Risk Factors for Community-Associated Clostridium difficile Infection in Children

      Objective

      To characterize the medication and other exposures associated with pediatric community-associated Clostridium difficile infections (CA-CDIs).

      Study design

      We performed a case-control study using billing records from the US military health system database. CA-CDI cases included children 1-18 years of age with an outpatient International Classification of Diseases, Ninth Revision, Clinical Modification diagnostic code for Clostridium difficile infection (CDI) from 2001 to 2013. Each case was matched to 3 controls without CDI by age and sex. Children hospitalized at any time before their CDI were excluded. Outpatient pharmacy records were used to identify medication exposures in the preceding 12 weeks. In addition, we evaluated recent outpatient healthcare exposure, exposure to a sibling younger than 1 year of age, or to a family member with CDI.

      Results

      A total of 1331 children with CA-CDI were identified and 3993 controls were matched successfully. Recent exposure to fluoroquinolones, clindamycin (OR 73.00; 95% CI 13.85-384.68), third-generation cephalosporins (OR 16.32; 95% CI 9.11-29.26), proton pump inhibitors (OR 8.17; 95% CI 2.35-28.38), and to multiple classes of antibiotics, each was associated strongly the subsequent diagnosis of CA-CDI. Recent exposure to outpatient healthcare clinics (OR 1.35; 95% CI 1.31-1.39) or to a family member with CDI also was associated with CA-CDI.

      Conclusions

      CA-CDI is associated with medications regularly prescribed in pediatric practice, along with exposure to outpatient healthcare clinics and family members with CDI. Our findings provide additional support for the judicious use of these medications and for efforts to limit spread of CDI in ambulatory healthcare settings and households.

      Keywords

      Abbreviations:

      CA-CDI (Community-associated Clostridium difficile infection), CDI (Clostridium difficile infection), ICD-9-CM (International Classification of Diseases, Ninth Revision, Clinical Modification), MHS (Military health system)
      Clostridium difficile is a spore-forming, toxin-producing, anaerobic bacillus responsible for a variety of gastrointestinal manifestations ranging from asymptomatic carriage to mild diarrhea, pseudomembranous colitis, and, very rarely in children, toxic megacolon, bowel perforation, and death. Clostridium difficile infection (CDI) is increasing among hospitalized children,
      • Nylund C.M.
      • Goudie A.
      • Garza J.M.
      • Fairbrother G.
      • Cohen M.B.
      Clostridium difficile infection in hospitalized children in the United States.
      leading to increased mortality, longer length of stay, and greater hospitalization costs.
      • Sammons J.S.
      • Localio R.
      • Xiao R.
      • Coffin S.E.
      • Zaoutis T.
      Clostridium difficile infection is associated with increased risk of death and prolonged hospitalization in children.
      In recent years, however, the epidemiology of this infection has shifted as CDI cases in both adults and children increasingly have been originating in the community.
      • Baker S.S.
      • Faden H.
      • Sayej W.
      • Patel R.
      • Baker R.D.
      Increasing incidence of community-associated atypical Clostridium difficile disease in children.
      • Benson L.
      • Song X.
      • Campos J.
      • Singh N.
      Changing epidemiology of Clostridium difficile-associated disease in children.
      These community-associated Clostridium difficile infections (CA-CDIs) now account for nearly one-third of all C difficile cases.
      • Chitnis A.S.
      • Holzbauer S.M.
      • Belflower R.M.
      • Winston L.G.
      • Bamberg W.M.
      • Lyons C.
      • et al.
      Epidemiology of community-associated Clostridium difficile infection, 2009 through 2011.
      Although previous antibiotic exposure is a well-established risk factor for the development of CDI, there are limited data on which antibiotic class exposures precede CDI in children and few studies evaluating these exposures in CA-CDI.
      • Crews J.D.
      • Anderson L.R.
      • Waller D.K.
      • Swartz M.D.
      • DuPont H.L.
      • Starke J.R.
      Risk factors for community-associated Clostridium difficile associated diarrhea in children.
      In addition, studies of CA-CDI in both adult and pediatric populations have identified a large subset of patients without preceding antibiotic exposures.
      • Benson L.
      • Song X.
      • Campos J.
      • Singh N.
      Changing epidemiology of Clostridium difficile-associated disease in children.
      • Chitnis A.S.
      • Holzbauer S.M.
      • Belflower R.M.
      • Winston L.G.
      • Bamberg W.M.
      • Lyons C.
      • et al.
      Epidemiology of community-associated Clostridium difficile infection, 2009 through 2011.
      These findings highlight the need to identify additional exposures contributing to this increase in community transmission of CDI. Two recent small studies of CA-CDI in children identified use of gastric acid-suppression therapy
      • Jimenez J.
      • Drees M.
      • Loveridge-Lenza B.
      • Eppes S.
      • del Rosario F.
      Exposure to gastric acid-suppression therapy is associated with health care- and community-associated Clostridium difficile infection in children.
      and the presence of a gastrointestinal feeding device
      • Crews J.D.
      • Anderson L.R.
      • Waller D.K.
      • Swartz M.D.
      • DuPont H.L.
      • Starke J.R.
      Risk factors for community-associated Clostridium difficile associated diarrhea in children.
      as additional risk factors for CA-CDI; however, the epidemiologic exposures and underlying risk factors leading to CA-CDI in children remain largely unknown. Using the large US military health system (MHS) database, we sought to characterize medication and other potential exposures associated with CA-CDI in the pediatric population.

      Methods

      We performed a case-control study using billing records from the TRICARE Management Activity MHS database, which includes all eligible military dependents cared for both in military and civilian facilities. Cases were selected from among children ages 1-18 years old who received care during the time period spanning October 1, 2001, to September 30, 2013, with an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic code 008.45. This is the only code representing CDI and has been validated previously in pediatric billing records.
      • Shaklee J.
      • Zerr D.M.
      • Elward A.
      • Newland J.
      • Leckerman K.
      • Asti L.
      • et al.
      Improving surveillance for pediatric Clostridium difficile infection: derivation and validation of an accurate case-finding tool.
      Children younger than 1 year old were excluded as C difficile frequently colonizes the intestine in healthy infants.
      • Al-Jumaili I.J.
      • Shibley M.
      • Lishman A.H.
      • Record C.O.
      Incidence and origin of Clostridium difficile in neonates.
      • Bryant K.
      • McDonald L.C.
      Clostridium difficile infections in children.
      • Viscidi R.
      • Willey S.
      • Bartlett J.G.
      Isolation rates and toxigenic potential of Clostridium difficile isolates from various patient populations.
      In addition, to evaluate only CA-CDI, children hospitalized at any time before their CDI were excluded from the analysis. Cases were then matched by age (date of birth) and sex with 3 controls without an ICD-9-CM code for CDI. We used outpatient pharmacy records for the selected cases and controls to identify medication exposures in the 12 weeks preceding the date of the first CDI among cases, including 10 classes of oral antibiotics (clindamycin, fluoroquinolones, sulfonamides, macrolides, penicillins, amoxicillin/clavulanate, tetracyclines, and first-, second-, and third-generation cephalosporins), 2 gastric acid-suppression medications (proton pump inhibitors, H2 receptor antagonists), and corticosteroids. Dosing or patient weight information was not available. All medication exposures were categorized as binary variables. In addition, we calculated the odds of CA-CDI with exposure to multiple classes of antibiotics. During the same 12-week time period, we also evaluated exposures to outpatient healthcare settings, siblings younger than 1 year of age, or to family members with a diagnosis of CDI. Each TRICARE dependent (either child or spouse) is tied to their sponsor in the MHS database by a unique identifier, which was used to identify and evaluate family member CDI exposures.
      Univariate and multivariable conditional logistic regression were performed to calculate unadjusted and adjusted OR and 95% CIs. The dependent variable was CDI, and the independent variables were various medication exposures, family exposures, and the number of outpatient healthcare encounters. In addition, 2-way interactions between all independent variables were evaluated and included in the final multivariable model if significant. The Cochran-Armitage test for trend was used to analyze the trend in CA-CDI during the 13-year study period. P values less than .05 were considered statistically significant. All analyses were performed with SAS 9.3 (SAS Institute, Cary, North Carolina). The study was approved by our institutional review board.

      Results

      A total of 1331 children with CA-CDI and 3993 controls were identified and matched successfully by sex and exact date of birth from the MHS database during the study period (Table I). The median (IQR) age in years of CA-CDI cases and controls was 7.0 (3.3-13.4). Cases were divided evenly between male (50.3%) and female (49.7%) children. A total of 163 (12.2%) of children with CA-CDI were hospitalized on the same day as their outpatient encounter for CDI, and an additional 55 (4.1%) were hospitalized within the following week. The majority (59.7%) of the 1331 children with CA-CDI was prescribed at least 1 antibiotic in the 12 weeks preceding their diagnosis; however, 40.3% had no preceding antibiotic exposure. Of the 795 children with an identified antibiotic exposure preceding their CA-CDI, 319 (40.1%) were prescribed multiple classes of antibiotics, including 252 (31.6%) with exposure to 2, and 68 (8.6%) with exposure to 3 or more different antibiotic classes. The median (IQR) time to CA-CDI after an antibiotic prescription was 33 days (17-54). Pediatric CA-CDI cases demonstrated a significantly increasing trend during the 12-year study period with an average annual increase of 47.9% (P < .001) (Figure 1).
      Table IExposures preceding community-associated C difficile infection in pediatric cases compared with controls
      ExposuresCases (n = 1331),

      n (%)
      Controls (n = 3993),

      n (%)
      Family member with CDI8(0.60)0(0)
      Fluoroquinolones51(3.83)0(0)
      Clindamycin103(7.74)8(0.20)
      Third-generation cephalosporins223(16.75)50(1.25)
      Proton pump inhibitors143(10.74)19(0.48)
      Second-generation cephalosporins41(3.08)20(0.50)
      H2 receptor antagonist69(5.18)13(0.33)
      Sulfonamides167(12.55)39(0.98)
      First-generation cephalosporins62(4.66)37(0.93)
      Amoxicillin/clavulanate194(14.58)88(2.20)
      Macrolides131(9.84)161(4.03)
      Penicillins206(15.48)282(7.06)
      Corticosteroids201(15.10)133(3.33)
      Tetracyclines17(1.28)30(0.75)
      Sibling younger than 1 y of age98(7.36)343(8.59)
      Figure 1
      Figure 1Trend in pediatric community-associated C difficile infections.
      The antibiotic class exposures associated most strongly with CA-CDI included fluoroquinolones (OR could not be calculated as 51 cases were exposed compared with 0 controls) (Table I), clindamycin (OR 73.00; 95% CI 13.85-384.68), and third-generation cephalosporins (OR 16.32; 95% CI 9.11-29.26) (Table II). In addition, children with recent exposure to multiple classes of antibiotics carried increased odds of developing CA-CDI, compared with those exposed to only 1 class of antibiotic (Figure 2). The odds of CA-CDI following exposure to proton pump inhibitors (OR 8.17; 95% CI 2.35-28.38) was comparable with that of antibiotic class exposures. Outpatient healthcare clinic visits also were associated with CA-CDI in children, with a 35% increase in the odds of CDI of for each additional visit (OR 1.35; 95% CI 1.31-1.39). Regarding household C difficile exposures, recent contact with a family member with CDI was associated with pediatric CA-CDI (8 cases were exposed compared with 0 controls) (Table I); however, recent exposure to a sibling younger than 1 year of age was not associated with CA-CDI in children (OR 0.83; 95% CI 0.57-1.22).
      Table IIOdds of community-associated C difficile infection in children by exposure
      * ORs could not be calculated for fluoroquinolones or exposure to a family member with CDI.
      ,
      Significant 2-way interactions were included in multivariable model.
      ExposuresUnadjusted OR (95% CI)Adjusted OR (95% CI)
      Clindamycin43.86(20.40-94.32)73.00(13.85-384.68)
      Third-generation cephalosporins16.62(11.82-23.36)16.32(9.11-29.26)
      Proton pump inhibitors29.92(17.28-51.82)8.17(2.35-28.38)
      H2 receptor antagonist17.08(9.25-31.53)3.33(1.18-9.40)
      Second-generation cephalosporins6.38(3.70-11.00)2.96(1.25-7.01)
      Sulfonamides14.46(10.00-20.92)2.53(1.33-4.79)
      First-generation cephalosporins5.43(3.56-8.30)2.28(1.13-4.60)
      Macrolides2.62(2.06-3.34)2.01(1.04-3.86)
      Amoxicillin/clavulanate7.54(5.76-9.87)1.50(0.66-3.43)
      Penicillins2.48(2.04-3.02)1.44(1.04-2.00)
      Outpatient visits
      Represents odds of C difficile infection for each additional outpatient healthcare visit.
      1.44(1.40-1.48)1.35(1.31-1.39)
      Corticosteroids5.12(4.05-6.45)1.22(0.79-1.90)
      Tetracyclines1.72(0.94-3.13)1.26(0.46-3.44)
      Sibling younger than 1 y of age0.84(0.66-1.07)0.83(0.57-1.22)
      * ORs could not be calculated for fluoroquinolones or exposure to a family member with CDI.
      Significant 2-way interactions were included in multivariable model.
      Represents odds of C difficile infection for each additional outpatient healthcare visit.
      Figure 2
      Figure 2Odds of community-associated C difficile infection in children by the number of preceding antibiotic class exposures.

      Discussion

      Our study supports the occurrence of CDI among a population of children who were never hospitalized previously and provides a broad characterization of the medication and epidemiologic exposures associated with pediatric CA-CDI cases. Recent exposure to fluoroquinolones, clindamycin, third-generation cephalosporins, and to multiple classes of antibiotics was associated strongly with the subsequent diagnosis of CA-CDI in children; however, a sizeable minority had no preceding antibiotic exposure. We identified exposure to proton pump inhibitors, outpatient healthcare clinics, and to family members with CDI as additional pediatric CA-CDI risk factors. Although not the primary aim of our study, we identified a significantly increasing trend in CA-CDI cases in children during the 12-year study period, consistent with that previously described.
      • Benson L.
      • Song X.
      • Campos J.
      • Singh N.
      Changing epidemiology of Clostridium difficile-associated disease in children.
      • Pant C.
      • Deshpande A.
      • Altaf M.A.
      • Minocha A.
      • Sferra T.J.
      Clostridium difficile infection in children: a comprehensive review.
      • Tschudin-Sutter S.
      • Tamma P.D.
      • Naegeli A.N.
      • Speck K.A.
      • Milstone A.M.
      • Perl T.M.
      Distinguishing community-associated from hospital-associated Clostridium difficile infections in children: implications for public health surveillance.
      This increase in CA-CDI among children is alarming, as Tschudin-Sutter et al
      • Tschudin-Sutter S.
      • Tamma P.D.
      • Naegeli A.N.
      • Speck K.A.
      • Milstone A.M.
      • Perl T.M.
      Distinguishing community-associated from hospital-associated Clostridium difficile infections in children: implications for public health surveillance.
      • Tschudin-Sutter S.
      • Tamma P.D.
      • Milstone A.M.
      • Perl T.M.
      Predictors of first recurrence of Clostridium difficile infections in children.
      have reported an association between CA-CDI and both recurrent CDI and severe complications of CDI such as toxic megacolon.
      Antibiotic exposure is the best-established risk factor for the development of CDI and was present in the majority of children with CA-CDI in our study. Nearly every class of antibiotic has been linked to the development of CDI; however, certain classes, especially those with a broader spectrum of action, including potent activity against enteric gram-negative and anaerobic bacteria, have been associated with a higher risk for this infection.
      • Brown K.A.
      • Khanafer N.
      • Daneman N.
      • Fisman D.N.
      Meta-analysis of antibiotics and the risk of community-associated Clostridium difficile infection.
      • Deshpande A.
      • Pasupuleti V.
      • Thota P.
      • Pant C.
      • Rolston D.D.
      • Sferra T.J.
      • et al.
      Community-associated Clostridium difficile infection and antibiotics: a meta-analysis.
      Few studies have evaluated which antibiotics pose the greatest risk for CDI in children. In a large retrospective cohort of children with a newly diagnosed malignancy, recent exposure to classes of antibiotics used commonly in patients with fever and neutropenia were evaluated as potential risk factors for the development of HA-CDI. The authors noted a significantly increased risk of HA-CDI among those patients who were treated with aminoglycosides, third-generation cephalosporins, and cefepime in the week preceding their CDI.
      • de Blank P.
      • Zaoutis T.
      • Fisher B.
      • Troxel A.
      • Kim J.
      • Aplenc R.
      Trends in Clostridium difficile infection and risk factors for hospital acquisition of Clostridium difficile among children with cancer.
      In a retrospective nested case-control study of children with HA-CDI and CA-CDI, receipt of fluoroquinolones or nonquinolone antibiotics in the preceding 4 weeks were both predictors of CDI.
      • Sandora T.J.
      • Fung M.
      • Flaherty K.
      • Helsing L.
      • Scanlon P.
      • Potter-Bynoe G.
      • et al.
      Epidemiology and risk factors for Clostridium difficile infection in children.
      Our study is the most complete description of antibiotic exposures in children with CDI and specifically characterizes those preceding infections acquired in the community, among presumably healthy, never-hospitalized children. Two recent meta-analyses studying the association between antibiotic class exposures and CA-CDI in adults closely matched the results of our study in children, identifying clindamycin, fluoroquinolones, and cephalosporins as having the greatest risk for CDI. Exposure to macrolides, sulfonamides, and penicillins also was associated with CDI in these studies in adults, but exposure carried a much lower risk.
      • Brown K.A.
      • Khanafer N.
      • Daneman N.
      • Fisman D.N.
      Meta-analysis of antibiotics and the risk of community-associated Clostridium difficile infection.
      • Deshpande A.
      • Pasupuleti V.
      • Thota P.
      • Pant C.
      • Rolston D.D.
      • Sferra T.J.
      • et al.
      Community-associated Clostridium difficile infection and antibiotics: a meta-analysis.
      In addition, we identified a significant number of children with CA-CDI who were exposed to multiple classes of antibiotics in the 12 weeks preceding their infection and demonstrated an increase in the odds of CA-CDI with each preceding antibiotic class exposure. Combination antibiotic therapies may have constituted a broader spectrum of antimicrobial activity, leading to a greater impact on host microbial diversity, or if experienced in series may have impaired recovery of healthy microbiota before the next antibiotic exposure. Alternatively, those exposed to multiple antibiotic classes may represent children with more severe underlying disease, who therefore are more vulnerable to CDI.
      More than 40% of the pediatric CA-CDI cases in our study had no preceding antibiotic exposure. This finding is consistent with a retrospective cohort of children with CDI who were evaluated in a hospital emergency department, of whom 43% had no recent documented antibiotic exposure.
      • Benson L.
      • Song X.
      • Campos J.
      • Singh N.
      Changing epidemiology of Clostridium difficile-associated disease in children.
      Our results shed light on several nonantibiotic exposures that may have contributed to CA-CDI in these children. There is mounting evidence that maintenance of a physiologic gastric acid barrier may be important in preventing CDI, although acid-suppression medications may increase the risk of infections through other mechanisms.
      • Stark C.M.
      • Nylund C.
      Side effects and complications of proton pump inhibitors: a pediatric perspective.
      Several studies of CDI in children support our findings that recent exposure to gastric acid-suppression medications is associated with CA-CDI.
      • Jimenez J.
      • Drees M.
      • Loveridge-Lenza B.
      • Eppes S.
      • del Rosario F.
      Exposure to gastric acid-suppression therapy is associated with health care- and community-associated Clostridium difficile infection in children.
      • Nylund C.M.
      • Eide M.
      • Gorman G.H.
      Association of Clostridium difficile infections with acid suppression medications in children.
      • Turco R.
      • Martinelli M.
      • Miele E.
      • Roscetto E.
      • Del Pezzo M.
      • Greco L.
      • et al.
      Proton pump inhibitors as a risk factor for paediatric Clostridium difficile infection.
      We found that the odds of CA-CDI development following proton pump inhibitor and H2 receptor antagonist exposure were similar to that seen with antibiotic exposures. Given the high frequency with which these medications are prescribed in pediatrics for conditions such as physiologic reflux, these exposures may be a contributor to the increasing number of CA-CDI cases seen in children.
      • Hassall E.
      Over-prescription of acid-suppressing medications in infants: how it came about, why it's wrong, and what to do about it.
      • Faden H.S.
      • Ma C.X.
      Trends in oral antibiotic, proton pump inhibitor, and histamine 2 receptor blocker prescription patterns for children compared with adults: implications for Clostridium difficile infection in the community.
      Although C difficile remains a common hospital-associated infection, with great emphasis placed on the importance of infection control measures used to prevent spread in the inpatient setting, our results indicate that many CA-CDI cases may be the result of transmission in the outpatient healthcare setting. Each additional outpatient clinic visit increased the odds by 35% of children in our study developing CA-CDI. C difficile spores have been shown to survive for months to years in the environment and have been recovered from multiple surfaces in healthcare settings, including commodes, blood pressure cuffs, and both rectal and oral thermometers. Healthcare workers' hands, however, seem to be the most important vehicle in person-to-person spread.
      • Cohen S.H.
      • Gerding D.N.
      • Johnson S.
      • Kelly C.P.
      • Loo V.G.
      • McDonald L.C.
      • et al.
      Clinical practice guidelines for Clostridium difficile infection in adults: 2010 update by the society for healthcare epidemiology of America (SHEA) and the infectious diseases society of America (IDSA).
      It is therefore reasonable to conclude that some C difficile transmission is occurring in outpatient healthcare clinics, which traditionally were thought to be lower risk settings for the acquisition of C difficile. The CDC “Guide to Infection Prevention for Outpatient Settings” stresses the importance of using soap and water for handwashing when caring for patients with CDI and gives generic guidance on environmental cleaning; however the prevention of C difficile transmission in outpatient healthcare settings remains an important area for future research.
      • Centers for Disease Control and Prevention
      Guide to infection prevention for outpatient settings: minimum expectations for safe care [page on the internet].
      It is possible, however, that increased outpatient visits among children in our study reflect an increased underlying disease severity, placing them at increased risk for acquiring CDI. Further studies to evaluate and quantify the risk of CA-CDI in children from outpatient healthcare exposure are needed.
      Recent contact with an immediate family member diagnosed with CDI was associated strongly with CA-CDI in our study but affected only a few children. Because we were only able to account for immediate family members with records in the MHS database, additional exposure to other family members and household contacts with CDI may have been missed, thereby underestimating the true impact of household CDI exposure. To help quantify the risk of CDI among household contacts of infected persons, Pepin et al
      • Pepin J.
      • Gonzales M.
      • Valiquette L.
      Risk of secondary cases of Clostridium difficile infection among household contacts of index cases.
      studied CDI attack rates among 1562 household contacts of 2222 index cases in a region of Quebec, Canada. They discovered that in the 3 months after CDI diagnosis in the index case, the attack rate for spouses and children was only 4.7 of 1000 and 5.9 of 1000, respectively. They concluded that interventions aimed at curbing household spread of C difficile were therefore not warranted, given the relatively low absolute risk of transmission. Others have argued, however, that household transmission of CDI should not be ignored and advocate for continued use of contact precautions for household contacts of infected patients while undergoing treatment at home.
      • Girotra M.
      • Abraham R.R.
      • Pahwa M.
      Clostridium difficile infection: how safe are the household contacts?.
      Formal guidance in this area is limited, with the US Centers for Disease Control and Prevention factsheet for CDI only mentioning the importance of hand hygiene for patients infected with C difficile and their household contacts after returning home from the hospital.
      • Centers for Disease Control and Prevention
      Clostridium difficile infection information for patients.
      As the epidemiology of CDI has shifted toward community acquisition, there has been increased recognition of asymptomatic C difficile carriage among both children and adults.
      • Alasmari F.
      • Seiler S.M.
      • Hink T.
      • Burnham C.A.
      • Dubberke E.R.
      Prevalence and risk factors for asymptomatic Clostridium difficile carriage.
      • Leibowitz J.
      • Soma V.L.
      • Rosen L.
      • Ginocchio C.C.
      • Rubin L.G.
      Similar proportions of stool specimens from hospitalized children with and without diarrhea test positive for Clostridium difficile.
      In a study of adults with CA-CDI, those with minimal or no outpatient healthcare contact were more likely than those with high level outpatient healthcare contact to have been exposed to infants younger than 1 year old and household contacts with CDI.
      • Chitnis A.S.
      • Holzbauer S.M.
      • Belflower R.M.
      • Winston L.G.
      • Bamberg W.M.
      • Lyons C.
      • et al.
      Epidemiology of community-associated Clostridium difficile infection, 2009 through 2011.
      In our analysis of pediatric CA-CDI cases, exposure to an infant younger than 1 year of age in the home was not associated with CDI; however, as mentioned previously, we were only able to capture infant exposure within the immediate family unit and cannot account for other exposure to infants in daycare settings or elsewhere outside the home. Regarding the risk of transmission from asymptomatic C difficile carriers, Durham et al
      • Durham D.P.
      • Olsen M.A.
      • Dubberke E.R.
      • Galvani A.P.
      • Townsend J.P.
      Quantifying transmission of Clostridium difficile within and outside healthcare settings.
      used a sophisticated model to demonstrate that hospitalized C difficile-infected patients transmit their infection 15 times more frequently than asymptomatic patients. In addition, they show that community transmission of CDI occurs at only 0.1% the rate of hospital transmission. Despite these low transmission rates, however, the authors rightly emphasize the significant contribution of asymptomatic C difficile carriage and community transmission toward CDI cases as a whole, given the substantial number of these C difficile sources present in the community. Further research on household CDI transmission is necessary to provide appropriate guidance to family member and close contacts of patients with CDI.
      Ours is one of the largest studies of CA-CDI in children and includes a population that is both geographically and demographically diverse, treated at both military and civilian facilities, with universal access to care. This allowed us to accurately exclude children with previous hospitalizations and comprehensively identify antibiotic prescriptions in the period leading up to their CDI. Our study was limited by the use of an ICD-9-CM code for case identification and therefore is vulnerable to misclassification bias; however, this code has been previously validated as a measure of CDI. Although the children in our study were never hospitalized before their CDI, we could not account for any hospital visits they made to visit friends or family members or hospital exposure apart from inpatient admission. In addition, as our measurement of antibiotic exposure was based on antibiotic prescriptions dispensed from pharmacies, and not medication compliance, the contribution of antibiotics to CA-CDI could be overestimated.
      Improved stewardship of antibiotics is clearly important, but decreasing antibiotic exposure alone will not eliminate CA-CDI cases in children. Our findings also support judicious use of acid-suppression medications and highlight the need for efforts to better quantify C difficile transmission and limit its spread both in ambulatory healthcare settings and in households.

      References

        • Nylund C.M.
        • Goudie A.
        • Garza J.M.
        • Fairbrother G.
        • Cohen M.B.
        Clostridium difficile infection in hospitalized children in the United States.
        Arch Pediatr Adolesc Med. 2011; 165: 451-457
        • Sammons J.S.
        • Localio R.
        • Xiao R.
        • Coffin S.E.
        • Zaoutis T.
        Clostridium difficile infection is associated with increased risk of death and prolonged hospitalization in children.
        Clin Infect Dis. 2013; 57: 1-8
        • Baker S.S.
        • Faden H.
        • Sayej W.
        • Patel R.
        • Baker R.D.
        Increasing incidence of community-associated atypical Clostridium difficile disease in children.
        Clin Pediatr (Phila). 2010; 49: 644-647
        • Benson L.
        • Song X.
        • Campos J.
        • Singh N.
        Changing epidemiology of Clostridium difficile-associated disease in children.
        Infect Control Hosp Epidemiol. 2007; 28: 1233-1235
        • Chitnis A.S.
        • Holzbauer S.M.
        • Belflower R.M.
        • Winston L.G.
        • Bamberg W.M.
        • Lyons C.
        • et al.
        Epidemiology of community-associated Clostridium difficile infection, 2009 through 2011.
        JAMA Intern Med. 2013; 173: 1359-1367
        • Crews J.D.
        • Anderson L.R.
        • Waller D.K.
        • Swartz M.D.
        • DuPont H.L.
        • Starke J.R.
        Risk factors for community-associated Clostridium difficile associated diarrhea in children.
        Pediatr Infect Dis J. 2015; 34: 919-923
        • Jimenez J.
        • Drees M.
        • Loveridge-Lenza B.
        • Eppes S.
        • del Rosario F.
        Exposure to gastric acid-suppression therapy is associated with health care- and community-associated Clostridium difficile infection in children.
        J Pediatr Gastroenterol Nutr. 2015; 61: 208-211
        • Shaklee J.
        • Zerr D.M.
        • Elward A.
        • Newland J.
        • Leckerman K.
        • Asti L.
        • et al.
        Improving surveillance for pediatric Clostridium difficile infection: derivation and validation of an accurate case-finding tool.
        Pediatr Infect Dis J. 2011; 30: e38-40
        • Al-Jumaili I.J.
        • Shibley M.
        • Lishman A.H.
        • Record C.O.
        Incidence and origin of Clostridium difficile in neonates.
        J Clin Microbiol. 1984; 19: 77-78
        • Bryant K.
        • McDonald L.C.
        Clostridium difficile infections in children.
        Pediatr Infect Dis J. 2009; 28: 145-146
        • Viscidi R.
        • Willey S.
        • Bartlett J.G.
        Isolation rates and toxigenic potential of Clostridium difficile isolates from various patient populations.
        Gastroenterology. 1981; 81: 5-9
        • Pant C.
        • Deshpande A.
        • Altaf M.A.
        • Minocha A.
        • Sferra T.J.
        Clostridium difficile infection in children: a comprehensive review.
        Curr Med Res Opin. 2013; 29: 967-984
        • Tschudin-Sutter S.
        • Tamma P.D.
        • Naegeli A.N.
        • Speck K.A.
        • Milstone A.M.
        • Perl T.M.
        Distinguishing community-associated from hospital-associated Clostridium difficile infections in children: implications for public health surveillance.
        Clin Infect Dis. 2013; 57: 1665-1672
        • Tschudin-Sutter S.
        • Tamma P.D.
        • Milstone A.M.
        • Perl T.M.
        Predictors of first recurrence of Clostridium difficile infections in children.
        Pediatr Infect Dis J. 2014; 33: 414-416
        • Brown K.A.
        • Khanafer N.
        • Daneman N.
        • Fisman D.N.
        Meta-analysis of antibiotics and the risk of community-associated Clostridium difficile infection.
        Antimicrob Agents Chemother. 2013; 57: 2326-2332
        • Deshpande A.
        • Pasupuleti V.
        • Thota P.
        • Pant C.
        • Rolston D.D.
        • Sferra T.J.
        • et al.
        Community-associated Clostridium difficile infection and antibiotics: a meta-analysis.
        J Antimicrob Chemother. 2013; 68: 1951-1961
        • de Blank P.
        • Zaoutis T.
        • Fisher B.
        • Troxel A.
        • Kim J.
        • Aplenc R.
        Trends in Clostridium difficile infection and risk factors for hospital acquisition of Clostridium difficile among children with cancer.
        J Pediatr. 2013; 163 (e1): 699-705
        • Sandora T.J.
        • Fung M.
        • Flaherty K.
        • Helsing L.
        • Scanlon P.
        • Potter-Bynoe G.
        • et al.
        Epidemiology and risk factors for Clostridium difficile infection in children.
        Pediatr Infect Dis J. 2011; 30: 580-584
        • Stark C.M.
        • Nylund C.
        Side effects and complications of proton pump inhibitors: a pediatric perspective.
        J Pediatr. 2016; 168: 16-22
        • Nylund C.M.
        • Eide M.
        • Gorman G.H.
        Association of Clostridium difficile infections with acid suppression medications in children.
        J Pediatr. 2014; 165 (e1): 979-984
        • Turco R.
        • Martinelli M.
        • Miele E.
        • Roscetto E.
        • Del Pezzo M.
        • Greco L.
        • et al.
        Proton pump inhibitors as a risk factor for paediatric Clostridium difficile infection.
        Aliment Pharmacol Ther. 2010; 31: 754-759
        • Hassall E.
        Over-prescription of acid-suppressing medications in infants: how it came about, why it's wrong, and what to do about it.
        J Pediatr. 2012; 160: 193-198
        • Faden H.S.
        • Ma C.X.
        Trends in oral antibiotic, proton pump inhibitor, and histamine 2 receptor blocker prescription patterns for children compared with adults: implications for Clostridium difficile infection in the community.
        Clin Pediatr (Phila). 2016; 55: 712-716
        • Cohen S.H.
        • Gerding D.N.
        • Johnson S.
        • Kelly C.P.
        • Loo V.G.
        • McDonald L.C.
        • et al.
        Clinical practice guidelines for Clostridium difficile infection in adults: 2010 update by the society for healthcare epidemiology of America (SHEA) and the infectious diseases society of America (IDSA).
        Infect Control Hosp Epidemiol. 2010; 31: 431-455
        • Centers for Disease Control and Prevention
        Guide to infection prevention for outpatient settings: minimum expectations for safe care [page on the internet].
        Centers for Disease Control and Prevention, Atlanta (GA)2016 (Updated 2016, Sept) (Accessed February 14, 2017)
        • Pepin J.
        • Gonzales M.
        • Valiquette L.
        Risk of secondary cases of Clostridium difficile infection among household contacts of index cases.
        J Infect. 2012; 64: 387-390
        • Girotra M.
        • Abraham R.R.
        • Pahwa M.
        Clostridium difficile infection: how safe are the household contacts?.
        Am J Infect Control. 2013; 41: 1140-1141
        • Centers for Disease Control and Prevention
        Clostridium difficile infection information for patients.
        ([page on the internet]. Atlanta (GA): Centers for Disease Control and Prevention. Updated 2015, Feb) (Accessed February 14, 2017)
        • Alasmari F.
        • Seiler S.M.
        • Hink T.
        • Burnham C.A.
        • Dubberke E.R.
        Prevalence and risk factors for asymptomatic Clostridium difficile carriage.
        Clin Infect Dis. 2014; 59: 216-222
        • Leibowitz J.
        • Soma V.L.
        • Rosen L.
        • Ginocchio C.C.
        • Rubin L.G.
        Similar proportions of stool specimens from hospitalized children with and without diarrhea test positive for Clostridium difficile.
        Pediatr Infect Dis J. 2014; 34: 261-266
        • Durham D.P.
        • Olsen M.A.
        • Dubberke E.R.
        • Galvani A.P.
        • Townsend J.P.
        Quantifying transmission of Clostridium difficile within and outside healthcare settings.
        Emerg Infect Dis. 2016; 22: 608-616