gram participants (P < .0001) compared with a group of patients with diabetes in A culturally tailored diabetes lifestyle intervention delivered by …
The Diabetes Educator
http://tdesagepubcom Hospitalization and Discharge Education of Emergency Department Patients With Hypoglycemia
Adit A Ginde, Daniel J Pallin and Carlos A Camargo, Jr The Diabetes Educator 2008; 34; 683 DOI: 101177/0145721708321022 The online version of this article can be found at: http://tdesagepubcom/cgi/content/abstract/34/4/683
Published by:
http://wwwsagepublicationscom
On behalf of:
American Association of Diabetes Educators
Additional services and information for The Diabetes Educator can be found at: Email Alerts: http://tdesagepubcom/cgi/alerts Subscriptions: http://tdesagepubcom/subscriptions Reprints: http://wwwsagepubcom/journalsReprintsnav Permissions: http://wwwsagepubcom/journalsPermissionsnav Citations this article cites 27 articles hosted on the SAGE Journals Online and HighWire Press platforms: http://tdesagepubcom
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
Education of Patients With Hypoglycemia
683
Hospitalization and
Discharge Education of Emergency Department Patients With Hypoglycemia
Purpose
The purpose of this study is to evaluate the content and adequacy of emergency department ED discharge instructions and factors associated with hospitalization in patients presenting with hypoglycemia
Adit A Ginde, MD, MPH Daniel J Pallin, MD, MPH Carlos A Camargo, Jr, MD, DrPH
From the Department of Emergency Medicine, University of Colorado Health Sciences Center, Aurora, Colorado Dr Ginde, Department of Emergency Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts Dr Ginde, Department of Emergency Medicine, Brigham and Womens Hospital, Boston, Massachusetts Dr Pallin, and Department of Emergency Medicine, Massachusetts General Hospital, Boston, Massachusetts Dr Camargo
Methods
This is a retrospective cohort study at 3 adult EDs A 1-year consecutive sample of hypoglycemia cases were identified using ICD-9-CM codes and were confirmed by chart review Clinical variables and written discharge instructions were analyzed by chart abstraction
Results
Six hundred thirty-six charts of patients with possible hypoglycemia were reviewed, of which 436 64 hypoglycemia cases were confirmed The
median age was 64 Hypoglycemia was associated with sulfonylurea use for 78 16 patients and insulin alone for 286 65 patients Written discharge instructions advised frequent blood glucose checks in 21 of patients and medication dose adjustment in 27 of patients and rarely recommended avoiding recurrent hypoglycemia 3, checking glucose before driving 04, or obtaining glucagon emergency kits 2 Hospitalization resulted from 177 41 visits and was associated with older age age 65-74 [odds ratio 57] and age 75 [odds ratio 79], sulfonylurea use odds ratio 35, 3 hypoglycemic episodes odds ratio 31, no documented diabetes medications
Correspondence to Adit A Ginde, MD, MPH, University of Colorado Health Sciences Center, Department of Emergency Medicine, Leprino Office Building, 7th Floor, 12401 E 17th Avenue, B-215, Aurora, CO 80045 aditginde@uchscedu
Acknowledgments: This study was funded by an investigator-initiated grant from Novo Nordisk, Inc Dr Ginde was supported by the Emergency Medicine Foundation Research Fellowship grant The authors thank Rebecca Lieberman, Roberta Capp, and Philip Blanc for their assistance with data collection
DOI: 101177/0145721708321022
Ginde et
al
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
The Diabetes EDUCATOR
684
odds ratio 21, current primary care provider odds ratio 42, and hypoglycemia as a secondary diagnosis odds ratio 47
Conclusions
ED written discharge instructions appeared inadequate in providing recommended education for patients with severe hypoglycemia Older age and sulfonylurea use were independently associated with hospital admission Although hypoglycemia is generally considered a selflimited condition, 2 of every 5 patients required hospitalization, which likely reflects an older and more complex patient population
and discharging the patient Sometimes a patient may require additional observation or hospital admission The factors associated with this prolonged observation are unknown Additionally, a hypoglycemic event requiring an ED visit often requires substantial education and outpatient referral The current content of discharge instructions after an ED visit for hypoglycemia has not
been reported, including advice on avoidance of recurrent hypoglycemia, medication adjustment, and close outpatient follow-up The purpose of this study is to evaluate the content and adequacy of ED discharge instructions and factors associated with hospitalization in patients presenting with hypoglycemia
Materials and Methods
ypoglycemia is a common complication of diabetes therapy and can have a profound impact on quality of life1 Although tight glycemic control is a hallmark of lower rates of complications, the barrier of hypoglycemia is the major limiting factor in maintaining the glycemic control necessary for improved clinical outcomes2-5 Mild or self-treated episodes of hypoglycemia are common, especially in type 1 diabetes, with reported rates of 2 episodes per week6 Severe hypoglycemia or episodes requiring external assistance may occur at least once a year and are a significant cause of morbidity6-8 Although some episodes of severe hypoglycemia are treated at home by administration of oral glucose or parenteral glucagon by family members, the most severe episodes require ambulance or emergency department ED visits9 Although ED visits for severe hypoglycemia are a small
percentage of the total episodes of hypoglycemia in diabetes, they serve as a good epidemiological marker of the complication and result in significant economic and psychological costs10 Most data on the incidence and distribution of hypoglycemia are based on highly selected patients in the setting of large randomized clinical trials2-4,8 and may not generalize to the entire population An ED visit for hypoglycemia is an important event for its significance to the patient, the cost to the system, and the opportunity for education and intervention The ED visit often consists of restoring normoglycemia, ensuring safe maintenance of normoglycemia,
H
Study Design
This is a multicenter, retrospective cohort study using a structured medical record review of ED patients presenting with hypoglycemia Institutional Review Board approval was obtained with waiver of informed consent at each site
Study Setting and Population
This study was conducted at 3 US academic hospitals that are active participants in the Emergency Medicine Network wwwemnet-usaorg The EDs have a combined annual visit volume of 175 000 adults and are staffed by emergency medicine, internal medicine, and surgery
residents; patient care is supervised by attending emergency physicians 24 h/d The electronic medical record systems at each site were searched for the following International Classification of Diseases, Ninth Revision, Clinical Modification ICD-9CM codes in any diagnosis field, which were used to identify possible hypoglycemia visits: 2503 diabetes with other coma, 2508 diabetes with other specified manifestations 2510 hypoglycemic coma, 2511 other specified hypoglycemia, 2512 hypoglycemia, unspecified, 2703 leucine-induced hypoglycemia, 7750 hypoglycemia in an infant born to a diabetic mother, 7756 neonatal hypoglycemia, and 9623 poisoning by insulin and antidiabetic agents Only ED-based admission codes were examined, to avoid inclusion of incident hypoglycemia that occurred during inpatient
Volume 34, Number 4, July/August 2008
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
Education of Patients With Hypoglycemia
685
hospitalization This strategy was based on detailed
examination of the ICD-9-CM coding manual,11 review of the experience from previously reported approaches,12-16 and discussion with coding experts For chart validation of hypoglycemia, all ED visits with candidate ICD-9-CM codes between July 1, 2005, and June 30, 2006, were identified at each site, and written ED charts were obtained For patients with multiple ED visits during the data collection period, only the first visit was included to avoid overrepresentation of demographic data by certain patients To enhance the reliability of our chart review, only charts with complete ED nursing notes and emergency physician notes were abstracted, and all other charts were considered incomplete The number of hypoglycemia visits for all nonabstracted visits was estimated ie, incomplete records and subsequent visits by same patient using the authors new ICD-9-CM coding algorithm17 As ED charts were reviewed, cases of hypoglycemia were confirmed based on the following criteria: 1 any documented prehospital or ED glucose value serum or capillary 70 mg/dL 39 mmol/L or 2 charted physician discharge diagnosis of hypoglycemia The glucose threshold was based on the consensus recommendation of the
American Diabetes Association Workgroup on Hypoglycemia18 Physician diagnosis of hypoglycemia was used to include clinically diagnosed cases for which hypoglycemia resolved prior to first blood glucose determination eg, patients who received oral or parenteral glucose for symptoms consistent with hypoglycemia prior to blood glucose determination and whose symptoms improved after treatment
Study Protocol
physician documentation Data not documented in the written or electronic medical record were presumed absent Patient characteristics included demographics; type 1 or type 2 diabetes, if specified; and current home diabetes medications insulin, sulfonylurea, and/or other oral hypoglycemics Patients with documentation of noninsulin-dependent diabetes or oral hypoglycemic use were considered to have type 2 diabetes, but diabetes type could not be identified for patients with documentation of insulin-dependent diabetes Visit characteristics were documented, including mode of arrival, any prehospital glucose values, prehospital treatment, ED glucose values, and ED treatment The prehospital setting was defined as home, outpatient, or ambulance events within 4 hours of ED presentation
Additionally, the presumed cause of the hypoglycemic episode, ED length of stay for discharged or observation patients, and patient disposition were recorded, if specifically documented Hospital admission was confirmed by electronic records and verified during chart review from physician and nursing documentation For nonadmitted patients, written discharge instructions for documented recommendations and follow-up care were evaluated There were no standardized written discharge instructions for hypoglycemia at any of the studied institutions Finally, EDderived ICD-9-CM codes were evaluated, from electronic medical records, to determine whether hypoglycemia was the primary first-listed diagnosis or a secondary diagnosis
Data Analysis
Three paid research assistants were taught data abstraction using 10 training charts, and their performance was monitored throughout the data collection Using a standardized data abstraction form, trained reviewers performed a detailed chart review for all confirmed cases of hypoglycemia, and the research group met weekly to maintain consistency in data collection and resolve disputes The primary reviewers were blinded to the study hypotheses
Additionally, 2 reviewers independently abstracted a 10 convenience sample of charts to evaluate interrater agreement in data collection Patient and visit-related characteristics were recorded, based on ED electronic records and written nursing and
Statistical analysis was performed using Stata 90 College Station, Tex, and data were summarized using basic descriptive statistics with 95 confidence intervals CI Continuous variables age and glucose values were presented using summary measures and further stratified into clinically meaningful categories Interrater agreement was measured for chart abstraction by calculating the statistic for the subgroup of double-abstracted charts Crude associations of specified patient and clinical variables with the outcome of hospital admission were measured using the chi-square test Variables with P 10 on univariate analysis were entered into a backwardselection, multivariate logistic regression model to determine independent associations with the primary outcome of hospital admission Age and glucose were analyzed as categorical variables, based on a priori grouping, to
Ginde et al
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER
HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
The Diabetes EDUCATOR
686
174,134 ED visits
172,995 visits excluded
1,139 candidate visits based on ICD-9-CM
238 recurrent visits
901 unique patient-visits
157 hypoglycemia presumed
81 not hypoglycemia
679 abstracted
222 incomplete
436 hypoglycemia confirmed
243 not hypoglycemia
170 hypoglycemia presumed
52 not hypoglycemia
Figure 1 Identification and chart abstraction of emergency department ED visits for hypoglycemia ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification Based on detailed chart review and ICD-9-CM coding algorithm
enhance interpretability Reference groups for categorical variables were chosen based on a prevalent, lower risk subgroup to facilitate comparison There were no interaction terms of clinical interest, so to avoid multiple testing, interaction terms were not used in the model Only variables with P 05 were retained in the final model, unless they significantly confounded associations of other retained variables
Results of regression are presented as odds ratios OR with 95 CIs Goodness of fit for the final model was evaluated using the HosmerLemeshow test
Results
Of the 174 134 ED visits at the 3 institutions during the data collection period, 901 patients with candidate ICD-9-CM codes were identified These patients accounted for 1139 visits with possible hypoglycemia ie, 05 of all ED visits Complete documentation was available for 679 75 of first visits by included patients Distribution of candidate ICD-9-CM codes and demographic data age, sex, and race/ethnicity were similar among abstracted and incomplete/missing charts data not shown
The initial chart review confirmed 436 hypoglycemia visits Diabetic ulcers and cellulitis identified by ICD-9CM code 2508 diabetes with other manifestations were the most common nonhypoglycemia diagnoses among flagged charts Confirmation of hypoglycemia had a very high interrater agreement 097, and data from detailed chart review were similarly reliable 092 Figure 1 displays confirmed and presumed cases of hypoglycemia based on detailed chart review and the authors new ICD-9-CM coding algorithm unpublished data, respectively An estimated 763
hypoglycemia visits 436 confirmed, 327 presumed occurred during the study, which represents 04 of total ED visits during the 1-year study period
Characteristics of Study Subjects
Patient characteristics for the 436 abstracted hypoglycemia visits are summarized in Table 1 The median age was 64 interquartile range [IQR] 46-76 Table 2 presents the clinical characteristics of the hypoglycemia visits The median initial prehospital glucose was 40 mg/dL 22 mmol/L [IQR 28-57], and initial ED glucose was 87 mg/dL 48 mmol/L [IQR 52-144] The most commonly
Volume 34, Number 4, July/August 2008
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
Education of Patients With Hypoglycemia
687
Table 1
Patient Characteristics of 436 Emergency Department Visits for Hypoglycemia
Total N 436 Variable Age 45 45-64 65-74 75 Sex Female Male Race/ethnicity White Black Hispanic Asian Other Insurance Medicare only Medicaid with or without Medicare Any private insurance No insurance Primary care
provider Yes No Diabetes Type 1 Type 2 Unspecified diabetes Not documented Diabetes medication Insulin only Sulfonylurea only Insulin sulfonylurea Neither documented
CI, confidence interval
Admitted n 177 n 95 CI n
Discharged/Other n 259 95 CI
n
95 CI
95 126 99 116 216 220 278 101 24 17 16 122 143 155 15 390 46 41 144 201 50 286 61 17 72
22 18-26 29 25-33 23 19-27 27 22-31 50 45-54 50 46-55 64 59-68 23 19-27 6 3-8 4 2-6 4 2-5 28 24-32 33 28-37 36 31-40 3 2-5 89 87-92 11 8-13 9 7-12 33 29-37 46 41-51 11 8-14 65 61-70 14 11-18 4 2-6 17 13-20
20 32 52 73 95 82 115 35 16 5 6 69 65 41 2 169 8 9 80 66 22 89 42 12 34
21 13-31 25 18-34 53 42-63 63 53-72 44 37-51 37 31-44 41 36-47 35 25-45 67 45-84 29 10-56 38 15-65 57 47-66 45 37-54 26 20-34 13 2-40 43 38-48 17 8-31 22 11-38 56 47-64 33 26-40 44 30-59 31 26-37 69 56-80 71 44-90 47 35-59
75 94 47 43 121 138 163 66 8 12 10 53 78 114 13 221 38 32 64 135 28 197 19 5 38
79 69-87 75 66-82 47 58-37 37 47-28 56 49-63 63 56-69 59 53-64 65 55-75 33 16-55 71 44-90 72 35-85 43 34-53 55 46-63 74 66-80 87 60-98 57 52-62 83 69-92 78 62-89 44 36-53 67 60-74 56 41-70 69 63-74 31 20-44 29 10-56 53 41-65
Ginde et al
Downloaded from
http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
The Diabetes EDUCATOR
688
Table 2
Clinical Characteristics for Emergency Department ED Visits Wit 
h Hypoglycemia
Total N 436 Variable Mode of arrival Ambulance Walk-in Not documented Prehospital glucose 50 mg/dL 50-69 mg/dL 70-99 mg/dL 100 mg/dL NA/not documented Prehospital treatment IV dextrose Glucagon Oral glucose Initial ED glucose 50 mg/dL 50-69 mg/dL 70-99 mg/dL 100 mg/dL None documented Total ED capillary glucose values 1 2 3 4 None documented Total glucose values 70 mg/dLa 1 2 3 None documented ED treatment IV dextrose Octreotide Glucagon Oral glucose ICD-9-CM code position First Second or later n 298 86 52 226 56 17 17 120 151 38 153 91 79 57 174 35 103 108 70 114 41 266 86 35 49 183 13 1 226 319 117 95 CI 68 64-73 20 16-24 12 9-15 52 47-57 13 10-16 4 2-6 4 2-6 28 23-32 35 30-39 9 6-12 35 31-40 21 17-25 18 15-22 13 10-17 40 35-45 8 6-11 24 20-28 25 21-29 16 13-20 26 22-31 9 7-13 61 56-66 20 16-24 8 6-11 11 8-15 42 37-47 3 2-5 0
0-1 52 47-57 73 69-77 27 23-31 n 121 32 24 88 20 4 5 60 67 17 41 39 43 27 54 14 31 45 31 54 16 103 39 22 13 94 8 1 76 103 74
Admitted n 177 95 CI 41 35-46 37 27-48 46 32-61 39 33-46 36 23-50 24 7-50 29 10-56 50 41-59 44 36-53 45 29-62 27 20-35 43 33-54 54 43-66 47 34-61 31 24-38 40 24-58 30 21-40 42 32-52 44 32-57 47 38-57 39 24-55 39 33-45 45 35-56 63 45-79 27 15-41 51 44-59 62 32-86 100 34 27-40 32 27-38 63 54-72 n
Discharged/Other n 259 95 CI 59 54-65 63 52-73 54 39-68 61 54-67 64 50-77 76 50-93 71 44-90 50 41-59 56 47-64 55 38-71 73 65-80 57 46-67 46 34-57 53 39-66 69 62-76 60 42-76 70 60-79 58 48-68 56 43-68 53 43-62 61 45-76 61 55-67 55 44-65 37 21-55 73 59-85 49 41-56 38 14-68 0 66 60-73 68 62-73 37 28-46
177 54 28 138 36 13 12 60 84 21 112 52 36 30 120 21 72 63 39 60 25 163 47 13 36 89 5 0 150 216 43
CI, confidence interval; NA, not applicable; IV, intravenous; ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification a Prehospital or emergency department
Volume 34, Number 4, July/August 2008
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of
Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
Education of Patients With Hypoglycemia
689
charted causes of hypoglycemia were missed or delayed meal 162 [37], new or changed dose of diabetes medication 56 [13, 95 CI], recent illness 16 [4], and alcohol use 11 [3] Etiology of hypoglycemia was unknown or not charted for 196 visits 45
Discharge Education
Table 3
Variables Independently Associated With Hospital Admission in Emergency Department ED Patients With Hypoglycemia
Variable Age reference: age 45 y 45-64 y 65-74 y 75 y Presence of PCP Type 2 diabetes Home diabetes medications reference: insulin only Sulfonylurea agents Other oral hypoglycemic agents None documented 3 prehospital/ED glucose values 70 mg/dL reference: 1 glucose value 70 mg/dL Hypoglycemia charted as secondary diagnosisa OR 95 CI
Among the 259 nonadmitted patients, the median ED length of stay was 55 hours IQR 37-86 Written ED discharge instructions were located for 214 83 nonadmitted patients Written instructions indicated that patients were usually advised to follow up with a primary care physician and/or endocrinologist 202 [94]
but were inconsistently advised to check blood glucose frequently 46 [21] or to consider medication dose adjustment 57 [27] Additionally, emergency physicians rarely recommended avoiding recurrent hypoglycemia to improve glucose counterregulation 7 [3], checking glucose before driving 1 [04], or obtaining a glucagon emergency kit 4 [2]
Clinical Data
15 07-32 57 27-122 79 36-174 42 16-111 08 04-15
35 15-70 35 09-131 21 11-40 31 13-74
Of the 436 abstracted visits, 177 41 resulted in hospital admission and 202 46 resulted in primary ED discharge; 44 10 visits resulted in patients being placed on ED observation status prior to discharge; and in 13 3 visits, patients left against medical advice Hypoglycemia was the primary diagnosis for 319 73 cases When hypoglycemia was a secondary diagnosis, the most common first-listed diagnoses were infection n 23, 20, syncope/presyncope n 19, 16, trauma n 12, 10, and renal insufficiency n 11, 9 Table 3 summarizes variables independently associated with hospital admission in the logistic regression model Hosmer-Lemeshow testing indicated appropriate goodness of fit for the logistic regression model P 13
47 28-81
OR, odds ratio; CI,
confidence interval; PCP, primary care provider Table is based on a multivariate logistic regression model a Most common other primary diagnoses were infection, syncope, trauma, and renal insufficiency
Discussion
To the authors knowledge, this is the largest study of ED visits for hypoglycemia Hypoglycemia was a relatively common presentation, accounting for an estimated 04 of all ED visits, which would translate to approximately 450 000 ED visits annually in the United States This reflects the increased focus on intensive glucose control in diabetes care, which reduces the rate of long-term complications but increases the rate of hypoglycemia3,8,12 Prior studies, mainly in the setting of clinical trials or longitudinal cohort studies, captured a wider array of
Ginde et al
hypoglycemia visits, including patients treated outside the ED2-6,8,10,12 Contrary to the patients in those studies, who were younger and predominantly had type 1 diabetes, half of our patients were age 65 or older and many had documented type 2 diabetes Additionally, two thirds of hypoglycemia patients were taking insulin only, whereas one third were not; this suggests that oral hypoglycemic use comprises an
important subgroup in evaluation of hypoglycemia in ED patients The rate of hospital admission in our analysis 41 was higher than anticipated Brackenridge et al19 found that only 11 of patients with hypoglycemia in a UK accident and emergency department were admitted, although the patient population was younger mean age 526 and predominantly insulin-treated 93 The data support our hypothesis that older age and sulfonylurea use are associated with higher odds of hospitalization Older
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
The Diabetes EDUCATOR
690
patients are at higher risk from hypoglycemia and additionally have more medical comorbidities, both of which likely drive the increased admission rate Sulfonylureaassociated hypoglycemia often requires prolonged observation, given the risk for recurrent episodes of hypoglycemia20 Accordingly, 69 of patients taking sulfonylurea agents were hospitalized Furthermore, recurrent hypoglycemia 3 documented prehospital or ED
episodes was independently associated with admission Although admission rates as high as 95 have been reported for hypoglycemia visits in patients with type 2 diabetes,10 diabetes type was not independently associated with admission in this analysis Although the crude admission rate was higher in patients with type 2 diabetes compared with others 56 vs 33, respectively, this association was not statistically significant when other factors such as age and sulfonylurea use were controlled These findings are limited, however, by the inability to determine diabetes type based on chart review in half of the patients When hypoglycemia was a secondary diagnosis, the odds of admission were 47 times higher The most likely explanation is that hypoglycemia can be caused by or result in other conditions that more likely prompt admission For instance, in these cases, the first-listed diagnoses were most often infection, syncope, trauma, or renal insufficiency, which were associated with higher rates of admission Patients with primary care providers had 4-fold higher odds of admission A possible explanation for this is that patients without a source of primary care may present for less severe
episodes of hypoglycemia, whereas those with primary care may be more likely to present for more severe or complicated episodes Patients without documentation of diabetes medications were twice as likely to be admitted as those on insulin Although some of these charts may lack documentation of actual diabetes medications, other patients may also have had hypoglycemia caused by other problems mandating admission eg, infection or renal insufficiency These potential explanations need further study Identifying the cause of hypoglycemia is an important component of management and prevention Half of the charts indicated a missed meal or a change in diabetes medication as the documented cause of hypoglycemia These causes are amenable to education and intervention to reduce of the risk of recurrent hypoglycemia Nearly half of the charts, however, lacked any documentation regarding cause or unknown cause If charting reflected actual medical decision making for these patients, a significant opportunity may be missed to identify and correct causes
of hypoglycemia Conversely, if care were more complete than documentation suggested, education about charting could mitigate medicolegal risk An
important component of care for patients with hypoglycemia should be education and careful instructions to help avoid recurrence of and morbidity from hypoglycemia21,22 Data from the outpatient setting suggest that educational intervention causes a sustained reduction in the incidence of severe hypoglycemia23 Our results reveal that emergency physicians provided inadequate written discharge instructions Although most patients were appropriately referred for outpatient follow-up, the importance of close blood glucose monitoring and avoidance of hypoglycemia was not documented in most discharge instructions Hypoglycemic autonomic failure may occur in patients with recent hypoglycemia and can lead to defective glucose counterregulation and hypoglycemic unawareness5,24,25 Prior studies suggest that 2 to 3 weeks of scrupulous avoidance of hypoglycemia reverses hypoglycemic unawareness and improves glucose counterregulation, which reduces risk of recurrent and severe episodes of hypoglycemia26-28 National guidelines recommend that all patients at risk for hypoglycemia be advised to check blood glucose before driving,21-22 but this was rarely documented in our sample of ED visits
Additionally, glucagon emergency kits were discussed in only 2 of discharge instructions These devices are recommended for any patient with an episode of severe hypoglycemia,22 in the same way that home epinephrine kits are recommended for patients with severe allergic reactions Discussion of glucagon kits has not been incorporated in ED practice, but the ED may be an important venue to emphasize their value Patient education is an important component of care for all diabetic patients with hypoglycemia and is an area for substantial improvement Template discharge handouts could easily mitigate the deficiencies that were observed The ideal content of these instructions to improve patient outcomes requires further study This study has some potential limitations The characteristics of hypoglycemia visits were based on ED visits at 3 academic medical centers in the United States and may not generalize to other geographic areas The case definition allowed identification only of those patients whose ICD-9-CM codes noted hypoglycemia Additionally, recurrent hypoglycemia visits and 25 of eligible charts with missing or incomplete data were excluded Although this lowered the number of total
cases identified, the distribution of ICD-9-CM codes and demographic characteristics were similar for abstracted and nonabstracted charts,
Volume 34, Number 4, July/August 2008
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
Education of Patients With Hypoglycemia
691
and the likelihood of biased estimates was small The accuracy of case validation and abstracted data depended on chart review, which is limited by the possibility of missing, incomplete, or unreliable information For example, this limited the ability to determine diabetes type, because terms such as insulin-dependent diabetes could not be used to reliably classify patients as having type 1 or type 2 diabetes Standardized definitions and training of abstractors limited the potential for bias, and high interrater agreement demonstrated internal reliability of the chart review Additionally, the content of verbal instructions could not be evaluated, which limited the ability to fully assess the adequacy of
discharge instructions
8 9
10 11 12
13
14
Conclusions
ED written discharge instructions appeared inadequate in providing recommended education for patients with severe hypoglycemia Older age and sulfonylurea use were independently associated with hospital admission Although hypoglycemia is generally considered a self-limited condition, 2 of every 5 patients required hospitalization, which likely reflects an older and more complex patient population Opportunities for improvement in the management of ED patients with hypoglycemia center on recognition and documentation of cause and improved discharge instructions
References
1 Gonder-Frederick LA, Clarke WL, Cox DJ The emotional, social, and behavioral implications of insulin-induced hypoglycemia Semin Clin Neuropsychiatry 1997;2:57-65 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 1993;329:977-986 The 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 1998;352:837-853 The 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 1998;352:854-865 Cryer PE Hypoglycaemia: the limiting factor in the glycaemic management of type I and type II diabetes Diabetologia 2002;45:937-948 MacLeod KM, Hepburn DA, Frier BM Frequency and morbidity of severe hypoglycaemia in insulin-treated diabetic patients Diabet Med 1993;10:238-245 Laing SP, Swerdlow AJ, Slater SD, et al The British Diabetic Association Cohort Study II: cause-specific mortality in patients with insulin-treated diabetes mellitus Diabet Med 1999;16:466-471
15
16
17
18
19
20
21 22 23
2
3
24
4
25
26
5
6
27
7
28
DCCT Research Group Hypoglycemia in the diabetes control and complications trial Diabetes 1997;46;271-286 Bell DS, Cutter G Characteristics of severe hypoglycemia in the patient with insulin-dependent diabetes South Med J 1994;87: 616-620 Holstein A, Plaschke A, Egberts EH Incidence and costs of severe hypoglycemia Diabetes Care 2002;25:2109 The International
Classification of Diseases, 9th revision, Clinical Modification 5th ed Salt Lake City, Utah: Medicode; 1997 Johnson ES, Koepsell TD, Reiber G, Stergachis A, Platt R Increasing incidence of serious hypoglycemia in insulin users J Clin Epidemiol 2002;55:253-259 Shorr RI, Ray WA, Daugherty JR, Griffin MR Incidence and risk factors for serious hypoglycemia in older persons using insulin or sulfonylureas Arch Intern Med 1997;15:1681-1687 Herings RMC, de Boer A, Stricker BHCh, Leufkens HGM, Porsuis A Hypoglycemia associated with use of inhibitors of angiotensin converting enzyme Lancet 1995;345:1195-1198 Morris AD, Boyle DIR, McMahon AD, et al ACE inhibitor use is associated with hospitalization for severe hypoglycemia in patients with diabetes Diabetes Care 1997;20:1363-1367 Shorr RI, Ray WA, Daugherty JR, Griffin MR Antihypertensives and the risk of serious hypoglycemia in older persons using insulin or sulfonylureas JAMA 1997;278:40-43 Ginde AA, Blanc PB, Lieberman RM, Camargo CA Jr Validation of ICD-9 coding to identify emergency department patients with hypoglycemia BMC Endocr Disord 2008;8:4 American Diabetes Association Workgroup on Hypoglycemia: Defining and reporting
hypoglycemia in diabetes Diabetes Care 2005;28:1245-1249 Brackenridge A, Wallbank H, Lawrenson RA, Russell-Jones D Emergency management of diabetes and hypoglycaemia Emerg Med J 2006;23:183-185 Harrigan RA, Nathan MS, Beattie P Oral agents for the treatment of type 2 diabetes mellitus: pharmacology, toxicity, and treatment Ann Emerg Med 2001;38:68-78 Cryer PE, Davis SN, Shamoon H Hypoglycemia in diabetes Diabetes Care 2003;26:1902-1912 American Diabetes Association Standards of medical care in diabetes–2007 Diabetes Care 2007;30:S4-S41 Nordfeldt S, Johansson C, Carlsson E, Hammersjo JA Persistent effects of a pedagogical device targeted at prevention of severe hypoglycaemia: a randomized controlled study Acta Paediatr 2005;94:1395-1401 Segel SA, Paramore DS, Cryer PE Hypoglycemia-associated autonomic failure in advanced type 2 diabetes Diabetes 2002;51:724-744 Dagogo-Jack SE, Craft S, Cryer PE Hypoglycemia-associated autonomic failure in insulin-dependent diabetes mellitus J Clin Invest 1993;91:819-828 Fanelli C, Pampanelli S, Epifano L, et al Long-term recovery from unawareness, deficient counterregulation and lack of cognitive dysfunction during hypoglycaemia following
institution of rational intensive insulin therapy in IDDM Diabetologia 1994;37:1265-1276 Cranston I, Lomas J, Maran A, Macdonald I, Amiel SA Restoration of hypoglycaemia awareness in patients with longduration insulin dependent diabetes Lancet 1994;344:283-287 Dagogo-Jack S, Rattarasarn C, Cryer PE Reversal of hypoglycemia unawareness, but not defective glucose counterregulation, in IDDM Diabetes 1994;43:1426-1434
Ginde et al
Downloaded from http://tdesagepubcom at UNIV OF COLORADO DENVER HEALTH SCIENCES LIBRARY on August 4, 2008 2008 American Association of Diabetes Educators; Published by SAGE Publications All rights reserved Not for commercial use or unauthorized distribution
