VERY Morbidity and Mortality Weekly Report Weekly January 12, 2007 / Vol. 56 / No. 1 Infant Deaths Associated with Cough and Cold Medications — Two States, 2005 Cough and cold medications that contain nasal deconges- In January 2006, NAME, in collaboration with CDC, tants, antihistamines, cough suppressants, and expectorants initiated an e-mail inquiry, requesting reports of deaths in commonly are used alone or in combination in attempts to infants aged <12 months for which cough and cold medica- temporarily relieve symptoms of upper respiratory tract tions were determined as the underlying cause. To identify infection in children aged <2 years. However, during 2004- additional cases, CDC examined media and medical-journal 2005, an estimated 1,519 children aged <2 years were treated reports of infant deaths suspected to be linked to cough and in U.S. emergency departments for adverse events, including cold medications during 2005. A total of 15 local medical overdoses, associated with cough and cold medications.* In examiners in 12 U.S. states and Canada responded to the response to reports of infant deaths after such events, CDC NAME survey. However, no cases other than those from and the National Association of Medical Examiners (NAME) media and published reports were identified. From these investigated deaths in U.S. infants aged <12 months associ- reports, CDC identified three cases of infant deaths in two ated with cough and cold medications. This report describes states during 2005 that were determined by a medical the results of that investigation, which identified deaths of examiner or coroner to have been caused by cough and cold three infants aged <6 months in 2005, for which cough and medications (Table 1). cold medications were determined by medical examiners or he three infants ranged in age from 1 to 6 months; two coroners to be the underlying cause. The dosages at which were male. All three infants had what appeared to be high cough and cold medications can cause illness or death in chil- levels of pseudoephedrine (a nasal decongestant) in postmor- dren aged <2 years are not known. Food and Drug tem blood samples. The blood levels of pseudoephedrine Administration (FDA)-approved dosing recommendations for ranged from 4,743 ng/mL to 7,100 ng/mL.’ One infant clinicians prescribing cough and cold medications do not (patient 2) had received both a prescription and an over-the- exist for this age group. Because of the risks for toxicity, counter cough and cold combination medication at the same absence of dosing recommendations, and limited published time; both medications contained pseudoephedrine (Table 1). eV idence of effectiveness of these medications in children aged In pharmacokinetic studies of children aged 2-12 years, the mean maximum <2 years, parents and other caregivers should not administer plasma concentrations of pseudoephedrine after therapeutic doses ranged from cough and cold medications to children in this age group with- 180 ng/mL to 500 ng/mL and were comparable to adults with current dosing out first consulting health-care provider and should follow regimens (FDA, unpublished data, 2006 the provider's instructions precisely (/). Clinicians should use caution when prescribing cough and cold medications to chil- INSIDE dren aged <2 years. Moreover, clinicians should always ask 4 National and State Medical Expenditures and Lost caregivers about their use of over-the-counter combination Earnings Attributable to Arthritis and Other Rheumatic medications to avoid overdose in children from multiple medi- Conditions — United States, 2003 cations that contain the same ingredient. Suicide Trends and Characteristics Among Persons in the Guarani Kaiowa and Nandeva Communities — Mato * Estimated from the National Electronic Injury Surveillance System—Cooperative Grosso do Sul, Brazil, 2000-2005 Adverse Drug Events Surveillance project, which is jointly operated by CDC, the Notices to Readers Food and Drug Administration, and the Consumer Product Safety Commission DEPARTMENT OF HEALTH AND HUMAN SERVICES CENTERS FOR DISEASE CONTROL AND PREVENTION MMWR January 12, 2007 lhe other two infants also had received pseudoephedrine- The MMWR series of publications is published by the Coordinating containing medications (one prescription and one over the Center for Health Information and Service, Centers for Disease 7] ( : 12) had | Control and Prevention (CDC), U.S. Department of Health and counter). lwo of the infants (patients and <) Nad been Human Services, Atlanta, GA 30333. administered prescription medications containing carbinox- Suggested Citation: Centers for Disease Control and Prevention. amine (an antihistamine), although neither had detectable [Article title]. MMWR 2007;56:[inclusive page numbers). postmortem blood levels of carbinoxamine. Two of the Centers for Disease Control and Prevention infants (patients 2 and 3) had detectable blood levels of Julie L. Gerberding, MD, MPH dextromethorphan (a cough suppressant) and acetaminophen Director (an antipyretic and analgesic Tanja Pi nic, MD, PhD ‘ ros ee Chief $ : Off ; All three infants were found dead in their homes. Autopsy \/ie ing Mey Science f ce lames W. Stephens, PhD and medical investigation records were obtained. A medical (Acting) Associate Director for Science examiner or coroner determined that cough and cold medica Steven L. Solomon, MD tion was the underlying cause of death for each of the three. Director, Coordinating Center for Health Information and Service None of the deaths were determined to be intentional. On Jay M. Bernhardt, PhD, MPH - ' ;, a? Sh heal eat D, irector, Na, tional Center for Health Marketing auuttooppssyy, twoo of of the infants p(aptatiieennt s | aandn c 2) had d eveividdeennc e o Judith R. Aguilar respiratory infection; no abnormalities in cardiac pathology . | (Acting) Director, Division of Health Information Dissemination (Proposed) were revealed in any of the Editorial and Production Staff Reported by: Frederic E. Shaw, MD, JD Vational Assoc of Editor, MMWR Series John S. Moran, MD, MPH i d ter jor lref Vetection, an Guest Editor, MMWR Series l es (proposed Col UD, EIS Offices Suza > M. He , MPA a . . Managing Editor scout Serie Editorial Note: Cough and cold medications (Table ee. ee widespread use to treat children, and the overall incidence of (Acting) Lead Technical Writer-Editor reported adverse events has been low. An estimated 1,519 chil- Catherine H. Bricker, MS dren aged <2 years we reated for adverse events from such Jude C. Rutledge medications during 2004—2005; infant deaths, although BevWerriltye rs]-.E dHiotlolras nd uWs'e og |N av‘e bDye en reportI j . The1 € casISeEs dje scrib1 e) d | inin tthhi RS Lead Visual Information Specialist report suggest that such deaths continue to be reported LyndaG . Cupell underscore the for clinicians to use caution when pre Malbea A. LaPe | Ibea A. LaPere scribing and caregivers t se caution when administering Visual Information Specialists Quang M. Doan. MBA cougihg n aann d cokl d medications to chcihillddrreenn aaggeecd <2 eayreasrs. rica R. Shaver dren aged >2 years, FDA approval of the use of overt Information Technology Specialists ed|i catti ons is }D ased laonn revViiee w oO! Editorial Board safety and efticac ita by an external advisory review panel. William L. Roper, MD, MPH, Chapel Hill, NC, Chairman However, in chcihilldd iv ed i years,I systematic revieowf sco n Virginia A. Caine, MD, Indianapolis, IN trolled trialso f « nter cough and | cold medications David W. Fleming, MD, Seattle, WA =e = ad er. il ape William E. Halperin, MD, DrPMPHH, N,ewar k, NJ have concluded 1 NOS. REE EELS Ce Fy oe Margaret A. Hamburg, MD, Washington, D¢ reducing acute cough (5) and other symptoms of King K. Holmes, MD, PhD, Seattle, WA ratory tract infte ct Because of the unproven Deborah Holtzman, PhD, Aclanta, GA ; | 2 j John K. Iglehart, Bethesda, MD the cough suppressants codeine and dextrome thorphan in Desnis G) Mai MD. Madicon. Wl young children and the potential for adverse events, in 1997 Sue Mallonee, MPH, Oklahoma City, OK the American A ademy of Pediatrics issued a policy statement Stanley A. Plotkin, MD, Doylestown, PA ivising ti a ee | 1 regarding the lack of Patricia Quinlisk, MD, MPH, Des Moines, IA ek. ea per eager imeien te pw,.comdber ions Patrick L. Remington, MD, MPH, Madison, WI intitussive effects, risk for adverse events, and pote ntial for Barbara K. Rimer, DrPH, Chapel Hill, N¢ overdose in children fron these medications (7). In 2006, the John V. Rullan, MD, MPH, San Juan, PR | American College of Chest Physicians released clinical prac- Anne Schuchat, MD, Atlanta, GA Li a Dixie E. Snider, MD. MPH. Atlanta, GA tice guidelines for management of cough, advising health-care John W. Ward, MD, Atlanta, GA providers to refrain from 1r ecommending cough suppressants Vol. 56 / No. 1 MMWR TABLE 1. Reported infant deaths for which cough and cold medications were determined the underlying cause of death* — United States, 2005 Significant Nasal Other medication medical conditions, decongestant Antihistamine (e.g., cough Age Underlying contributing factors, postmortem postmortem suppressant or antipyretic) Patient (mos) Sex cause of deatht and findings on autopsy blood levels blood levels postmortem blood levels 1 1 Male Pseudoephedrine Interstitial pneumonia, recent Pseudoephedrine None detected None detected intoxication hospitalization for fever 4,743 ng/mL Female Pseudoephedrine Bronchopneumonia and Pseudoephedrine None detected Dextromethorphan and dextromethorphan empyema on autopsy 6,832 ng/mL 1,909 ng/mL, intoxication acetaminophen 35 pg/ml Drug poisoning Infant found lying in crib in Pseudoephedrine Doxylamine Dextromethorphan prone position, reported history 7,100 ng/mL 1,000 ng/mL 390 ng/mL, of colic, born preterm (33 weeks) acetaminophen small fracture of left distal tibia, 1.9 yo/mL acute anoxic encephalopathy on autopsy * As determined by medical examiner or coroner The three infants were known to have received the following medications: patient 1 received a prescription medication containing pseudoephedrine, carbinoxamine, and dextromethorphan; patient 2 received a prescription medication containing pseudoephedrine, carbinoxamine, and dextromethorphan and also received a nonprescription medication containing pseudoephedrine and acetaminophen; patient 3 received a nonprescription medication containing pseudoephedrine and acetaminophen. The nonpre- scription medications might also have contained other ingredients; exact formulations are unknown TABLE 2. Examples of common ingredients in cough and cold medications that contain pseudoephedrine, which can be used medications, by class of medication to make methamphetamine. Because of this act, pseudoephe- Class Examples drine has been removed as an ingredient in many cough and Antihistamine Acrivastine, brompheniramine, carbinoxamine, cold medications and replaced with other nasal decongestants. (first generation) chlorpheniramine, cyproheptadine, However, some pediatric cough and cold medications con- diphenhydramine, doxylamine, triprolidine taining pseudoephedrine still might be sold behind the counter. Antipyretic Acetaminophen, ibuprofen As an alternative to pseudoephedrine and other nasal decon- and analgesic gestants, caregivers might consider clearing nasal congestion Cough suppressant Benzonatate, codeine, dextromethorphan, in infants with a rubber suction bulb; secretions can be soft- (antitussive) hydrocodone ened with saline nose drops or a cool-mist humidifier. Expectorant Guaifenesin Few data exist regarding the therapeutic or toxic levels of Nasal decongestant Ephedrine, phenylephrine, cough and cold medications in children aged <2 years (2,3, 10). phenylpropanolamine, pseudoephedrine Blood levels of cough and cold medications revealed in post- mortem studies might not reflect levels in the bloodstream at and other over-the-counter cough medications for young the time of administration (/). However, in this report, the children because of associated morbidity and mortality (8). blood levels of pseudoephedrine found in the three patients In addition to advising caregivers and health-care providers aged 1-6 months were approximately nine to 14 times the regarding the risks of administering cough and cold medica- levels resulting from administration of recommended doses tions to children aged <2 years, public health officials have to children aged 2-12 years. taken steps to improve the safety of these medications. On The findings in this report are subject to at least two limita- June 8, 2006, FDA took enforcement action to stop the manu- tions. First, because no universally accepted criteria exist for facture of carbinoxamine-containing medications that had not attributing deaths to cough and cold medications, the cause of been approved by the agency; FDA noted that many of the death in these cases was based on the report of the medical exam- medications were inappropriately labeled for use in infants iner or coroner. However, the actual cause of death might have and young children despite safety concerns regarding use of been overdose of one drug, interaction of different drugs, an carbinoxamine in children aged <2 years (9). Although manu- underlying medical condition, or a combination of drugs and facturers were required to cease production by September 6, underlying medical conditions. Second, the findings are limited 2006, some products might still be in distribution. In by the low response rate and absence of identified cases from the another action, the availability of pseudoephedrine- NAME survey, which might underestimate the number of deaths containing medications has been affected by the federal Com- in infants attributed to cough and cold medications. bat Methamphetamine Epidemic Act, which was signed into No FDA-approved dosing recommendations exist for law March 9, 2006. This act bans over-the-counter sales (but administering over-the-counter cough and cold medications permits behind-the-counter sales in limited amounts) of cold MMWR January 12, 2007 to children aged <2 years, and proper dosing for children in National and State Medical this age group has not been studied. Instructions on over-the- Expenditures and Lost Earnings counter medications advise consumers to “consult a doctor” for children in th} is age group (/). Suggested dosing for some Attributable to Arthritis ough and cold medications can be found in parenting and and Other Rheumatic Conditions — yrescribing guides, ind clinicians commonly extrapolate a dose United States, 2003 based on the weight or age of children aged <2 years from Arthritis is the leading cause of disability in the United States dosing guidelines for adults and older children (7). Such (1), potentially limiting affected persons from walking a few extrapolation is based on the assumption that the pathophysi blocks or climbing a flight of stairs. Using Medical Expendi- logy of the disease and the effects of the drug are similar in ture Panel Survey (MEPS) data, CDC analyzed national and idult and pediatric patients. state-specific direct costs (i.e., medical expenditures) and Caregivers and clinicians should be aware of the risk for indirect costs (i.e., lost earnings) attributable to arthritis and rious illness or fatal overdose from administration of cough j | 1 other rheumatic conditions (AORC) in the United States dur- ind cold medications to children aged <2 years. Caregivers q ! ] ing 2003. This report describes the results of that analysis, hould only administer cough and cold medications to chil which indicated that, in 2003, the total cost of AORC in the rroup when following the exact advice of a United States was approximately $128 billion ($80.8 billion linicians should be certain that caregivers under in direct and $47.0 billion in indirect costs), equivalent to mnppo< rtana liceO Ft ad | ministering cough and cold medi 1 1.2% of the 2003 U.S. gross domestic product. Total costs ind 2) th risk for overdose tf they attributable to AORC, by state/area, ranged from $225.5 ynal medications that might contain the same : ,; ; ' ' million in the District of Columbia to $12.1 billion in Calli- s should always inform their health-care z | " 1] fornia. Total costs attributable to AORC have increased sub tions they are administetor ai cnhigld . stantially since 1997 and that increase is expected to continue because of the aging of the population and increases Acknowledgment © } OF ) ' in obesity and physical inactivity. These findings signal the need for broader implementation of effective public health Aan interventions, such as arthritis and chronic disease self References management programs, which can reduce medical expendi- tures (3) among persons v ith AORG National direct and indirect costs were derived from the house 1 hold component of the 2003 MEPS (MEPS-H¢ , an annual household interview surve medical conditions, medical sys tem expenditures and utilization, and earnings and employ ment history (4). MEPS signed to be representative of the U.S. civilian, noninstitutionalized population; each year’s MEPS panel is a subsample of the previous year’s National Health Interview Survey. The 2003 MEPS did not include a nursing home component; thus, costs among nursing home residents were excluded from the analysis. During the household inter view, MEPS respondents described all medical conditions for which they had sought care from a health-care provider. Each of these medical conditions was later assigned an /nternational Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) code by medical coders. The 2003 MEPS-H(¢ response rate was 71.9%. AORC cases in MEPS were identi fied using three-d'eit ICD-9-CM codes selected by the National Arthritis Data Workgroup* (5). The 2003 MEPS sample con sisted of 23,352 participants aged $18 years, and 4,801 partici pants met the case definition for AORC. Vol. 56 / No. 1 MMWR Direct per-person costs attributable to AORC were estimated reporting AORC, and average per-person direct costs of $1,752 using a series of four-stage regression analyses (6) that mod- (Table 1). Ambulatory care accounted for the highest per-person eled the probability and magnitude of medical care expendi- direct costs ($914), followed by emergency department and tures among adults aged >18 years. This modeling included inpatient services ($352), prescriptions ($338), and other costs adjustment for the following variables: age (18-44, 45-64, or ($146) (Table 1). Total indirect costs attributable to AORC were >65 years), sex, race (white or nonwhite), ethnicity (Hispanic $47.0 billion; average per-person lost earnings were $1,590 among or non-Hispanic), marital status (single, currently married, 29.5 million working-age adults. National direct and indirect widowed, separated, or divorced), highest educational attain- costs totaled $128 billion. Among states/areas, total costs attrib- ment (less than high school, high school graduate, some col- utable to AORC ranged from $225.5 million in the District of lege, college graduate, or graduate school), health-insurance Columbia to $12.1 billion in California; New York and Texas status (no insurance, public insurance only, or any private in- had the next highest total costs at $8.7 billion (Table 2). surance), and the presence of nine other high-cost chronic Reported by: F Yelin, PhD, Univ of California at San Francisco; conditions (hypertension, other forms of heart disease, pul- M Cisternas, MA, A Foreman, MA, D Pasta, MS, Ovation Research monary disease, stroke, other neurologic conditions, diabe- Group, San Francisco, California. L Murphy, PhD, CG Helmick, MD, tes, cancer, mental illness, or non-AORC musculoskeletal Div of Adult and Community Health, National Center for Chronic conditions). The average per-person direct cost attributable Disease Prevention and Health Promotion, CDC. to AORC was the difference between the observed and corre- Editorial Note: The analysis described in this report estimates sponding expected medical costs. Expected costs simulated total AORC-attributable costs for 2003 at $128 billion. Per- costs among persons with AORC as ift hey did not have AORC person direct and indirect estimates were derived from a re- (2). Average per-person direct costs were generated for overall gression-based econometric approach, which enabled expenditures and for each of the following four cost catego- adjustment for various confounders (e.g., comorbidities) and ries: 1) ambulatory care, 2) emergency department and inpa- thus allowed estimation of the costs attributable to AORC. tient services, 3) prescriptions, and 4) other costs (i.e., home National direct costs attributable to AORC increased 24% health care, vision aids, dental visits, and medical devices). during 1997-2003, from $64.8 billion (adjusted to 2003 Finally, total national direct costs were calculated as the prod- dollars using the medical care component of the Consumer uct of the number of persons aged >18 years reporting AORC Price Index) to $80.8 billion (2). Despite changes in the treat- and the average per-person direct costs. ment of persons with AORC during 1997-2003, including Indirect per-person costs attributable to AORC were introduction of expensive medications in two drug classes, derived from a similar four-stage analysis that modeled the the coxibs" (e.g., Celebrex. “ [Pfizer Pharmaceutical ompany; probability of employment and the magnitude of lost earn- New York, New York]) and biologics (e.g., TINF® inhibitors), ings among persons aged 18-64 years. However, age was cat- Cyclooxvgenase (COX)-2-selective inhibitors egorized differently (18-34, 35—44, 45-54, and 55—64 years), lumor necrosis factor and no adjustment was made for health-insurance status. Total indirect costs were the product of the number of per- TABLE 1. National direct and indirect costs attributable to sons aged 18-64 years with AORC who had ever worked and arthritis and other rheumatic conditions (AORC), by cost the average lost earnings per person attributable to AORC. category — Medical Expenditure Panel Survey, United States, All analyses were conducted using statistical software that 2003 adjusted for the clustered sampling design of MEPS. The sta- No. of adults Average Total costs affected cost per (in billions) tistical methods used to derive the national direct and indi- Cost category (in millions) person (S$) (S) rect cost estimates are described elsewhere (2). Direct (medical expenditures) 46.1" 1,752 80.8t Direct and indirect costs for each state were estimated by Ambulatory care 914 42.1 Emergency department applying the state’s proportion of overall doctor-diagnosed and inpatient care - 352 16.2 arthritis (from the 2003 Behavioral Risk Factor Surveillance Prescription --- 338 15.6 System [BRFSS]) to the MEPS-derived national cost estimates. Others 146 6.7 State-specific direct costs were estimated for those aged >18 Indirect (lost earnings) 29.51 1,590 47.0 years and lost earnings for the working-age population aged * Persons aged >18 years Medical costs for each of the four categories do not total to $80.8 billion 18-64 years. The 2003 BRFSS response rates ranged from because estimates for each category were derived from separate regres- 34.4% to 80.4% among states. sion models; the discrepancy results from consolidation of variance across . regression models In 2003, total direct costs attributable to AORC were $80.8 ~ Includes home health care, vision aids, dental visits, and medical devices billion, with an estimated 46.1 million persons aged >18 years Persons aged 18-64 years with AORC who had ever worked MMWR January 12, 2007 TABLE 2. Direct and indirect costs attributable to arthritis and other rheumatic conditions, by state/area — Medical Expenditure Panel Survey and Behavioral Risk Factor Surveillance System (BRFSS), United States, 2003 Direct costs (medical expenditures) Indirect costs (lost earnings) No. of adults affected Costs (in millions) No. of adults affected Costs (in millions) Total costs (in millions) State/Area (in thousands)* (S)* (in thousands) (S) ($) 1,617.9 763 978.9 2,596.8 160.0 114.6 274.7 1,517.9 824.6 2,342.5 908.5 415 532. 1,441.3 863.3 3.4 4,273.3 12,136.6 1,186.2 : 733.3 1,919.5 343.1 389 499 ¢ 1,442.5 232.7 ( 130 363.4 146 ) 225.5 7,623.5 3,910.7 375.3 563.6 2,670.0 3,180.9 1,249.7 1,106.0 2,426.4 2,036.3 650.0 2,479.3 2,734.2 5,557.2 2,171.9 1,494.5 2,873.7 395.6 757.0 1,021.6 573.5 3,543.7 770.0 8,725.5 4,111.6 284.7 5,745.0 1,627.7 1,609.1 6,577.7 510.8 2,132.9 351.4 3,271.3 8,706.2 819.9 289.8 3,465.9 2,786.9 1,188.2 2,444.9 243.1 United States - 80.800.0 127, 800.0 Vol. 56 / No. 1 MMWR and an increase in the number of knee and hip joint replace- 3. Kruger JM, Helmick CG, Callahan LF, Haddix AC. Cost-effectiveness of the arthritis self-help course. Arch Intern Med 1998;158:1245-9. ments performed, the average per-person direct costs were simi- 4. Agency for Healthcare and Research Quality. Medical Expenditure Panel lar ($1,762 in 1997 [in 2003 dollars] and $1,752 in 2003). Survey—household. Available at http: www.meps.ahrq.gov/mepsweb The increase in total direct costs resulted from the increase in survey_comp/householdjs.p . . CDC. Arthritis prevalence and activity limitations—United States, 1990. the number of persons (9 million) with AORC in 2003, at- MMWR 1994;43:433-8. tributable to the increase in population (predominantly in the . Duan N, Manning W, Morris C, Newhouse J. Comparison of alterna ages 45-64 years cohort) and the increased prevalence of tive models for the demand of medical care. Journal of Business and self-reported AORC among adults aged >50 years (2). Economic Statistics 1983;1:115—26. 7. Quandt SA, Chen H, Grzywacz JG, Bell RA, Lang W, Arcury TA. Use The findings in this report are subject to at least two limita- of complementary and alternative medicine by persons with arthritis: tions. First, direct costs likely were underestimated because results of the National Health Interview Survey. Arthritis Rheum MEPS does not capture costs associated with complementary 2005:53:748-—55. 8. Hootman JM, Helmick CG. Projections of US prevalence of arthritis and alternative medicines (persons with arthritis are among and associated activity limitations. Arthritis Rheum 2006;54:226-9. the major consumers of these medicines [7]), long-term men- tal health services, and nondurable medical goods. Similarly, indirect costs did not capture loss of unpaid work such as homemaking, child care, and volunteer work. Other expenses Suicide Trends and Characteristics associated with treatment of illness, such as transportation, Among Persons in the accommodation, and lost wages among family members were not measured by MEPS and therefore were not included in Guarani Kaiowa and Nandeva this analysis. Second, state-specific direct and indirect cost Communities — Mato Grosso do Sul, estimates were simply derived using state-level prevalences and Brazil, 2000-2005 were not adjusted for differences among states in provider Suicide rates among indigenous communities around the charges, treatment resources, or wage differentials. Thus, costs world vary substantially; in many nations these groups have among states with medical expenditures or wages higher than the highest suicide risk of any identifiable cultural or ethnic the national mean likely are underestimated, and costs among group (/). Mato Grosso do Sul is a state in the southwest cor- those below the mean overestimated. Deriving cost estimates ner of Brazil that borders Bolivia and Paraguay. In 2004, the from state-ley el MEPS data was not possible because these Guarani, an indigenous ethnic group in the region (Figure 1), data were available for only 30 of the largest states. accounted for 2.6% of Mato Grosso do Sul’s population lhe substantially increased costs of AORC in 2003 were (approximately 2,230,702).* During 1975-2000, the infant driven by an increase in number of persons with AORC. Costs mortality rate decreased, and overall life expectancy increased likely will continue rising because the number of persons with in Mato Grosso do Sul; however, suicide increased as a propor- arthritis is projected to continue to increase, with another 8 tion of overall mortality among the Kaiowa and Nandeva com- million arthritis cases anticipated during 2005-2015 (8). munities of the Guarani population (2). In 2000, the National Without cost-reduction strategies, the economic burden of Health Foundation (FUNASA) of the Brazilian Ministry of AORC will continue to increase. This trend underscores the Health (BMH) initiated a study of suicide trends and charac- need for wide-scale implementation of interventions that re- teristics in these two Guarani communities; data were collected duce medical expenditures and lost earnings among persons during 2000-2005, and epidemiologic assistance was provided with AORC. Self-management programs such as the Arthritis by CDC. This report summarizes the results of that study, which Self-Help Program are cost-effective strategies to reduce di- suggested that the suicide rate among Guarani was 19 times rect costs associated with arthritis (3). Self-management pro- higher than the national rate in Brazil and 10 times higher than srams foster skills in coordinating work accommodations and the rate in Mato Grosso do Sul and that suicides disproportion- pain management (through physical activity and weight man- ately affected Guarani adolescents and young adults. To agement) and are essential for reducing the economic and so- decrease suicide rates, BMH initiated research and prevention cietal burden of AORC. programs among the Guarani, and the Guarani initiated mea- References sures to increase their economic self-sufficiency. 1. CDC. Prevalence of disabilities and associated health conditions among adults—United States, 1999. MMWR 2001;50:120-—5 2. Yelin E, Murphy L, Cisternas M, Foreman A, Pasta D, Helmick ¢ Medical care expenditures and earnings losses of persons with arthritis iro de Geografia e Estatistica. Population estimates for Brazilian and other rheumatic conditions in 2003 with comparisons to 1997 2004. Available at http://www.ibge.gov.br/english/estatistica Arthritis Rheum 2007. In Press tiva2004/de| taultIr.ss} htm MMWR January 12, 2007 FIGURE 1. Geographic distribution of Guarani population — (hanging); three were caused by pesticide ingestion. Annual Mato Grosso do Sul, Brazil, Argentina, and Paraguay suicide rates in the Guarani ranged from 121.5 per 100,000 population for males in 2000 to 113.2 in 2005 and from 63.7 per 100,000 population for females in 2000 to 59.1 in 2005 Brazil 5 (Figure 2). The ratio of male to female suicide rates was wae approximately 2 to | In 2005, the overall Guarani suicide rate was 86.3 per 100,000 population. By comparison, in 2004, the most recent year for which Brazil national and state data were available, suicides represented 1.5% of deaths in Mato Grosso do Sul and 0.8% of deaths in Bolivia Brazil (4). Therefore, the proportion of deaths caused by sui cide in the Guaran{ population during 2000-2005 was Mato Grosso do Sul approximately six times the proportion in Mato Grosso do Sul and 12 times the proportion overall in Brazil in 2004. The sui cide rate overall in Mato Grosso do Sul was 8.6 }p er 100,000 population, and the national rate was 4.5. The Guarani suicide rate in 2005 was aptp roxiimm ately 10 timetsh e ratei n Mato Grosso do Sul and 19 times 1 val rate in 2004. P)e rsons a" ged <30 iccoumnetr ed }f or 70% of the Guarani j ) population and 85 suicides. Among persons aged 20-29 years in 2005, the suicide rate was 159.9 per 100,000 population; among perso! 20 years, suicide accounted for 28.0% of the deaths suicide rate was 90.3 pet study were from the Kaiowa and Nandeva 100,000 population rani popi ulation coll1} ectitHv ely v rreetfeer rred | Reported by: ( this report. FUNASA medical teams dete | J using Categories from the / Healt was defined as a de poisoning odes X60—X84 and Y87 drawn from routin mv FIGURE 2. Suicide rates in Guarani population, by sex and year — Mato Grosso do Sul, Brazil, 2000-2005 y etnnograpnik iWOormation ON Ca ncluded political iI t ronm have beet suicide 2000, intormatio re was avail ises Of suicide, and information on sex one. Case. T) iLJemogr, apniL, c Goant a |f or a|e atn|s omplet Crude rates and ind sex 100,000 population 000 OOS 1 total ol 5004 deaths were recorde| d population in Mato Grasso do Sul, including ides; 190 (66 of the suicide victims were *Per 100,000 populatior ree suicides were caused Dy suffocation SOURCE: National He Vol. 56 / No. 1 MMWR Editorial Note: The findings in this report indicate that youths groups, crisis response, recreational activities, volunteer sup- and young adults in Guarani communities had higher suicide port systems, and cultural-heritage education. For example, a rates than older members of their population and that the suicide-prevention program for an American Indian commu- Guarani suicide rate overall was higher than the rates in Mato nity in the United States included a comprehensive strategy Grosso do Sul and Brazil overall. In the United States, suicide involving schools, community outreach to persons at risk for rates among some American Indian and Alaska Native (AI/AN) suicide and their families, improved infrastructure of local indigenous populations have been similar to those of the overall mental health services, and interventions addressing common U.S. population, whereas in certain other AIl/AN popuiations, suicide risk factors in the community (e.g., alcohol abuse, rates have been seven times higher than the U.S. rate overall (5). family violence, and unemployment) (/0). Historically, the highest suicide rate nationally in Brazil has To better address health disparities among indigenous com- been among those aged >65 years (6). However, among the munities, BMH established DSEI and placed multidisciplinary Guarani, the highest rates were among adolescents and young indigenous health-care teams in municipal governments. These adults. Among Guarani aged 20-29 years, the 2005 suicide teams periodically visit area villages and, with the support of rate was 159.9 per 100,000 population, compared with 6.1 trained local residents, provide health care. In 2000, FUNASA for the same age group nationwide in 2004 (4). began a mental health initiative to address alcohol abuse and Previous studies have identified multiple factors that might suicidal behaviors in indigenous ethnic communities. This be associated with suicidal behaviors among the Guarani. These initiative involved developing the Monitoring Center for factors, some resulting from colonization, include rapid Indigenous Mental Health, which includes physicians, behav- sociocultural change, disturbances in traditional social life, ioral scientists, social workers, and linguists. The center sup- progressive dismantling of extended family structure, and ports research and encourages a multidisciplinary approach forced relocation to reservations (7). Exposure to different and to assessing and preventing suicidal behavior. Its projects are conflicting cultures, perspectives, and belief systems exacer- developed with community participation and designed so that bates challenges faced by these communities and contributes they can be sustained by the local community. In addition, to intergenerational conflict. Studies conducted among other the Guarani have been organizing to recover, through court indigenous communities indicate that youths often perceive challenges and legislation, most of their former territories; this themselves as marginalized from mainstream society and their is expected to increase their economic self-sufficiency and re- own communities (7), resulting in a sense of social isolation duce poverty and unemployment, improvements associated with decreased risk for suicide and suicidal behaviors (9). that might contribute to an increased rate of suicide. The findings in this report are subject to at least four limita- References 1. Leenaars AA. Suicide among indigenous peoples: introduction and call tions. First, counting the number of suicides based on death to action. Arch Suicide Res 2006;10:103-15 certificates might underestimate the true number because of Coloma C, Hoftman |S, Crosby A. Suicide among Guarani Kaiowa misclassification of cause of death (8). Many decedents were und Nandeva youth in Mato Grasso do Sul, Brazil. Arch Suicide Res 2006:10:191—207 buried privately by their families; in 2000, 16.5% of deaths World Health Organization. International statistical classification of among the Guarani occurred without medical attention or cer- diseases and related health problems, 10th rev. 2nd ed. (ICD-10 tification by a coroner or medical examiner, so the cause of Geneva, Switzerland: World Health Organization; 2004 death could not be established. In 2001, DSEI began provid- Brazil Ministry of Health. Information mortality system (DATASUS), Ministry of Health. Available at http://www.datasus.gov.br ing financial support for burials, thereby increasing death regis- CDC. Injury mortality among American Indian and Alaska Native trations, and by 2003, cause of death could not be established hildren and youth—United States, 1989-1998. MMWR 2003;52 for only 5.7% 0 of deaths. Second, because data were not specifi- 69 0] Mello Santos ¢ Bertolote IM W YP. I pidemiology of suicide in cally collected at the municipal or national level, suicide rates of Brazi 1 1980-2000 cl haracterization of age and | gender Irt ates of indigenous and nonindigenous populations could not be com- Rev Bras Psiquiatr 2005;27:131 pared. Third, temporal data were limited and insufficient fot L, Brym RJ. The return of the native: a cultural and psycho comprehensive analysis. Finally, the small absolute number of ique of Durkheim's suicide based on the Guarani-Kaiow4a of estern Brazil. Sociological Theory 2006;24:42—57 suicides in the Guarani limited the detail of this analysis. rs CD, Fat DM, Inoue M, et al. Counting the dead and what Suicide is a complex, multifaceted problem influenced by { 1 from: an assessment of the global status of cause of death risk factors among persons, families, communities, and soci- data. Bull World Health Organ 2006;83:171 Advisory Groupon Suicide Prevention. Actingo n what we know: pre eties. Studies on indigenous populations in other nations have venting youth suicide in first nations. Ottawa, Canada: Health Canada found that community-based, comprehensive suicide- ind the Assembly of First Nations; 2003 prevention programs are the most promising (9). Such pro- May PA, Serna P, Hurt L, DeBruyn LM. Outcome evaluation of a grams vary, but strategies typically include counseling, support public health approach to suicide prevention in an American Indian tribal nation. Am ] Pub Health 2005;95:1238-—44 10 MMWR January 12, 2007 Notice to Readers References 1. CSTE. Position statement No. 06-1D-15. Inclusion of poliovirus infection Changes to National Notifiable Infectious reporting in the National Notifiable Diseases Surveillance System. Avail able at http://www.cste.org/PS/2006pdts/PSFINAL2006/06-ID-15FINAI pdf Disease List and Data Presentation, CSTE. Position statement No. 06-1D-05. National reporting for non as of January 2007 cholera Vibrio infections (Vibriosis). Available at http://www.cste.org PS/2006pdfs/PSFINAL2006/06-ID-05 FINAL. pdf. | his issue of MM\ R incorporates modifications to Table | CDC. Case definitions for nationally notifiable infectious diseases. Avail Provisional cases of infrequently reported notifiable diseases, able at http://www.cdc.gov/epo/dphsi/nndsshis.htm. United States), Table II (Provisional cases of selected notifi- CSTE. Position statement No. 06-ID-14. Enhancing local, state, and territorial-based surveillance for invasive pneumococcal disease in chil ible diseases, United States), and Figure I (Selected notifiable dren less than five years of agi Available at http www.cste.org/PS disease reports, United States, comparison of provisional 2006pdts/PSFINAL2006/06-ID-14FINAL.pdf i-week totals with historical data). This year, the modifica- Averoff F, Zucker J, Vellozzi C, et al. Adequacy of surveillance to detect endemic rubella transmission in the United States. Clin Infect Dis tions add conditions designated as nationally notifiable by the 2006;43(Supp! 3):S152 Council of State and Territorial Epidemiologists (CSTE) in CDC. Achievements in public health: elimination of rubella and 1 con conjunction with CDC (/,2). genital rubella syndrome—United States, 1969-2004. MMWR 2005 $4:279-82 Hlavsa MC, Watson Jt Giardiasis surveillance—United Modifications to Table | and Table Il States, 1998-2002. In: Surveillance summaries January sarv 2288, 2004 [wo new conditions have been added to the list of nationally MMWR 2004;53(No. SS notifiable infectious diseases: nonparalytic poliovirus infection and vibriosis (non-cholera Vibrio species infections). Incidence data Notice to Readers for both of these conditions will appear in Table |. The surveil- Satellite Broadcast: Epidemiology ince case definitions adopted for these conditairoe nlisste d within and Prevention of Vaccine-Preventable their respective CSTE position statements (/,2) and are posted Diseases 2007 o the case definitions section of the National Notifiable Diseases Surveillance System (NNDSS) website (3 CDC and the Public Health Training Network will present he CSTE position statement Enhancing local, state, and the satellite broadcast webcast series, Epide miology and Dd . ry rritorial-based surveillance for invasive pneumococcal dis Prevention of Vaccine entable Diseases. This series is sched- of age 4), includes uled for four consecutive Thursdays from 12 noon to 4 p.m., case definitions for on January 25 and February 1, 8, and 15, 2007. dé invasive disease Session | includes an overview of general immunization con NeuUINONI de cepts and principles and vaccine safety, storage and handling, event code 1171 and administration. Session 2 topics include pertussis, pneu- isional data tables published in mococcal disease (childhood), polio, rotavirus, and Haemophilus 1 in Table II in two col influenzae type b. Session 3 topics include measles, rubella, nd a second column varicella, zoster, and meningococcal disease. Session 4 topics include hepatitis B, hepatitis A, influenza, human papillo mavirus, and pneumococcal disease (adult). A live question and-answet session will be conducted Via toll-tree telephone ifications to Figure | lines. Continuing Education (CE) credits will be provided. bella has been deleted from Figure I and replaced with \dditional information regarding the series is availablea t http: because of low incidence www2.cdc.gov/phtn/epv07/default.asp. Information for site yn as no longer endemic in the United States administrators regarding establishing and registering a viewing istrointestinal illness, is caused by the pro location is available at http://www.cdc.gov/phtnonline. This ia intestinalis. This pathogen has a low website also is appropriate for individual participants who wish tious dose, protracted communicability, and moderate to register to view the broadcast from a specific location, or who nce to chlorine, which makes it ideally suited for trans seek CE credit. No registration is necessary tO access the W ebcasts SIO! through drinking and recreational water, food and both via an Internet connection. The link to the live webcasts is avail erson and animal-to-person contact. Transmission able at http: IwWWww 2.cdc.gov phtn/webcast/epv07/default.asp. giardiasis occurs throughout the United States with a marked lhe webcasts will be accessible through an Internet connection sonality peaking in summet through early fall until March 15, 2007. The program will become available as a self-study DVD and Internet-based program in March 2007.