabetic population or emergency department visits did not change significantly (P = 0.70 and 0.14 for trend, respectively). Hypoglycemia visit rates per diabetic population were three times higher in younger (aged 0–44 years) and older (aged ≥75 years) patients compared with those aged 45–74 years (both P < 0.001). Visit rates per diabetic population for female, black, and Hispanic patients were higher than those for male, white, and non-Hispanic patients (all P < 0.001). Additionally, emergency department visit rates for hypoglycemia were higher in the South and Midwest than in the West (both P < 0.001). Rates per diabetic population or emergency department visits stratified by sex, race/ethnicity, and region subgroups did not change significantly over time (all P > 0.05 for trend).

This study is the first report of national and emergency department–based epidemiological data for severe hypoglycemia. Prior epidemiologic data suggest that the incidence of severe hypoglycemia has increased in the era of greater attention to tight glycemic control (2,3,8,9). Our data indicate that this increased emphasis has not changed the incidence of hypoglycemia requiring emergency care. Between 1993 and 2005, we found an increasing number of hypoglycemia visits, but the rate remained unchanged, suggesting that the increased overall prevalence of diabetes accounts for the growth in hypoglycemia visits.

However, we did find demographic disparities in emergency department visits for hypoglycemia by age, sex, race, ethnicity, and region. An increased incidence of severe hypoglycemia has been previously observed in older patients and those of lower socioeconomic status (9–12). While demographic disparities clearly exist, current data do not explain the cause for these differences, and further evaluation of these observations is warranted.

With the large number of annual visits for hypoglycemia and the majority (75%) resulting in discharge, the emergency department should be an important venue for education and referral. Indeed, a hypoglycemic event requiring an emergency care is a clinical red flag for diabetes, which requires substantial education on self-management, possible dietary or medication adjustment, avoidance of recurrent hypoglycemia, and close outpatient monitoring (18,19).

This study has a few limitations. The ICD-9 classification system is imperfect for case identification, as it was created for reimbursement rather than research purposes. However, we validated our coding algorithm and found a high level of accuracy by chart review (16). Because NHAMCS lacks individual identifiers, data pertain to emergency department visits, not individual patients. Additionally, a limitation of the NHAMCS study design is that data may be missed due to relying on medical records, the and cause of hypoglycemia is not known. Nevertheless, NHAMCS provides important epidemiologic data collected using rigorous methodology (14).

In summary, the rate of emergency department visits for hypoglycemia has not changed in the past decade; however, we identified demographic disparities that merit further study.

Table 1—

Emergency department visits for hypoglycemia according to patient and hospital characteristics, 1993–2005

nEstimated total number of cases (95% CI)Rate per 1,000 of the diabetic population (95% CI)Rate per 1,000 emergency department visits (95% CI)
Overall 1,303 4,960 (4,460–5,460) 34 (30–37) 3.7 (3.4–4.1) 
Years     
    1993–1995 215 773 (578–969) 30 (23–38) 2.8 (2.1–3.5) 
    1996–1997 152 672 (524–820) 43 (32–54) 3.6 (2.8–4.4) 
    1998–1999 141 740 (543–937) 26 (19–33) 3.6 (2.7–4.6) 
    2000–2001 222 897 (663–1,130) 40 (30–50) 4.2 (3.1–5.2) 
    2002–2003 262 776 (633–920) 32 (24–39) 3.5 (2.8–4.1) 
    2004–2005 311 1,100 (864–1,340) 33 (26–39) 4.9 (3.8–5.9) 
Age (years)     
    <45 401 1,550 (1,330–1,780) 62 (53–71) 1.7 (1.5–2.0) 
    0–19 78 359 (229–489) — 0.9 (0.6–1.2) 
    20–44 323 1,200 (1,020–1,370) — 2.3 (2.0–2.7) 
    45–64 364 1,230 (1,060–1,400) 19 (17–22) 5.5 (4.7–6.2) 
    65–74 219 845 (698–991) 25 (20–29) 10 (8.5–12) 
    ≥75 319 1,330 (1,090–1,580) 54 (44–64) 12 (9.4–14) 
Sex     
    Female 730 2,840 (2,470–3,200) 37 (32–42) 4.0 (3.5–4.5) 
    Male 573 2,120 (1,870–2,380) 30 (26–34) 3.4 (3.0–3.8) 
Race     
    White 829 3,270 (2,910–3,640) 28 (25–31) 3.6 (3.2–4.0) 
    Black 326 1,210 (970–1,450) 50 (40–60) 4.8 (3.9–5.8) 
    Other 36 104 (50–159) 14 (6.5–21) 3.4 (1.6–5.1) 
Ethnicity     
    Hispanic 100 329 (241–416) 21 (15–27)* 2.5 (1.8–3.2) 
    Non-Hispanic 985 3,720 (3,290–4,150) 12 (10–13)* 3.9 (3.5–4.4) 
Insurance     
    Private 285 1,140 (963–1,300) NA 2.6 (2.2–3.0) 
    Public 643 2,430 (2,090–2,760) NA 5.8 (5.0–6.6) 
    Self-pay 111 414 (316–513) NA 2.2 (1.7–2.7) 
    Other/unknown 124 416 (316–515) NA 2.4 (1.8–3.0) 
U.S. region     
    Northeast 319 903 (747–1,060) NA 3.4 (2.8–4.0) 
    Midwest 299 1,320 (1,080–1,570) NA 3.9 (3.2–4.6) 
    South 485 2,010 (1,630–2,380) NA 4.1 (3.3–4.8) 
    West 200 726 (551–901) NA 3.0 (2.3–3.8) 
Urban status     
    Urban 1,078 3,730 (3,270–4,190) NA 3.5 (3.1–4.0) 
    Non-urban 225 1,230 (806–1,660) NA 4.5 (3.0–6.1) 
Disposition     
    Admitted 318 1,230 (1,060–1,400) NA 7.2 (6.0–8.3) 
    Not admitted 969 3,630 (3,450–3,810) NA 3.2 (2.8–3.5) 
nEstimated total number of cases (95% CI)Rate per 1,000 of the diabetic population (95% CI)Rate per 1,000 emergency department visits (95% CI)
Overall 1,303 4,960 (4,460–5,460) 34 (30–37) 3.7 (3.4–4.1) 
Years     
    1993–1995 215 773 (578–969) 30 (23–38) 2.8 (2.1–3.5) 
    1996–1997 152 672 (524–820) 43 (32–54) 3.6 (2.8–4.4) 
    1998–1999 141 740 (543–937) 26 (19–33) 3.6 (2.7–4.6) 
    2000–2001 222 897 (663–1,130) 40 (30–50) 4.2 (3.1–5.2) 
    2002–2003 262 776 (633–920) 32 (24–39) 3.5 (2.8–4.1) 
    2004–2005 311 1,100 (864–1,340) 33 (26–39) 4.9 (3.8–5.9) 
Age (years)     
    <45 401 1,550 (1,330–1,780) 62 (53–71) 1.7 (1.5–2.0) 
    0–19 78 359 (229–489) — 0.9 (0.6–1.2) 
    20–44 323 1,200 (1,020–1,370) — 2.3 (2.0–2.7) 
    45–64 364 1,230 (1,060–1,400) 19 (17–22) 5.5 (4.7–6.2) 
    65–74 219 845 (698–991) 25 (20–29) 10 (8.5–12) 
    ≥75 319 1,330 (1,090–1,580) 54 (44–64) 12 (9.4–14) 
Sex     
    Female 730 2,840 (2,470–3,200) 37 (32–42) 4.0 (3.5–4.5) 
    Male 573 2,120 (1,870–2,380) 30 (26–34) 3.4 (3.0–3.8) 
Race     
    White 829 3,270 (2,910–3,640) 28 (25–31) 3.6 (3.2–4.0) 
    Black 326 1,210 (970–1,450) 50 (40–60) 4.8 (3.9–5.8) 
    Other 36 104 (50–159) 14 (6.5–21) 3.4 (1.6–5.1) 
Ethnicity     
    Hispanic 100 329 (241–416) 21 (15–27)* 2.5 (1.8–3.2) 
    Non-Hispanic 985 3,720 (3,290–4,150) 12 (10–13)* 3.9 (3.5–4.4) 
Insurance     
    Private 285 1,140 (963–1,300) NA 2.6 (2.2–3.0) 
    Public 643 2,430 (2,090–2,760) NA 5.8 (5.0–6.6) 
    Self-pay 111 414 (316–513) NA 2.2 (1.7–2.7) 
    Other/unknown 124 416 (316–515) NA 2.4 (1.8–3.0) 
U.S. region     
    Northeast 319 903 (747–1,060) NA 3.4 (2.8–4.0) 
    Midwest 299 1,320 (1,080–1,570) NA 3.9 (3.2–4.6) 
    South 485 2,010 (1,630–2,380) NA 4.1 (3.3–4.8) 
    West 200 726 (551–901) NA 3.0 (2.3–3.8) 
Urban status     
    Urban 1,078 3,730 (3,270–4,190) NA 3.5 (3.1–4.0) 
    Non-urban 225 1,230 (806–1,660) NA 4.5 (3.0–6.1) 
Disposition     
    Admitted 318 1,230 (1,060–1,400) NA 7.2 (6.0–8.3) 
    Not admitted 969 3,630 (3,450–3,810) NA 3.2 (2.8–3.5) 
*

Data available 1997–2005. NA, data not available.

View Large

This study was supported by an investigator-initiated grant from Novo Nordisk, Inc.

1.
The Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus.
N Engl J Med
 
329
:
977
–986,
1993
2.
DCCT Research Group: Hypoglycemia in the Diabetes Control and Complications Trial.
Diabetes
 
46
:
271
–286,
1997
3.
United Kingdom Prospective Diabetes Study Research Group: Intensive blood glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes.
Lancet
 
352
:
837
–853,
1998
4.
United Kingdom Prospective Diabetes Study Research Group: Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes.
Lancet
 
352
:
854
–865,
1998
5.
MacLeod KM, Hepburn DA, Frier BM: Frequency and morbidity of severe hypoglycaemia in insulin-treated diabetic patients.
Diabet Med
 
10
:
238
–245,
1993
6.
Laing SP, Swerdlow AJ, Slater SD, Botha JL, Buden AC, Waugh NR, Smith AW, Hill RD, Bingley PJ, Patterson CC, Qiao Z, Keen H: The British Diabetic Association Cohort Study II: cause-specific mortality inpatients with insulin-treated diabetes mellitus.
Diabet Med
 
16
:
466
–471,
1999
7.
Holstein A, Plaschke A, Egberts EH: Incidence and costs of severe hypoglycemia.
Diabetes Care
 
25
:
2109
,
2002
8.
Johnson ES, Koepsell TD, Reiber G, Stergachis A, Platt R: Increasing incidence of serious hypoglycemia in insulin users.
J Clin Epidemiol
 
55
:
253
–259,
2002
9.
Bulsara MK, Holman CDJ, Davis EA, Jones TW: The impact of a decade of changing treatment on rates of severe hypoglycemia in a population based cohort of children with type 1 diabetes.
Diabetes Care
 
27
:
2293
–2298,
2004
10.
Holstein A, Plaschke A, Egberts EH: Clinical characterization of severe hypoglycemia: a prospective population-based study.
Ex Clin Endocrinol Diabetes
 
111
:
364
–369,
2003
11.
Leese GP, Wang J, Broomhall J, Kelly P, Mardsen A, Morrison W, Frier BM, Morris AD, DARTS/MEMO Collaboration: Frequency of severe hypoglycemia requiring emergency treatment in type 1 and type 2 diabetes: a population-based study of health service resource use.
Diabetes Care
 
26
:
1176
–1180,
2003
12.
Donnelly LA, Morris AD, Frier BM, Ellis JD, Donnan PT, Durrant R, Band MM, Reekie G, Leese GP, DARTS/MEMO Collaboration: Frequency and predictors of hypoglycaemia in type 1 and insulin-treated type 2 diabetes: a population-based study.
Diabet Med
 
22
:
749
–755,
2005
13.
Nawar EW, Niska RW, Xu J: National Hospital Ambulatory Medical Care Survey: 2005 emergency department summary.
Adv Data
 
386
:
1
–32,
2007
14.
McCaig LF, McLemore T: Plan and operation of the National Hospital Ambulatory Medical Survey. Series 1. Programs and collection procedures.
Vital Health Stat
 
34
:
1
–78,
1994
15.
The International Classification of Diseases, 9th rev. Clinical Modification, 5th ed. Salt Lake City, UT, Medicode Publications,
1997
16.
Ginde AA, Blanc PB, Lieberman RM, Camargo CA: Validation of coding algorithm to improve identification of hypoglycemia visits (Abstract).
Ann Emerg Med
 
50 (Suppl.)
:
S46
,
2007
17.
National Diabetes Surveillance System: Prevalence of diabetes [article online], 2005. Available from http://www.cdc.gov/diabetes/statistics/prev/national/index.htm. Accessed 30 August
2007
18.
Cryer PE, Davis SN, Shamoon H: Hypoglycemia in diabetes.
Diabetes Care
 
26
:
1902
–1912,
2003
19.
American Diabetes Association: Standards of Medical Care in Diabetes—2007.
Diabetes Care
 
30(Suppl. 1)
:
S4
–S41,2007

Published ahead of print at http://care.diabetesjournals.org on 19 November 2007. DOI: 10.2337/dc07-1790.

A.A.G. has received investigator-initiated research funding from Bayer. C.A.C. has received support from AstraZeneca, Bayer GlaxoSmithKline, Merck, Novartis, and Novo Nordisk.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C Section 1734 solely to indicate this fact.

DIABETES CARE
2008