Recognizing Hypoglycemia in Children Through
Automated Adverse-Event Detection
WHAT’S KNOWN ON THIS SUBJECT: Automated adverse-event detection using triggers derived from the electronic health record is an efficient method of identifying adverse events, including hypoglycemia. However, there has been limited investigation using this system to detect adverse events in hospitalized children.
WHAT THIS STUDY ADDS: Hypoglycemia is common in
hospitalized children, particularly neonates and those receiving insulin therapy. An electronic health record– driven automated adverse-event detection system was effective in identifying hypoglycemia in this population and will augment the safety programs of organizations that have adopted the electronic health record.
abstract
BACKGROUND:Automated adverse-event detection using triggers de-rived from the electronic health record (EHR) is an effective method of identifying adverse events, including hypoglycemia. However, the true occurrence of adverse events related to hypoglycemia in pediatric inpatients and the harm that results remain largely unknown.
OBJECTIVE:We describe the use of an automated adverse-event detec-tion system to detect and categorize hypoglycemia-related adverse events in pediatric inpatients.
METHODS:A retrospective observational study of all hypoglycemia triggers generated by an EHR-driven surveillance system was con-ducted at a large urban children’s hospital during a 1-year period. All hypoglycemia triggers were investigated to determine if they repre-sented a true adverse event and if that event followed or deviated from the local standard of care. Clinical and demographic variables were analyzed to identify subpopulations at risk for hypoglycemia.
RESULTS:Of the 1254 hypoglycemia triggers produced, 198 were verse events (positive predictive value: 15.8%). No hypoglycemic ad-verse events were identified via the hospital’s voluntary incident-reporting system. The majority of hypoglycemia-related adverse events occurred in the NICU (n⫽123 of 198 [62.1%]). A total of 154 (77.8%) of the 198 adverse events hospital-wide and 102 (83%) of the 123 adverse events in the NICU occurred in patients who were receiving insulin therapy.
CONCLUSIONS:Hypoglycemia is common in hospitalized children, par-ticularly neonates and those who receive insulin. An EHR-driven auto-mated adverse-event detection system was effective in identifying hypoglycemia in this population. Automated adverse-event detection holds great promise in augmenting the safety program of organiza-tions who have adopted the EHR.Pediatrics2011;127:e1035–e1041
AUTHORS:Mindy J. Dickerman, MD,a,bBrian R. Jacobs,
MD,c,d,eHima Vinodrao, RN,cand David C. Stockwell, MDd,e
aDivision of Pediatric Critical Care, St Christopher’s Hospital for
Children, Philadelphia, Pennsylvania;bDepartment of Pediatrics,
College of Medicine, Drexel University, Philadelphia,
Pennsylvania;cCenter for Pediatric Informatics andeDivision of
Pediatric Critical Care, Children’s National Medical Center, Washington, DC; anddDepartment of Pediatrics, George
Washington University School of Medicine, Washington, DC
KEY WORDS
trigger tool, adverse event, patient safety, hypoglycemia, pediatrics, NICU, electronic health record
ABBREVIATION
EHR—electronic health record
www.pediatrics.org/cgi/doi/10.1542/peds.2009-3432
doi:10.1542/peds.2009-3432
Accepted for publication Dec 23, 2010
Address correspondence to Mindy J. Dickerman, MD, Division of Pediatric Critical Care Medicine, St Christopher’s Hospital for Children, 3601 A St, Philadelphia, PA 19134.
E-mail: [email protected]
PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).
Copyright © 2011 by the American Academy of Pediatrics
FINANCIAL DISCLOSURE:The authors have indicated they have no financial relationships relevant to this article to disclose.
experience an adverse event each year, 60% of which may be prevent-able.1,2 Although much of the patient safety research has focused on adult patients, several recent studies de-scribe the frequency and epidemiol-ogy of adverse events and their asso-ciated harm in pediatric inpatient settings.2–11
The Institute of Medicine released guidelines urging the medical commu-nity to develop and improve surveil-lance systems to detect adverse events.12Traditional methods used to detect adverse events in the inpatient population, such as manual chart re-view and voluntary incident report-ing, significantly underestimate the number and prevalence of adverse events, are subject to bias, and are time-consuming.13,14
A newer detection strategy relying on the trigger method has been shown to augment traditional methods. A trig-ger is defined as an “occurrence, prompt, or flag found on manual re-view of the medical chart that ‘trig-gers’ additional investigation to deter-mine the presence or absence of an adverse event.”15,16An example is the administration of naloxone, an opiate antidote, to a child who has respira-tory depression after an opiate over-dose. Most triggers developed target adult inpatient adverse events. More recently, trigger tools have been de-signed specifically to identify adverse events in hospitalized neonates and children.10,17
Computerized adverse-event surveil-lance relies on the automated genera-tion of a trigger report to indicate the possibility that an adverse event has occurred, requiring additional investi-gation. This method has proven to be an efficient and cost-effective way to detect adverse events.18–22 Although
very few pediatric health care facilities have adopted this method.19
Trigger methods to detect hypoglyce-mia have been studied in adults but have undergone limited investiga-tion in children.19,20 A symptomatic hypoglycemia-related event (eg, sei-zures, mental status changes, apnea) occurring in a pediatric hospitalized patient is clearly an adverse event and 1 that results in immediate harm. How-ever, long-term harm resulting from hypoglycemia, as well as harm result-ing from asymptomatic hypoglycemia, remains speculative. There is evidence that hypoglycemia in patients with dia-betes, even if asymptomatic, leads to dysfunctional glucose counterregula-tion, hypoglycemia unawareness, and increased risk of subsequent hypogly-cemia.23–25Previous studies have found an association with hypoglycemia and neurodevelopmental deficits in nondi-abetic preterm infants and preterm in-fants small for their gestational age.26,27This association was reported more recently in term infants without hypoxic-ischemic encephalopathy af-ter symptomatic hypoglycemia.28
The increasing use of insulin protocols in our hospital and worldwide to main-tain euglycemia in nondiabetic hospi-talized children has created an added risk for hypoglycemia-related adverse events and potential harm.29,30A recent multicenter trial evaluating the effect of early insulin replacement in very low birth weight infants was sus-pended because of the increased inci-dence of brain parenchymal lesions in the early insulin group, without any im-provement in predefined outcomes.31
The purpose of this study was to deter-mine the effectiveness of an auto-mated adverse-event detection system in detecting hypoglycemia-related ad-verse events in pediatric inpatients. A
events in an effort to determine which patients are most at risk for hypoglycemia and to identify poten-tial prevention strategies.
METHODS
We conducted a retrospective observa-tional study of all hypoglycemia trig-gers generated over a 1-year period by using an EHR-driven surveillance sys-tem currently in use at Children’s Na-tional Medical Center (Washington, DC). Children’s National Medical Cen-ter is an urban, academic, freestand-ing children’s hospital with 283 total inpatient beds, 41 PICU beds, 48 NICU beds, and⬎13 000 admissions annu-ally. The study period was from Sep-tember 4, 2007, to August 31, 2008. The hypoglycemia trigger was defined as a glucose valueⱕ50 mg/dL determined from whole blood point-of-care testing or serum samples processed in the laboratory. Data were electronically collected from the emergency depart-ment and all inpatient clinical areas, including medical, surgical, psychiat-ric, and intensive care units.
Adverse-event severity and associated patient harm were assessed using a modified index from the National Coor-dinating Council for Medication Error Reporting and Prevention.32 (3) The clinical analyst examined the hospi-tal’s Web-enabled voluntary incident-reporting system (Web Envision [Envi-sion Web Technologies, Midvale, UT]) to determine if a voluntary incident re-port was completed for each trigger. (4) The clinical analyst met with a multidisciplinary Automated Adverse Event Detection Steering Committee monthly to present the data and adju-dicate questions related to classifica-tions assigned to each event.
An adverse event was defined as hypo-glycemia associated with medical therapy (eg, insulin use) or lack of ap-propriate medical therapy (eg, hypo-glycemia that resulted from intrinsic disease that was not appropriately treated). Initial hypoglycemic events that occurred as a result of the pa-tient’s disease or condition (eg, sepsis or hyperinsulinemia), as well as trig-gers determined to be false-positive laboratory results, were not adverse events.
Adverse events were further classified as following or deviating from the local standard of care. Events following the standard of care were those that oc-curred despite the medical team ad-hering to the locally accepted prac-tices. Hypoglycemia as a result of receiving a 10-fold insulin overdose or not receiving any intravenous dextrose while enteral nutrition was held in preparation for a procedure were ex-amples of events caused by care that deviated from the local standard. Hy-poglycemic events that occurred in children administered an appropriate insulin regimen according to hospital protocol with the diagnosis of either diabetic ketoacidosis or hyperglyce-mia of critical illness were events in which the standard of care was
fol-lowed. When there was uncertainty re-garding the appropriate classification of a trigger, the Automated Adverse Event Detection Steering Committee discussed the event and came to a con-sensus. If local standard of care was not clearly delineated, relevant physi-cian divisions were asked to clarify their standards and then applied accordingly.
The institutional review board at Chil-dren’s National Medical Center evalu-ated and approved this study. Descrip-tive statistics were used for much of the data, with means (SDs) where ap-propriate. Student’sttest was used to assess differences in continuous vari-ables, and2analysis or Fisher’s exact test was used to compare categorical data. APvalue ofⱕ.05 was considered significant.
RESULTS
The EHRs of all patients admitted to the hospital from September 4, 2007, through August 31, 2008 (79 282 total patient-days, 25 856 ICU patient-days, and 13 526 admissions), were queried daily for investigation. Weekend days from September 2007 through Febru-ary 2008 were excluded because of in-vestigator resource limitations. A total of 1254 hypoglycemia triggers were detected. Of the 1254 triggers gener-ated, 198 (15.8%) were determined to be adverse events, and 1056 (84.2%) were not adverse events. The positive predictive value of the electronic trig-ger for identifying a hypoglycemia-related adverse event was 15.8%. Of the 198 adverse events, 95 (48%) were determined to be associated with devi-ation from the standard of care and 103 (52%) followed the standard of care. The hospital’s Web-enabled vol-untary incident-reporting system had no entries for hypoglycemic adverse events during this same time period.
The 1056 triggers that were deter-mined not to be adverse events were
either false-positive laboratory values (ie, specimen was incorrectly pro-cessed and simultaneous laboratory serum value or whole blood point-of-care value was⬎50 mg/dL) or lacked objective evidence of being caused by medical care or lack of medical care and were deemed to be secondary to the patient’s illness (ie, septic shock). A portion of the false-positive values lacked confirmatory testing because of patient refusal, discharge, or a limi-tation of care order that precluded fur-ther testing.
To quantify resource utilization, the time required for investigation of trig-gers was recorded for all trigtrig-gers gen-erated from April 24, 2008, through Au-gust 31, 2008 (n ⫽ 594 triggers evaluated). A mean of 8.6 (SD: 3.4) min-utes was expended for trigger evalua-tion overall, 9.6 (SD: 4.3) minutes for adverse events, and 8.4 (SD: 3.2) min-utes for false-positive triggers.
Adverse-event rates were calculated on the basis of data from March 1, 2008, through August 31, 2008. We cal-culated adverse-event rates per 1000 patient-days and per 100 patient ad-missions (Table 1). Adverse-event rates were stratified according to hos-pital location. The locations were di-vided into NICU, PICU/cardiac ICU (com-bined and denoted as PICU), and acute care units (including all general pedi-atric and subspecialty pedipedi-atric non-ICU units).
TABLE 1 Hypoglycemia Adverse-Event Rates According to Hospital Location (March 1, 2008, to August 31, 2008)
Hospital Location
Adverse-Event Rate per 1000 Patient-Days
Adverse-Event Rate per 100
Patient Admissions
All hospital locations
3.5 2.1
All ICUs 9.3 8.3
NICU 13.4 28.2
PICU/cardiac ICU 4.5 2.4
Acute care units 0.8 0.4
Total number of adverse events⫽142.
CHARACTERISTICS OF ADVERSE EVENTS
The 198 adverse events affected 68 pa-tients. Distribution of glucose values of these adverse events are displayed in Fig 1. Of the 198 adverse events identi-fied in this study, all resulted in the need for intervention and were catego-rized as resulting in temporary harm. Temporary harm included the need for additional phlebotomy and intrave-nous dextrose boluses. One patient ex-perienced a seizure attributable to hy-poglycemia that required temporary seizure management but did not re-quire increased respiratory support. There were no hypoglycemia-related adverse events resulting in coma or death. We were unable to evaluate long-term harm because of lack of follow-up beyond inpatient discharge.
The majority of hypoglycemic events
occurred in the NICU (n ⫽ 123
[62.1%]). Of the 123 adverse events that occurred in the NICU, 48 (39%) were associated with deviation from the standard of care and 75 (61%) fol-lowed the standard of care. Of the 123 total adverse events investigated in the NICU, 92 (75%) occurred in prema-ture infants. The risk for hypoglycemia
in an infant born prematurely and ad-mitted to the NICU was significantly higher than the risk for an infant born term and admitted to the NICU during this same time period (P ⬍ .001). Forty-four percent of all adverse events in the hospital (n ⫽ 88) oc-curred in low birth weight infants (ie,
⬍2500 g).
A total of 154 adverse events (77.8%) occurred in patients receiving insulin therapy, via continuous intravenous in-sulin infusion (n ⫽ 127 [82.5%]) or subcutaneous injection (n ⫽ 27 [17.5%]). Eighty-nine of the 127 events that occurred during an insulin infu-sion occurred in premature neonates, representing 97% of all adverse events in premature infants. Eighty-three per-cent (n⫽102) of the adverse events in the NICU occurred in neonates receiv-ing continuous insulin infusion. The majority of these events in the NICU oc-curred while standard insulin infusion protocols were being followed. The in-sulin protocol for infants with sus-tained hyperglycemia persus-tained to pa-tients with glucose values⬎150 mg/dL and was targeted to achieve goal glu-cose concentrations between 60 and 150 mg/dL.
related adverse event hospital-wide to be 9.0 events per 100 patients receiv-ing any insulin therapy durreceiv-ing their ad-mission. There was a significant differ-ence in the rate of hypoglycemia in patients receiving an insulin infusion versus the rate of hypoglycemia in those receiving subcutaneous insulin hospital-wide (13.5 events per 100 pa-tients receiving insulin infusion, 3.8 events per 100 patients receiving sub-cutaneous insulin;P⫽.001). The risk of hypoglycemia during admission to the NICU was 167 events per 100 pa-tients receiving insulin therapy.
Several demographic and clinical vari-ables were evaluated to identify risk factors for hypoglycemia-related ad-verse events associated with deviance from the standard of care compared with events adherent to the standard of care. Events associated with care deviant from the standard were more likely to occur on the weekends (0.464 adverse events per weekend day ver-sus 0.250 adverse events per weekday; P⫽ .048.) In contrast, there was no significant difference in the proportion of hypoglycemic events associated with deviations in the standard of care compared with events when the stan-dard of care was followed for day shift versus night shift, or for subcutaneous or continuous insulin use.
Multiple hypoglycemic events oc-curred in 33 patients. Sixty-five percent of these events occurred despite ad-herence to the standard of care. Six children had⬎6 hypoglycemic events. All 6 children were premature neo-nates receiving continuous insulin in-fusion for the treatment of nondiabetic hyperglycemia.
DISCUSSION
To our knowledge, this study, from which we describe hypoglycemic
FIGURE 1
events that occurred in 79 282 pediat-ric inpatient-days from a single ter-tiary care children’s hospital over a 1-year period, is the largest epidemio-logic description of hypoglycemia-related adverse events in pediatrics published to date.
Our study supports the use of auto-mated adverse-event detection as an effective method for detecting adverse events in hospitalized children. Al-though the positive predictive value of 15.8% may seem low for a surveillance system, the merits of this system, in-cluding reasonable time required for investigation of each trigger (mean: 8.6⫾3.4 minutes) and complete cap-ture of every hypoglycemic event that occurred in 1 year, supports the utility of this method.
The detection rate for adverse events in an automated adverse-event detec-tion system depends on the trigger pa-rameter design. The hypoglycemia adverse-event rate determined in our study, 3.5 events per 1000 patient-days, is greater than the rate of 0.19 events per 1000 patient-days reported in an-other pediatric inpatient population.19 Ferranti et al19defined the trigger rule to require both hypoglycemia (plasma blood glucose⬍50 mg/dL) and docu-mented administration of an intrave-nous dextrose 50% bolus. We did not include a dextrose bolus in our trigger to increase the adverse-event capture rate, lowering the specificity of this trigger. Although distinct patient pop-ulations may explain the inconsistent rates, the different design of the auto-mated trigger rules used by the 2 or-ganizations likely influenced both the trigger rate and the subsequent posi-tive predicposi-tive value.
Many hypoglycemic events identified in this study were associated with glu-cose values⬍40 mg/dL. Although the establishment of a low safe limit value for blood glucose concentration re-mains controversial, many authorities
believe that in neonates, any value
⬍40 mg/dL be viewed as a risk for long-term neurodevelopmental
se-quelae and should be treated
appropriately.33
Our study identified a substantial hypoglycemia-related adverse-event rate of 28.2 per 100 NICU admissions. Premature and low birth weight neo-nates were found to be at highest risk for hypoglycemia-related adverse events. On the basis of studies by Lucas et al26and Duvanel et al,27it is likely this population is vulnerable to long-term harm from hypoglycemia. The majority of events occurring in the NICU were associated with insulin therapy, a practice whose efficacy in this particu-lar population is unclear and not with-out potential risk.31The rate of hypogly-cemia in NICU patients while receiving continuous insulin in our study (167 events per 100 patients) is difficult to compare with rates reported by Beardsall et al31because of the differ-ent definitions of hypoglycemia used in each study.
In all patients, insulin therapy was identified by our study as a risk factor for hypoglycemia-related adverse events. Despite 77.8% of adverse events occurring while patients were receiving insulin therapy, only 6.6% of all adverse events occurred in patients with a diagnosis of diabetes. The 13.5% hospital-wide risk of hypoglycemia while receiving continuous insulin in-fusion we report is consistent with the 6% to 16% incidence of hypoglycemia related to intensive glycemic control with insulin therapy reported in adult inpatient literature.34–37
The only statistically significant vari-able associated with adverse events occurring in the context of a deviation from standard of care was the day of the week, suggesting preventable events are more likely to occur on weekend days. This finding will need to be investigated in future studies.
Finally, our data revealed a subset of patients who experienced recurrent episodes of hypoglycemia, which has implications for children both with and without diabetes. In preterm infants without diabetes, recurrent episodes of hypoglycemia are a predictor for neurodevelopmental deficits when compared with a single hypoglycemic episode.27For patients with diabetes, the American Diabetes Association has established a goal for glycemic man-agement to “reduce the frequency and severity of hypoglycemia.”38
The data collected on NICU patients were disseminated to the neonatolo-gists in our institution, leading to the implementation of a new insulin proto-col for NICU patients with hyperglyce-mia. Preliminary results show de-creased episodes of hypoglycemia while using this protocol. This retro-spective work serves to highlight the importance of real-time monitoring and interventions by an adverse-event detection team or perhaps a workflow-integrated adverse-event alert within the EHR.
There are some important limitations to our study. When the investigator was unable to confirm a false-positive trigger because of lack of confirma-tory testing, we deferred to the clini-cian’s determination and did not cate-gorize these triggers as adverse events. This potentially underesti-mated the true rate of hypoglycemia-related adverse events. Furthermore, the inclusion of point-of-care testing in addition to laboratory serum values to define hypoglycemia may misrepsent the adverse-event rates we re-port. Recent studies report a statisti-cally significant difference between glucose values obtained by a point-of-care device and values obtained by a laboratory, with 1 study revealing 20% of values differing by ⬎20 mg/dL.39,40 Finally, adverse-event rates were cal-culated on the basis of data collected
on the basis of data available from April 2008 through August 2008. Al-though data collection began in Sep-tember 2007, complete data for all triggers, including those occurring on weekends and holidays, were not available before March 2008 because of clinical analyst availability. How-ever, because of the monthly volume of the hypoglycemia trigger, the val-ues are likely a representative sample.
ing diagnosis, gestational age) as well as specifying the need for increased phlebotomy, increased intravenous ac-cess, and hyperglycemia after a dex-trose infusion.
CONCLUSIONS
An EHR-derived automated adverse-event detection system is an efficient and effective method to identify hypo-glycemia in hospitalized children. Hy-poglycemia is a common adverse
therapy are at the highest risk for hy-poglycemia. Although our results are based on a single institution’s experi-ence, this information should signifi-cantly aid in the development of strat-egies to reduce hypoglycemia in hospitalized children.
ACKNOWLEDGMENT
This study was supported in part by a research grant from the Cerner Corporation.
REFERENCES
1. Thomas EJ, Studdert D, Burstin HR, et al. Incidence and types of adverse events and negligent care in Utah and Colorado.Med Care. 2000;38(3):261–271
2. Woods D, Thomas E, Holl J, Altman S, Bren-nan T. Adverse events and preventable ad-verse events in children.Pediatrics. 2005; 115(1):155–160
3. Goldmann D, Kaushal R. Time to tackle the tough issues in patient safety.Pediatrics. 2002;110(4);823– 826
4. Bates DW, Cullen DJ, Laird N, et al. Incidence of adverse drug events and potential ad-verse drug events: implications for preven-tion. ADE Prevention Study Group.JAMA. 1995;274(1):29 –34
5. Kaushal R, Bates DW, Landrigan C, et al. Medication errors and adverse drug events in pediatric inpatients.JAMA. 2001;285(16): 2114 –2120
6. Fortescue EB, Kaushal R, Landrigan C, et al. Prioritizing strategies for preventing medi-cation errors and adverse drug events in pediatric inpatients.Pediatrics. 2003;111 (4 pt 1):722–729
7. Larsen GY, Donaldson AE, Parker HB, Grant MJ. Preventable harm occurring to criti-cally ill children.Pediatr Crit Care Med. 2007;8(4):331–336
8. Miller MR, Elixhauser A, Zhan C. Patient safety events during pediatric hospitaliza-t i o n s . P e d i a t r i c s. 2 0 0 3 ; 1 1 1 ( 6 p t 1 ) : 1358 –1366
9. Kugelman A, Inbar-Sanado E, Shinwell E, et al. Iatrogenesis in neonatal intensive care units: observational and interven-tional, prospective, multicenter study. Pedi-atrics. 2008;122(3):550 –555
10. Sharek PJ, Horbar JD, Mason W, et al. Ad-verse events in the neonatal intensive care
unit: development, testing, and findings of an NICU-focused trigger tool to identify harm in North American NICUs.Pediatrics. 2006;118(4):1332–1340
11. Agarwal S, Classen D, Larsen G, et al. Preva-lence of adverse events in pediatric inten-sive care units in the United States.Pediatr Crit Care Med. 2010;11(5):568 –578
12. Institute of Medicine.Preventing Medica-tion Errors: Quality Chasm Series. Washing-ton, DC: National Academy Press; 2006
13. Cullen DJ, Bates DW, Small SD, Cooper JB, Nemeskal AR, Leape LL. The incident report-ing system does not detect adverse drug events: A problem for quality improvement.
Jt Comm J Qual Improv. 1995;21(10): 541–548
14. Silver MP, Antonow JA. Reducing medica-tion errors in hospitals: a peer review orga-nization collaboration.Jt Comm J Qual Im-prov. 2000;26(6):332–340
15. Rozich JD, Haraden CR, Resar RK. Adverse drug event trigger tool: a practical method-ology for measuring medication related harm. Qual Saf Health Care. 2003;12(3): 194 –200
16. Resar RK, Rozich JD, Classen D. Methodol-ogy and rationale for the measurement of harm with trigger tools. Qual Saf Health Care. 2003;12(suppl 2):ii39 –ii45
17. Takata GS, Mason W, Taketomo C, Logsdon T, Sharek PJ. Development, testing, and find-ings of a pediatric-focused trigger tool to identify medication-related harm in US chil-dren’s hospitals. Pediatrics. 2008;121(4): e927– e935
18. Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of adverse drug events in hospitalized patients. 1991.
Qual Saf Health Care. 2005;14(13):221–225
19. Ferranti J, Horvath MM, Cozart H, White-hurst J, Eckstrand J. Reevaluating the safety profile of pediatrics: a comparison of computerized adverse drug event surveil-lance and voluntary reporting in the pediatric environment.Pediatrics. 2008; 121(5). Available at: www.pediatrics.org/ cgi/content/full/121/5/e1201
20. Ferranti J, Horvath M, Cozart H, et al. A multifaceted approach to safety: the syn-ergistic detection of adverse drug events in adult inpatients. J Patient Saf. 2008; 4(3):184 –190
21. Jha AK, Kuperman GJ, Teich J, et al. Identify-ing adverse drug events: development of a computer-based monitor and comparison with chart review and stimulated voluntary report.J Am Med Inform Assoc. 1998;5(3): 305–314
22. Bates DW, Evans RS, Murff H, Stetson PD, Pizziferri L, Hripcsak G. Detecting adverse events using information technology.J Am Med Inform Assoc. 2003;10(2):115–128
23. Cryer PE, Davis SN, Shamoon H. Hypoglyce-mia in diabetes.Diabetes Care. 2003;26(6): 1902–1912
24. Cryer PE. Diverse causes of hypoglycemia-associated autonomic failure in diabetes.N Engl J Med. 2004;350(22):2272–2279
25. Dagogo-Jack SE, Craft S, Cryer PE. Hypoglycemia-associated autonomic fail-ure in insulin-dependent diabetes mellitus. Recent antecedent hypoglycemia reduces autonomic responses to, symptoms of, and defense against subsequent hypoglycemia.
J Clin Invest. 1993;91(3):819 – 828
27. Duvanel CB, Fawer CL, Cotting J, Hohlfeld P, Matthieu JM. Long-term effects of neonatal hypoglycemia on brain growth and psy-chomotor development in small-for-gestational-age preterm infants.J Pediatr. 1999;134(4):492– 498
28. Burns CM, Rutherford MA, Boardman JP, Cowan FM. Patterns of cerebral injury and neurodevelopmental outcomes after symp-tomatic neonatal hypoglycemia.Pediatrics. 2008;122(1):65–74
29. Wintergerst KA, Buckingham B, Gandrud L, Wong BJ, Kache S, Wilson DM. Association of hypoglycemia, hyperglycemia, and glucose variability with morbidity and death in the pediatric intensive care unit. Pediatrics. 2006;118(1):173–179
30. Hermanides J, Bosman RJ, Vriesendorp TM, et al. Hypoglycemia is associated with inten-sive care unit mortality.Crit Care Med. 2010; 38(6):1430 –1434
31. Beardsall K, Vanhaesebrouck S, Ogilvy-Stuart A, et al. Early insulin therapy in
very-low birth-weight infants.N Engl J Med. 2008; 359(18):1873–1884
32. NCC MERP index for categorizing medica-tion errors. Available at: www.nccmerp. org/medErrorCatIndex.html. Accessed Sep-tember 21, 2009
33. Cornblath M. Neonatal hypoglycemia 30 years later: does it injure the brain? Histor-ical summary and present challenges.Acta Paediatr Jpn. 1997;39(suppl 1):S7–S11
34. Preissig CM, Hansen I, Roerig PL, Rigby MR. A protocolized approach to identify and man-age hyperglycemia in a pediatric critical care unit.Pediatr Crit Care Med. 2008;9(6): 581–588
35. NICE-SUGAR Study Investigators; Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360(13): 1283–1297
36. Arabi YM, Tamim HM, Rishu AH. Hypoglyce-mia with intensive insulin therapy in criti-cally ill patients: predisposing factors and
association with mortality.Crit Care Med. 2009;37(9):2536 –2544
37. Preiser JC, Devos P, Ruiz-Santana S, et al. A prospective randomised multi-centre con-trolled trial on tight glucose control by in-tensive insulin therapy in adult inin-tensive care units: the Glucocontrol study.Intensive Care Med. 2009;35(10):1738 –1748
38. Workgroup on Hypoglycemia, American Dia-betes Association. Defining and reporting hypoglycemia in diabetes: a report from the American Diabetes Association Workgroup on Hypoglycemia. Diabetes Care. 2005; 28(5):1245–1249
39. Cook A, Laughlin D, Moore M, et al. Differ-ences in glucose values obtained from point-of-care glucose meters and labora-tory analysis in critically ill patients.Am J Crit Care. 2009;18(1):65–71
40. Shearer A, Boehmer M, Closs M, et al. Com-parison of glucose point-of-care values with laboratory values in critically ill patients.
Am J Crit Care. 2009;18(3):224 –230
DOI: 10.1542/peds.2009-3432 originally published online March 14, 2011;
2011;127;e1035
Pediatrics
Services
Updated Information &
http://pediatrics.aappublications.org/content/127/4/e1035
including high resolution figures, can be found at:
References
http://pediatrics.aappublications.org/content/127/4/e1035#BIBL
This article cites 37 articles, 17 of which you can access for free at:
Subspecialty Collections
rds_sub
http://www.aappublications.org/cgi/collection/electronic_health_reco
Electronic Health Records
chnology_sub
http://www.aappublications.org/cgi/collection/health_information_te
Health Information Technology following collection(s):
This article, along with others on similar topics, appears in the
Permissions & Licensing
http://www.aappublications.org/site/misc/Permissions.xhtml
in its entirety can be found online at:
Information about reproducing this article in parts (figures, tables) or
Reprints
http://www.aappublications.org/site/misc/reprints.xhtml
DOI: 10.1542/peds.2009-3432 originally published online March 14, 2011;
2011;127;e1035
Pediatrics
Mindy J. Dickerman, Brian R. Jacobs, Hima Vinodrao and David C. Stockwell
Detection
Recognizing Hypoglycemia in Children Through Automated Adverse-Event
http://pediatrics.aappublications.org/content/127/4/e1035
located on the World Wide Web at:
The online version of this article, along with updated information and services, is
by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.
