HW Assignment 2

A R T I C L E I N F O

Keywords: Analgesics, opioid Drug overdose Pediatrics Epidemiology Healthcare disparity

A B S T R A C T

1. Introduction

Since the 1990s, an upsurge of opioid misuse among patients of all ages have contributed to the epidemic of opioid overdose in the United States [1–3]. In 2016, > 11 million people aged 12 years and older reported opioid misuse in the prior year, with the highest-risk popu- lation being those 18 to 25 years of age [4]. In the last two decades, hospitalizations for opioid poisonings nearly doubled among per- sons < 20 years of age [4,5]. Adolescents are especially susceptible to the devastating effects of opioid misuse, including but not limited to: opioid use disorder, psychosocial impairment, hepatitis C and human immunodeficiency viral infection following intravenous drug use,

overdose, and death [5,6]. While recent studies suggest that the ma- jority of opioid overdoses in adolescents are intentional, accidental overdoses have increased at a faster rate, tripling between 1997 and 2012 [4]. Unintentional overdoses make up 80% of drug overdose deaths, with the majority attributed to opioid use [4–7].

Despite increasing rates of opioid-related hospitalizations in recent years, there is limited data on the clinical profile of pediatric patients admitted to the hospital for opioid overdose. Gaither et al. evaluated trends in rates of hospitalizations for opioid poisonings in the pediatric population and found that hospital admissions due to opioid overdose nearly doubled between 1997 and 2012 [5]. The Center for Disease Control and Prevention (CDC) estimates that while a decline in opioid-

⁎

T

We used the International Classification of Disease, Ninth Revision (ICD-9) procedure codes to extract records of pediatric patients > 1 year of age who were admitted for prescription or illicit opioid over- dose [10]. Fig. 1 shows the study inclusion criteria. KID generates data every three years; our analysis includes data for years 2000, 2003, 2006, 2009, and 2012. KID data were not available beyond 2012. Prescription opioid overdose (POD) was defined with the following ICD-9-CM codes: 9650.00 (poisoning with opium), 965.02 (poisoning by methadone), 965.09 (poisoning by other opiates and related nar- cotics), E850.1 (accidental poisoning by methadone), and E850.2 (ac- cidental poisoning by other opiates and related narcotics). Illicit opioid overdose (IOD) was defined as 965.01 (poisoning by heroin) and

E850.0 (accidental poisoning by heroin). Furthermore, we utilized ICD- 9 diagnosis codes to identify risk factors of inpatient mortality. Table 1 lists the ICD-9 codes utilized to extract the diagnoses and procedures for each case.

2.2. Statistical analysis

All statistical analysis was performed using R, a software environ- ment for statistical computing (R version 3.5.1). We applied discharge weights to each patient recorded provided by KID to reflect national estimates. Pearson chi-square test and Wilcoxon rank sum test were performed for comparisons among opioid overdose cohorts (prescrip- tion versus illicit). We excluded patients < 1 years of age. To determine trend in opioid overdose admissions with year, we report the Pearson's correlation coefficient. We used the trend weight data element “DISCWT” to create national estimates for trends analysis. Cases with > 5% missing values were recorded as “unknown” and included in the analysis. Otherwise, cases with missing values were excluded from the analysis. Two-sided p < 0.00125 (correcting for multiple com- parisons) was considered statistically significant.

3. Results

The final analysis included 15,884 pediatric patients admitted to a United States hospitals with opioid overdose. Fig. 2 shows the pre- valence of opioid overdose in the United States among the pediatric population by hospital admission year. Opioid overdose hospital ad- missions increased from 2010 to 2012. The rate of POD and IOD in- creased during this period.

Table 2 shows the sociodemographic variables among cases with POD compared to the IOD cohort. Black, Asian or Pacific Islander,

All pediatric patients in KID 2000 – 2012: 32,011,902

Population excluded: 97

Study population of opioid overdose: 15,841

Prescription opioid overdose: 13,348

Illicit opioid overdose: 2,536

All pediatric patients with ICD-9 diagnosis code of prescription or illicit opioid overdose: 15,981

Fig. 1. All data weighted for national estimates. Patient enrollment flow chart. KID = Kids' Inpatient Database. ICD – 9 = International Classification of Diseases, Ninth Revision.

related deaths among adolescents 15 to 19 years of age began at the end of 2007, the rate began to rise again in 2015 [7]. The primary objective of this study was to explore the demographics and comorbid conditions among pediatric patients admitted for POD or IOD.

2. Materials and methods

2.1. Data collection and study population

Data were obtained from the publicly available dataset Kids' Inpatient Database (KID) of the Healthcare Cost and Utilization Project (HCUP) [8]. KID is the largest pediatric inpatient database in the United States. KID includes de-identified d ata o f d iagnoses a nd procedures, demographic characteristics, hospital-related factors, payment source, total charges, discharge status, and length of hospital stay [9]. The database is de-identified and meets the criteria of the Health Insurance P ortability and Accountability Act to protect personal information. Waiver of consent was granted by the University of California San Diego institutional review board.

Native American, Multi-race, and Unknown race had higher proportion of POD versus IOD (p < 0.001). The median (interquartile range [IQR]) age was 18 years old (15, 19 years old) and 19 years old (18, 20 years old) among patients with POD compared to IOD, respectively (p < 0.001). Compared to IOD, patients with POD have higher rates of Medicaid insurance (35% versus 21.1%, p < 0.001). Compared to POD, the rate of IOD was highest in Northeast (29.2% versus 14.3%, p < 0.001) and Midwest (31.6%versus 26.1%, (p < 0.001) regions of the country.

Table 3 illustrates the differences in patient comorbidities between POD versus IOD populations. Compared to patients with IOD, the rate of benzodiazepine poisoning, aromatic analgesic poisoning, depressive disorder, antidepressant poisoning, and suicidal ideation was higher in patients with POD. The rate of central nervous system stimulant poi- soning, opioid abuse history, alcohol, tobacco, and cannabis use were highest among IOD versus POD.

4. Discussion

Comorbidities ICD-9 diagnosis code

Prescription opioid overdose 965.00, 965.02, 965.09, E850.1, E850.2

Illicit opioid overdose 965.01, E850.0 Benzodiazepine-based tranquilizers

poisoning 969.4

Aromatic analgesics poisoning 965.4 Depressive disorder 311, 296.20, 296.33, 296.20 CNS stimulant poisoning 970.8 Antidepressant poisoning 969 Acute respiratory failure 5188.1 Pneumonitis 507 Suicidal ideation V628.4 Mixed drug abuse 305.9 Opioid abuse 305.50, 304.01 Metabolic acidosis 276.2 Alcohol abuse 305 Tachycardia 785 Tobacco use 305.1 Hallucinogen poisoning 969.6 Cannabis abuse 305.2 Rhabdomyolysis 728.88 Acute kidney injury 584.9 Hypokalemia 276.8 Anemia 285.9 Leukocytosis 288.6 Asthma 493.9 ADHD 314.01 Anxiety 300 Cardiac dysrhythmias 427.89 Cocaine Abuse 305.6 Hypotension 458.9

Procedures ICD-9 procedure codes

Pulmonary complication Mechanical ventilation (967.1, 967.2), acute respiratory failure (51881), respiratory arrest (799.1)

Insertion of Senstaken tube 960.4 Drug detoxification 946.5 Gastric lavage 963.3 Temporary tracheostomy 311

Abbreviations = ICD 9 (international classification of disease, ninth revision), CNS (central nervous system), ADHD (attention - deficit hyperactivity dis- order).

Table 1 International classification o f d iseases, n inth r evision c odes u sed t o extract medical history.

Adolescence is an important period of psychological and emotional development; an increased interest in subjective reward combined with an underdeveloped prefrontal cortex places adolescents at high risk for nonmedical opioid use and overdose, as they are seeking the social reward while unable to process fully the consequences of their actions [16]. Adolescents from low socioeconomic statuses have additional burdens which may predispose them to opioid dependence. Barriers to health care access and high-quality care for patients of low socio- economic status increases the risk of acquiring comorbid conditions that complicate successful pain management. Furthermore, adolescents at a socioeconomic disadvantage often face circumstances that predis- pose them to poor health and chronic pain in which prescription opioids may be indicated [17]. Due to these risks for opioid overdose, the CDC does provide detailed guidelines for prescribing opioids in the pediatric population [18].

Prescription opioids serve as an important gateway drug to more potent drugs as well as substance abuse in adulthood. Multiple studies have indicated that approximately 80% of heroin users reported using prescription opioids prior to abusing illicit opioids [19]. Addressing addiction during childhood or adolescence is critical for limiting the occurrence of IOD in addition to POD [20,21].

There are several important limitations in our study. Because we did

not analyze cases of opioid overdose that were successfully managed in the ED or following hospital discharge, this limits the generalizability of our results. Another limitation is the use of ICD-9-CM, whereby the accuracy of the data is subject to miscoding and omissions. However, the use of ICD-9-CM codes to identify overdoses from acetaminophen and other commonly used pediatric medications has been authenticated through validation studies that support the use of this coding for re- search analysis in the pediatric population [22]. Finally, the use of the KID database does not identify the original indication of the prescrip- tion opioids that resulted in hospitalization due to POD. Individuals who use prescription opioids for chronic pain may differ from those who use these medications for opioid dependence; however, we con- trolled for comorbidities to limit confounders that may have occurred due to potential differences.

In this retrospective study, we evaluated demographic and co- morbidity trends in opioid overdose in the inpatient setting. Our find- ings reinforce existing studies that report a continued rise in opioid morbidity and mortality while providing new insights into socio- demographic patterns clinical factors associated with POD versus IOD. Medical and public health interventions that target harmful prescrip- tion opioid exposure will likely provide a mutual advantage in curbing illicit opioid use in children and adolescents.

0.03

0.04

0.05

0.06

2000 20003 2006 2009 2012

Year of Admission

P re

va le

n ce

o f O

p io

id O

ve rd

o se

( %

)

Prevalence of Overall Opioid Overdose A

0.01

0.02

0.03

0.04

0.05

2000 2003 2006 2009 2012

Year of Admission

P re

va le

n ce

o f O

p io

id O

ve rd

o se

( %

)

Overdose IOD POD

Trend in Opioid Overdose B

Fig. 2. Prevalence of Opioid Overdose in the Hospitalized Pediatric Population. A) Rate of all opioid overdose admissions from 2000 to 2012, B) Rate of illicit opioid overdose (IOD) and prescription opioid overdose (POD) from 2000 to 2012.

Disclosures

No conflict of interest. Acknowledgements

No acknowledgements.

Ethics

No necessary ethical approval(s) required. Data is obtained from national administrative database and therefore is de-identified. This study was exempt from institutional review board approval.

Funding

The study was funded by departmental resources.

Reprints

Send requests to corresponding author.

References

[1] Hedegaard H, Warner M, Minino AM. Drug overdose deaths in the United States, 1999-2016. NCHS Data Brief 2017:1–8.

[2] Burton BN, Lin TC, Said ET, Gabriel RA. National trends and factors associated with inpatient mortality in adult patients with opioid overdose. Anesth Analg

Prescription opioid overdose

Illicit opioid overdose

p value

Total, n 13,348 2536 Race, n (%) < 0.001 White 8014 (60.0) 1777 (70.1) Black 1066 (8.0) 58 (2.3) Hispanic 941 (7.0) 181 (7.1) Asian or Pacific islander 152 (1.1) 14 (0.5) Native American 138 (1.0) a

Multi-race 392 (2.9) 58 (2.3) Unknown 2647 (19.8) 442 (17.4)

Age (years), (median [IQR]) 18 [15, 19] 19 [18, 20] < 0.001 Female gender, n (%) 6470 (48.5) 868 (34.2) < 0.001 Insurance status, n (%) < 0.001 Medicare 67 (0.5) a

Medicaid 4614 (35) 535 (21.1) Private 6057 (45) 1321 (52.1) Uninsured 1884 (14) 531 (20.9) No Charge 97 (1) 44 (1.7) Other 629 (5) 104 (4.1)

Weekend hospital admission, n (%)

3965 (30) 865 (34.1) < 0.001

Hospital region, n (%) < 0.001 Northeast 1915 (14.3) 741 (29.2) Midwest 3478 (26.1) 800 (31.6) South 5075 (38.0) 479 (18.9) West 2880 (21.6) 515 (20.3)

Hospital location and teaching, n (%)

< 0.001

Rural 1947 (14.6) 136 (5.3) Urban nonteaching 4809 (36.0) 1204 (47.5) Urban teaching 6239 (46.7) 1181 (46.6) Unknown 353 (2.6) 16 (0.6)

Admission year, n (%) < 0.001 2000 678 (5.1) 126 (5.0) 2003 2509 (18.8) 402 (15.9) 2006 2911 (21.8) 491 (19.4) 2009 3768 (28.2) 627 (24.7) 2012 3483 (26.1) 889 (35.1)

Abbreviations = IQR (interquartile range). a Cell count ≤10 were removed in accordance with Healthcare Cost and

Utilization Project to protect patient identity.

Table 3 Distribution of comorbidities among pediatric patients with opioid overdose.

Prescription opioid overdose

Illicit opioid overdose

P value

Total, n 13,348 2536 Benzodiazepine poisoning, n

(%) 3145 (23.6) 333 (13.1) < 0.001

Aromatic analgesics poisoning, n (%)

1781 (13.3) 39 (1.5) < 0.001

Depressive disorder, n (%) 3516 (26.3) 498 (19.6) < 0.001 CNS stimulant poisoning, n

(%) 554 (4.1) 235 (9.3) < 0.001

Antidepressant poisoning, n (%)

686 (5.1) 20 (0.8) < 0.001

Pneumonitis, n (%) 1475 (11.0) 541 (21.3) < 0.001 Suicidal ideation, n (%) 507 (3.8) 66 (2.6) 0.013 Mixed drug abuse, n (%) 1250 (9.4) 251 (9.9) 0.504 Opioid abuse history, n (%) 1049 (7.9) 913 (36.0) < 0.001 Metabolic acidosis, n (%) 785 (5.9) 278 (11.0) < 0.001 Alcohol abuse, n (%) 999 (7.5) 257 (10.1) < 0.001 Tachycardia, n (%) 468 (3.5) 89 (3.5) 0.968 Tobacco use, n (%) 2546 (19.1) 835 (32.9) < 0.001 Hallucinogen poisoning, n

(%) 1088 (8.1) 196 (7.7) 0.56

Cannabis abuse, n (%) 1275 (9.6) 307 (12.1) 0.002 Rhabdomyolysis, n (%) 572 (4.3) 256 (10.1) < 0.001 Acute kidney injury, n (%) 623 (4.7) 289 (11.4) < 0.001 Hypokalemia, n (%) 960 (7.2) 235 (9.3) 0.003 Anemia, n (%) 285 (2.1) 75 (3.0) 0.037 Leukocytosis, n (%) 320 (2.4) 154 (6.1) < 0.001 Asthma, n (%) 885 (6.6) 160 (6.3) 0.632 ADHD, n (%) 786 (5.9) 133 (5.2) 0.29 Anxiety, n (%) 708 (5.3) 180 (7.1) 0.004 Cardiac dysrhythmias, n (%) 645 (4.8) 154 (6.1) 0.034 Cocaine abuse, n (%) 423 (3.2) 245 (9.6) < 0.001 Hypotension, n (%) 352 (2.6) 144 (5.7) < 0.001 Pulmonary complications, n

(%) 2976 (22.3) 864 (34.1) < 0.001

Insert Senstaken tube, n (%) 1625 (12.2) 459 (18.1) < 0.001 Drug detoxification, n (%) 109 (0.8) 91 (3.6) < 0.001 Gastric lavage, n (%) 181 (1.4) 15 (0.6) 0.008 Tracheostomy, n (%) 49 (0.4) 25 (1.0) 0.001 Length of stay (days),

(median [IQR]) 2 [1, 3] 2 [1, 4] 0.001

Mortality, n (%) 212 (1.6) 72 (2.9) 0.001

Abbreviation = ADHD (attention-deficit hyperactivity disorder), CNS (central nervous system), IQR (interquartile range), SD (standard deviation).

Table 2 Distribution of characteristics among pediatric patients with opioid overdose.

65

[14] The National Institute on Drug Abuse at The National Institute of Health. The University of Michigan Institute of Social Research. Monitoring the future national survey results on drug use: 1975–2014: overview, key findings on adolescent drug use. http://www.monitoringthefuture.org/pubs/monographs/mtf-vol2_2016.pdf. Published July 2017. Accessed March 14, 2018.

[15] Osborne V, Striley CW, Nixon SJ, Winterstein AG, Cottler LB. Sex differences in patterns of prescription opioid non-medical use among 10-18year olds in the US. Addict Behav 2019;89:163–71.

[16] Blakemore SJ, Mills KL. Is adolescence a sensitive period for sociocultural proces- sing? Annu Rev Psychol 2014;65:187–207.

[17] Poleshuck EL, Green CR. Socioeconomic disadvantage and pain. Pain 2008;136:235–8.

[18] Schechter NL, Walco GA. The potential impact on children of the CDC guideline for prescribing opioids for chronic pain: above all, do no harm. JAMA Pediatr 2016;170:425–6.

[19] Jones CM. Heroin use and heroin use risk behaviors among nonmedical users of prescription opioid pain relievers - United States, 2002-2004 and 2008-2010. Drug Alcohol Depend 2013;132:95–100.

[20] Substance Abuse and Mental Health Service Administration. Associations of non- medical pain reliever use and initiation of heroin use in the United States. https:// www.samhsa.gov/data/sites/default/files/DR006/DR006/nonmedical-pain- reliever-use-2013.htm. Published August 2013. Accessed March 14, 2018.

[21] Al-Tayyib AA, Koester S, Riggs P. Prescription opioids prior to injection drug use: comparisons and public health implications. Addict Behav 2017;65:224–8.

[22] de Achaval S, Feudtner C, Palla S, Suarez-Almazor ME. Validation of ICD-9-CM codes for identification of acetaminophen-related emergency department visits in a large pediatric hospital. BMC Health Serv Res 2013;13:72.

B.

2019;128:152–60. [3] Office of the Chief Medical Examiner DC.gov. Opioid-related fatal overdoses:

January 1, 2014 to November 30, 2016. https://search.usa.gov/search?query= opioid&affiliate=dc_ocme&emb=1, Accessed date: 21 May 2018.

[4] Centers for Disease Control and Prevention. 2018 annual surveillance report of drug-related risks and outcomes — United States. Surveillance special report. 2018 [Published August 31, 2018. Accessed March 13, 2018].

[5] Gaither JR, Leventhal JM, Ryan SA, Camenga DR. National trends in hospitaliza- tions for opioid poisonings among children and adolescents, 1997 to 2012. JAMA Pediatr 2016;170:1195–201.

[6] McCabe SE, West BT, Boyd CJ. Leftover prescription opioids and nonmedical use among high school seniors: a multi-cohort national study. J Adolesc Health 2013;52:480–5.

[7] Sharma B, Bruner A, Barnett G, Fishman M. Opioid use disorders. Child Adolesc Psychiatr Clin N Am 2016;25:473–87.

[8] Healthcare Cost and Utilization Project - Kids' Inpatient Database (KID). Overview of the Kid's inpatient database (KID). https://www.hcup-us.ahrq.gov/kidoverview. jsp. Last Modified September 2018. Accessed March 14, 2018.

[9] Healthcare Cost and Utilization Project. KID database documentation. https://www. hcup-us.ahrq.gov/db/nation/kid/kiddbdocumentation.jsp. Last Modified October 18, 2018. Accessed March 14, 2018.

[10] Control CfD, Prevention. International classification of diseases, ninth revision, clinical modification (ICD-9-CM). URL: http://www/ cdc gov/nchs/about/ otheract/icd9/abticd9 htm [accessed 2004 Dec 16]. 2013.

[11] Cerda M, Gaidus A, Keyes KM, Ponicki W, Martins S, Galea S, et al. Prescription opioid poisoning across urban and rural areas: identifying vulnerable groups and geographic areas. Addiction 2017;112:103–12.

[12] Seth P, Rudd RA, Noonan RK, Haegerich TM. Quantifying the epidemic of pre- scription opioid overdose deaths. Am J Public Health 2018;108:500–2.

[13] Substance Abuse and Mental Health Service Administration. Behavioral health

trends in the United States: results from the 2014 National Survey on Drug Use and Health. https://www.samhsa.gov/data/sites/default/files/NSDUH-FRR1-2014/ NSDUH-FRR1-2014.pdf. Published September 2015. Accessed March 14, 2018.