LETTERS TO THE EDITOR
Statements appearing here are those of the writers and do not represent the official position of the American Academy of Pediatrics or its Committees. Comments on any topic, including the contents of PEDIATRICS, are invited from all members of the profession; those accepted for publication will not be subject to major editorial revision but generally must be no more than 400 words in length. The editors reserve the right to publish replies and may solicit responses from authors and others.
Please see www.pediatrics.org for instructions on submitting letters.
Emergency Contraception
To the Editor.—
The American Academy of Pediatrics Committee on Ad-olescence policy statement on emergency contraception1 reports the effectiveness of the Yuzpe regimen (ethinyl estradiol and levonorgestrel) in terms of a pregnancy reduction of 70% to 80% and of levonorgestrel-only emergency contraception of 85%. These estimates are outdated. Using current methods for estimating effec-tiveness, the effectiveness rates seem to be in the range of 50% to 66% and 72% to 80%, respectively.2–5 Be-cause there are no randomized trials with a placebo arm, considerable uncertainty remains about the effectiveness of emergency contraception.3,5
The policy statement also proposes that “[e]mergency contraception has tremendous potential to reduce unin-tended pregnancy rates in teens and adults.” This state-ment remains, as yet, a hypothesis that is unsupported by empirical evidence. Several studies have failed to document a decrease in rates of unintended pregnancy or abortion in populations that are provided with ad-vance access to emergency contraception.6–8 This sug-gests that the studies that have demonstrated no changes in sexual behavior with advance access (other than creased use of emergency contraception) have used in-adequate surrogate end points or have failed to detect small changes in sexual behavior that were nevertheless sufficient to negate any decrease in unintended preg-nancy.
Joseph B. Stanford, MD, MSPH
Department of Family and Preventive Medicine University of Utah Salt Lake City, UT 84108
Rafael T. Mikolajczyk, MD
Department of Public Health Medicine School of Public Health University of Bielefeld D-33501 Bielefeld, Germany
REFERENCES
1. American Academy of Pediatrics, Committee on Adolescence. Emergency contraception.Pediatrics.2005;116:1026 –1035 2. Trussell J, Ellertson C, von Hertzen H, et al. Estimating the
effectiveness of emergency contraceptive pills. Contraception.
2003;67:259 –265
3. Raymond E, Taylor D, Trussell J, Steiner MJ. Minimum effec-tiveness of the levonorgestrel regimen of emergency contracep-tion.Contraception.2004;69:79 – 81
4. Mikolajczyk RT, Stanford JB. A new method for estimating the effectiveness of emergency contraception that accounts for vari-ation in timing of ovulvari-ation and previous cycle length. Fertil Steril.2005;83:1764 –1770
5. Stanford JB, Mikolajczyk RT. Methodologic review of the effec-tiveness of emergency contraception.Curr Womens Health Rev.
2005;1:119 –129
6. Hu X, Cheng L, Hua X, Glasier A. Advanced provision of emer-gency contraception to postnatal women in China makes no difference in abortion rates: a randomized controlled trial. Con-traception.2005;72:111–116
7. Glasier A, Fairhurst K, Wyke S, et al. Advanced provision of emergency contraception does not reduce abortion rates. Con-traception.2004;69:361–366
8. Raine TR, Harper CC, Rocca CH, et al. Direct access to emer-gency contraception through pharmacies and effect on unin-tended pregnancy and STIs: a randomized controlled trial.
JAMA.2005;293:54 – 62
doi:10.1542/peds.2005-2491
To the Editor.—
pe-diatricians routinely counsel about and prescribe EC to adolescents.
I would like to commend the Committee on Adoles-cence for highlighting the following in their policy state-ment:
1. Although approved by the Food and Drug Adminis-tration (FDA) for use within 72 hours of unprotected intercourse, EC is effective up to 120 hours (or 5 days) after unprotected intercourse.
2. Pills with norethindrone (not just those with norges-trel and levonorgesnorges-trel) can be used for EC.
3. Plan B is the EC regimen of choice because of its higher efficacy and less adverse effects compared with combination methods of EC.
4. Plan B should be prescribed as 2 tablets in a single dose instead of splitting the dose 12 hours apart with-out compromising efficacy or inducing more adverse effects.
5. The policy statement includes the indications for us-ing EC when newer contraceptive methods such as the birth control patch and vaginal ring fail to provide adequate protection.
6. The statement emphasizes that a urine pregnancy test is not required to prescribe EC and that timing of unprotected intercourse during the menstrual cycle should not determine if EC is prescribed.
7. The statement recommends prescribing EC over the telephone, even to new patients, so as not to delay access.
8. The summary statement for the policy statement rec-ommends educating and counseling about EC at an-nual visits for all teens and young adult patients when sexuality issues are addressed, which would include male adolescents and those who have not yet become sexually active.
I would also like to comment on a few inaccuracies that I found in the policy statement, fact sheet, and speaking points that are primary a result of the policy already being out of date.
1. All 3 resources (the policy statement, fact sheet, and speaking points) discuss Preven (a combination oral contraceptive pill) as a recommended EC product. It should be noted that Preven was bought by the mak-ers of Plan B and subsequently taken off the market in 2004, because the company felt that Plan B was a superior product in terms of its efficacy and adverse-effect profile. Currently, there are very few pharma-cies that have any remaining Preven in stock. Once the residual stock is gone, Preven will no longer be available. A prescription for Preven might prove very difficult to fill and will thus create an additional bar-rier to EC access. I worry that pediatricians may not
be aware that Preven is no longer being produced. It should be emphasized that pediatricians should avoid writing prescriptions for Preven and rather should always provide prescriptions for Plan B.
2. The policy statement appropriately emphasizes that Plan B is the preferred regimen for EC. However, neither the speaking points nor the fact sheet for parents and adolescents have the same emphasis. Plan B is the regimen of choice for all prescriptions for EC, and the instructions should read as follows: “Take 2 tablets as soon as possible after unprotected inter-course up to 120 hours from unprotected sex. Fol-low-up in 10 to 14 days.”
3. The fact sheet and the speaking points do not empha-size that EC can be prescribed up to 120 hours after unprotected intercourse (as the policy statement uni-formly emphasizes). Pediatricians should be aware that restricting EC to 72 hours after unprotected in-tercourse limits access to EC. However, the message that the sooner EC is taken, the more likely it is to be effective should not be lost.
4. In Table 2 of the policy statement, Women’s Capitol Corporation is shown as the owner of Plan B, but it is now owned by Barr Laboratories.
5. The policy statement comments on the FDA activities related to changing Plan B to over-the-counter status but neglects to provide the following most recent news: Barr Laboratories submitted an application to the FDA on July 22, 2004, for a unique and unprec-edented dual prescribing status for Plan B in which they requested that the product be sold over-the-counter for women aged ⱖ16 years and remain a prescription-only product for womenⱕ15 years. The response to the application on August 26, 2005, from the FDA concluded that the available scientific data were sufficient to support the safe use of Plan B as an over-the-counter product, but only for women who areⱖ17 years of age. However, they were unable to make a decision because the FDA has never deter-mined whether a drug may be both prescription and over-the-counter on the basis of the age of the indi-vidual using the drug. A related concern is how an age-based distinction could be enforced. Public com-ments on questions related to the marketing of pre-scription and over-the-counter versions of the same active ingredient in a single package and enforcement of the age limitation were taken for 60 days; that period ended on November 1, 2005.
6. The Society for Adolescent Medicine published a po-sition paper that was not cited in this policy statement that affirms that EC should be available over-the-counter without an age restriction.2
statement. Raine et al3assessed the sexual and con-traceptive behaviors of young women aged 15 to 24 years who were randomly assigned to 1 of 3 EC-access groups: advance provision (with 3 packs of EC), pharmacy access, or clinic access (control group). Those in the advance-provision group were almost twice as likely to use EC as those in the clinic-access group at the 6-month follow-up. The advance-provision group did not have significantly higher fre-quency of unprotected intercourse compared with the clinic-access group. No other differences in con-traceptive or condom use or other sexual behaviors by group were found. At the 6-month follow-up, 8% of the young women in the study had become preg-nant, and 12% had acquired a sexually transmitted infection. There were no reductions in pregnancies or increases in sexually transmitted infections by group (advance provision, pharmacy access, or clinic access).
It is my understanding that the American Academy of Pediatrics will soon be publishing a pamphlet on EC for adolescents. This is a sorely needed resource for the pediatrician in practice, and I look forward to its avail-ability.
Melanie A. Gold, DO
Adolescent Health Services Children’s Hospital of Pittsburgh University of Pittsburgh School of Medicine Pittsburgh, PA 15213
REFERENCES
1. American Academy of Pediatrics, Committee on Adolescence. Emergency contraception.Pediatrics.2005;116:1026 –1035 2. Gold MA, Sucato GS, Conard LA, Hillard PJ; Society for
Ado-lescent Medicine. Provision of emergency contraception to adolescents: position paper of the Society for Adolescent Medi-cine.J Adolesc Health.2004;35:67–70
3. Raine TR, Harper CC, Rocca CH, et al. Direct access to emer-gency contraception through pharmacies and effect on contra-ception and STIs: a randomized controlled trial.JAMA. 2005; 293:54 – 62
doi:10.1542/peds.2005-2811
In Reply.—
The American Academy of Pediatrics (AAP) Committee on Adolescence welcomes Dr Gold’s letter and the op-portunity to correct the dose and product recommenda-tions of readily available emergency contraceptive (EC) products. Although the position paper1 was current at the time it was written and approved by the AAP, she is correct in noting that the availability of products has changed.
Drs Stanford and Mikolajczyk identify several addi-tional references that selectively address the lack of ev-idence for reductions in abortion rates for populations for which EC is available. However, whether
popula-tions in Scotland and China are reasonably extrapolated to the United States is not clear; in addition, recent studies suggesting that EC is associated with responsible sexual behavior in Scotland were not reviewed in our article or in the letter.
Their reading of the policy also somewhat misses the point. It is appropriate for primary care organizations to judge the available evidence and make recommenda-tions for best clinical practices even when extensive efficacy data are not available. The AAP believes that “[e]mergency contraception has the potential to further decrease the rate of unintended teen pregnancies in the United States.”
Jonathan Klein, MD, MPH
Chair, AAP Committee on Adolescence
REFERENCE
1. American Academy of Pediatrics, Committee on Adolescence. Emergency contraception.Pediatrics.2005;116:1026 –1035
doi:10.1542/peds.2005-2980
Perceived Increase in Mortality
After Process and Policy Changes
Implemented With Computerized
Physician Order Entry
To the Editor.—
We read the article by Han et al1with great interest. The authors of this study concluded that the implementation of computerized physician order entry (CPOE) in a pe-diatric critical care unit was associated with an unantic-ipated increase in mortality in children admitted via interfacility transport over a 5-month period. Although we acknowledge a growing concern that information technology that was intended to improve clinical care may actually facilitate certain types of errors,2,3 we are concerned that the authors’ conclusion is not justified by their data.
recti-fied later, although they did not clarify whether it was done before or after the study period ended.
The authors also describe 2 critical deficiencies that are reflective of inadequate preparation for implemen-tation of an electronic medical chart, not of CPOE per se. First was the lack of bandwidth capacity on their wireless network. This infrastructure deficiency likely contrib-uted to decreased face-to-face interactions between the nurse and the physician when orders were being entered and highlights the important concept that CPOE is not a replacement for verbal communication. Second, the au-thors state in their methods section that “no ICU-specific order sets had been programmed at the time of CPOE implementation but instead were developed over time after CPOE implementation.” In a consensus statement on successful CPOE implementation, Ash et al wrote that “order sets . . . must be developed, reviewed, and maintained for personal and/or departmental usage.”4In fact, the authors later acknowledge in their discussion section that “ongoing development of preprogrammed order sets has helped to reduce some of the upfront time cost of order entry,” but again, it is not clear whether this effort was made before or after the study period ended. In summary, we believe that inadequate preparation for CPOE implementation, highlighted by inappropriate policy and process changes, insufficient infrastructure, and lack of critical care order sets, contributed substan-tially to the increased mortality rate reported by Han et al. We applaud the authors for highlighting the impor-tance of careful workflow redesign in CPOE implemen-tation, particularly in the intensive care environment, but we caution other readers not to dismiss the impor-tance of CPOE as a tool to help prevent medical errors and enhance patient safety. Indeed, other authors from the same institution have demonstrated a hospital-wide decrease in harmful adverse drug events,5 and other studies have described decreased medical errors in both a PICU6and an NICU7with successful CPOE implemen-tation.
Chris Longhurst, MD, MS Paul Sharek, MD, MPH
Department of Pediatrics
Jin Hahn, MD
Departments of Neurology and Pediatrics, Stanford University School of Medicine, Stanford, CA 94305
Jill Sullivan, RN, MSN
Department of Nursing Lucile Packard Children’s Hospital, Palo Alto, CA 94304
David Classen, MD, MS
Department of Medicine, University of Utah School of Medicine, Salt Lake City, UT 84132
REFERENCES
1. Han YY, Carcillo JA, Venkataraman ST, et al. Unexpected in-creased mortality after implementation of a commercially sold computerized physician order entry system.Pediatrics.2005;116: 1506 –1512
2. Koppel R, Metlay JP, Cohen A, et al. Role of computerized physician order entry systems in facilitating medication errors.
JAMA.2005;293:1197–1203
3. Berger RG, Kichak JP. Computerized physician order entry: helpful or harmful?J Am Med Inform Assoc.2004;11:100 –103 4. Ash JS, Stavri PZ, Kuperman GJ. A consensus statement on
considerations for a successful CPOE implementation.J Am Med Inform Assoc.2003;10:229 –234
5. Upperman JS, Staley P, Friend K, et al. The impact of hospital-wide computerized physician order entry on medical errors in a pediatric hospital.J Pediatr Surg.2005;40:57–59
6. Potts AL, Barr FE, Gregory DF, Wright L, Patel NR. Computer-ized physician order entry and medication errors in a pediatric critical care unit.Pediatrics.2004;113:59 – 63
7. Cordero L, Kuehn L, Kumar RR, Mekhjian HS. Impact of com-puterized physician order entry on clinical practice in a newborn intensive care unit.J Perinatol.2004;24:88 –93
doi:10.1542/peds.2005-3048
To the Editor.—
The article by Han et al1 reports a “direct association” between increased mortality rate and the implementa-tion of a commercial computerized physician order entry (CPOE) system in a tertiary care PICU. The Children’s Hospital of Pittsburgh (CHP) investigators observed a significant increase in raw mortality and in severity-adjusted mortality rates. The investigators concluded that the problems they encountered were unanticipated and embedded in the CPOE system rather than being a result of choices in how this technology was adopted. Problems specifically mentioned were the inability to preregister patients, additional time needed to enter or-ders, the need for a second physician devoted solely to entering orders, nurses spending too much time at the computer and away from the bedside, delays in admin-istration of critical medications, premature termination of standing medication orders, mistiming of medication administration, and using a generic platform in a highly specialized setting.
decentralized to a centralized pharmacy and acceptance of a CPOE platform that did not allow simultaneous users to enter/review medications were counterintuitive in a busy critical care unit. Furthermore, it seems that computer terminals were not optimally placed or were immobile and insufficient in number to allow nurses to carry on their work at the bedside, thus interfering with patient care.
Between April and December 2002, the Cincinnati Children’s Hospital Medical Center (CCHMC) imple-mented CPOE throughout the institution. CCHMC is a 423-bed facility with 1045 physicians and 1750 nurses. The PICU has a capacity of 25 beds and an average daily census of 18 children. The CPOE application is part of a larger integrating clinical information system, the core applications of which include a Web-based portal, CPOE, clinical documentation, and a data repository (Invision; Siemens Medical Solutions, Malvern, PA). These core ap-plications interface with other hospital-based information systems including radiology PACS, laboratory, pharmacy, admissions, and dietary. The CPOE platform is linked to numerous intranet- and Internet-based resources and clin-ical decision-support tools such as medication dose-range checking, hospital policies, the medication formulary, a discharge summary system, and an Internet search engine. The CPOE system is accessed through fixed and wireless workstations throughout the institution. Each week, 30 000 orders are generated through the system, 90% of which are entered directly into the computer by physicians or advance practice nurses.
Our experience with implementing CPOE in a PICU setting was quite different from that noted by Han et al. Critical care physicians, nurses, and respiratory thera-pists were heavily involved in our CPOE system de-sign before implementation. Pediatric-specific informa-tion (including critical care order sets, pathways for the entry of continuous infusions, and critical care–specific medication dose-range checking) was developed by critical care clinicians and incorporated into the sys-tem well in advance of implementation. Workflow in-compatibility with the CPOE system was analyzed and deficiencies corrected in advance of implementation. The system was implemented in a stepwise fashion, with entry into the critical care unit only after a successful pilot in a medical unit. Live user support was offered 24 hours/day, 7 days/week, until no longer needed. Ongo-ing feedback was not only accepted, but was solicited. Other important issues such as optimal workstation (fixed and wireless) numbers and location and standard-ized nomenclature (ie, centigrade versus Fahrenheit) were addressed before implementation.
Unlike CHP, the CCHMC did not observe a rise in mortality rates in its critical care unit after implementing CPOE. In fact, the raw mortality rate has declined, and the severity-adjusted mortality ratio has remained stable (Fig 1).
The CHP is to be congratulated for adopting an im-portant technology and persevering despite imim-portant challenges in their critical care setting. The pediatric community will benefit from the lessons they have learned. The study confirms the importance of having a well-supported plan for adopting and implementing CPOE. Our experience at the CCHMC validates the no-tion that CPOE can be implemented successfully with numerous benefits and without an adverse impact on mortality rates.
Brian R. Jacobs, MD Richard J. Brilli, MD Kim Ward Hart, MA
Division of Critical Care Medicine Cincinnati Children’s Hospital Cincinnati, OH 45229
REFERENCES
1. Han YY, Carcillo JA, Venkataraman ST, et al. Unexpected in-creased mortality after implementation of a commercially sold computerized physician order entry system.Pediatrics.2005;116: 1506 –1512
2. Ash JS, Stavri PZ, Kuperman GJ. A consensus statement on considerations for a successful CPOE implementation.J Am Med Inform Assoc.2003;10:229 –234
3. Ahmad A, Teater P, Bentley TD, et al. Key attributes of a successful physician order entry system implementation in a multi-hospital environment. J Am Med Inform Assoc. 2002;9: 16 –24
doi:10.1542/peds.2005-3116
To the Editor.—
Han et al1describe a retrospective study in which the mortality rate for interfacility transfers into an ICU was compared before and after implementation of a comput-erized provider order entry (CPOE) system. The authors found that the mortality rate increased from 2.8% (30 deaths of 1394 patient transfers during 13 months) be-fore CPOE implementation to 6.6% (36 deaths of 548 FIGURE 1
transfers during 5 months) after. The authors conclude that the increased mortality was associated directly with modifications in standard clinical processes, including the following changes: (1) not allowing order commu-nication until the patient was physically present and registered in the admitting system; (2) relocating medi-cation dispensing to a central (rather than a satellite) pharmacy; (3) increasing the physical separation of nursing and physician staff during the time that orders were generated; (4) implementing computerized order entry; and (5) system-wide provider role changes to support the CPOE system.
Perhaps the most important lesson from this study is that there exists an intimate association between care-delivery processes and health information technology. Any shift in the methods used to manage patient care (such as implementing and using a CPOE system) is associated with significant changes in clinical workflows, communication among providers, and distribution of responsibilities.2–4 Decades of research in medical infor-matics have underscored the importance of this ob-servation, a message that was not lost on the authors. In this study, they note that the increased unadjusted mortality may reflect problems with the process of change, including the extremely rapid implementation plan adopted by their organization. The authors de-scribe other major changes in workflow and patient care processes that occurred coincident with the CPOE sys-tem implementation. For example, their institution changed its policies to prohibit providers from entering patient orders before the patient had physically arrived and had been registered. This change was not a function of the CPOE system but rather of how the institution chose to implement it. The authors list other similar workflow changes that occurred coincident with the CPOE system implementation, such as the elimination of bedside stocks of critical drugs, resulting in the need to request these drugs when needed from a remote phar-macy.
The authors’ findings merit additional scrutiny be-cause, as they noted, there are numerous limitations to the study’s design. The primary study outcome, mortal-ity rates, naturally change over time with trends that are sensitive to many factors including policy changes at the hospital, staffing ratios, and seasonal variations in dis-ease. In a pre-post single-crossover design, as this study used, there is no way to disentangle trends associated with time from the effects of the CPOE system-imple-mentation process. In cases such as this, investigators must aggressively adjust for all possible measured con-founders, taking pains to identify and include all that might reflect changes occurring over calendar time, in-cluding seasonal effects, primary diagnoses, a large num-ber of comorbidities,5staffing levels, and any changes in care-delivery and hospital-workflow processes. This lim-itation is amplified by the concern that many of the
statistical methods used in this study have known prob-lems that can invalidate apparently significant statistical associations.6–8 For observational studies such as this one, in which the number of deaths is limited, the pre-ferred method for adjusting for confounding is propen-sity score analysis.9The propensity model should include all measured confounders regardless of whether they are statistically significant. A more rigorous approach that pays attention to identifying and including all known confounders in an appropriate statistical analysis might influence the outcomes observed in this study. It is also important to realize the very unique population of crit-ically ill children in this study. Interfacilty transfers occur under a variety of conditions. Mortality in this group may be a result of factors including suboptimal diagnosis and management at the transferring site, delays in trans-fer, problems during transit, and other factors remote from the study site.
Implementing health information technology such as CPOE and electronic health record systems is a complex process.10In the commentary that accompanied the ar-ticle by Han et al, Gesteland et al emphasized this lesson: “Deploying a sophisticated clinical-applications platform including CPOE in 6 days is an audacious task and leaves little margin for error in adapting highly evolved work processes to the new environment.”11 There exist well-established approaches to enabling high-risk organizational change.12 Untoward effects have been well described when the process is not performed appropriately.13–15 Additional studies need to be designed and conducted to understand whether there are unique implications of process change in ICU and other acute care environ-ments and to determine how pediatric health care affects these implications.16,17However, as noted recently by the Institute of Medicine,18,19the preponderance of evidence to date strongly suggests that CPOE systems can reduce medication-related errors of commission and may be useful for improving compliance with errors of omis-sion.20,21Readers of the article by Han et al should not abort plans to implement CPOE systems under the belief that CPOE itself can increase harm; rather, they should proceed with implementations cautiously, applying the lessons about systems implementation learned and pub-lished by others. In this regard, a more appropriate title for the article might have emphasized that any increase in observed mortality at the authors’ institution was associated with the process of implementing change rather than with the CPOE system itself.
con-ducted throughout the design and implementation pro-cess, usability testing in a controlled environment, and a stepwise rollout in which each subsequent institutional unit learns from challenges experienced by the previous unit. Pertinent system evaluations should supplement classic technical software and hardware analyses with human-interface and human-interaction testing. There also should be extensive end-user testing in realistic situations that takes place with a “frozen” software de-sign. Finally, there must be well-designed “break-the-glass” functionality that allows users to do what they think is best for the patient even when the computer system does not allow it (eg, writing orders before a critically ill patient is registered in the hospital).22
The CPOE system-implementation process described by Han et al did not incorporate steps or elements known to ensure system dependability and usability. For example, CPOE systems commonly include tools de-signed to improve safety and time efficiency, such as “order sets.”23 Order sets allow providers to select and order multiple related items such as medications and diagnostics tests with only a few mouse clicks or key-strokes. According to the authors, their institution chose an implementation approach that did not include order sets, although the commercially available system they implemented offered the ability to construct them. The authors also describe unforeseen technical problems that occurred during the rapid implementation process (eg, overloaded wireless networks that slowed down CPOE systems). Technical problems such as these should have been evident in testing that typically takes place before system implementation.
Over the past 5 years, numerous investigators have outlined case studies documenting successful and failed CPOE implementations.14,24Many successful implemen-tations have received the Nicholas E. Davies award (these case studies are available at www.himss.org/ASP/ daviesAward.asp). Fortunately, there now exists enough shared and published experience with CPOE implemen-tation that each institution should not need to rediscover problems that others have found. However, with the increasing demand for technology infrastructure in med-icine, there continues to be an insufficient number of individuals trained in biomedical informatics to provide consultation. Efforts such as the American Medical In-formatics Association’s 10 ⫻ 10 program and the Na-tional Library of Medicine’s short course in medical in-formatics, fellowship training programs, and career development grants all promise to improve the penetra-tion of informatics knowledge in the pediatric workforce. Awards such as the Davies award recognize best practices for CPOE and electronic health record system implemen-tation. Ongoing efforts by the Certification Commission for Health Information Technology (CCHIT) are designed to inform purchasers about the quality of commercial sys-tems. These nationally visible products and programs add
to the knowledge that can be applied to CPOE and elec-tronic health record system implementation.
S. Trent Rosenbloom, MD, MPH
Departments of Biomedical Informatics and Pediatrics Vanderbilt University School of Medicine Nashville, TN 37232
School of Nursing Vanderbilt University Nashville, TN 37235
Frank E. Harrell, Jr, PhD
Department of Biostatistics Vanderbilt University School of Medicine Nashville, TN 37232
Christoph U. Lehmann, MD
Department of Pediatrics and Division of Health Information Sciences Johns Hopkins University Baltimore, MD 21287
Joseph H. Schneider, MD, MBA
Department of Pediatrics University of Texas Southwestern Medical Center Dallas, TX 75390
Children’s Medical Center Dallas, TX 75235
S. Andrew Spooner, MD, MS
Department of Pediatrics University of Tennessee College of Medicine Memphis, TN 38163
Kevin B. Johnson, MD, MS
Departments of Biomedical Informatics and Pediatrics Vanderbilt University School of Medicine Nashville, TN 37232
ACKNOWLEDGMENT
The project was supported by US National Library of Med-icine grant 1K22 LM008576-01 (to Dr Rosenbloom).
REFERENCES
1. Han YY, Carcillo JA, Venkataraman ST, et al. Unexpected increased mortality after implementation of a commercially sold computerized physician order entry system. Pediatrics.
2005;116:1506 –1512
2. Braude RM. People and organizational issues in health infor-matics.J Am Med Inform Assoc.1997;4:150 –151
3. Lorenzi NM, Riley RT. Knowledge and change in health care organizations.Stud Health Technol Inform.2000;76:63– 69 4. Lorenzi NM, Riley RT. Organizational issues ⫽ change. Int
J Med Inform.2003;69:197–203
5. Stukenborg GJ, Wagner DP, Harrell FE Jr, et al. Hospital dis-charge abstract data on comorbidity improved the prediction of death among patients hospitalized with aspiration pneumonia.
J Clin Epidemiol.2004;57:522–532
6. Sun GW, Shook TL, Kay GL. Inappropriate use of bivariable analysis to screen risk factors for use in multivariable analysis.
J Clin Epidemiol.1996;49:907–916
8. Greenland S. When should epidemiologic regressions use ran-dom coefficients?Biometrics.2000;56:915–921
9. D’Agostino RB Jr. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group.Stat Med.1998;17:2265–2281
10. Kuperman GJ, Gibson RF. Computer physician order entry: benefits, costs, and issues.Ann Intern Med.2003;139:31–39 11. Gesteland PH, Nebeker JR, Gardner RM. These are the
tech-nologies that try men’s souls: common-sense health informa-tion technology.Pediatrics.2006;117:216 –217
12. Stablein D, Welebob E, Johnson E, Metzger J, Burgess R, Classen DC. Understanding hospital readiness for computer-ized physician order entry. Jt Comm J Qual Saf. 2003;29: 336 –344
13. N.V.: Cedars slips on CPOE; The editors and reporters ofModern Physicianchose the following as the 12 most important stories of the past year.Modern Physician.2003;7:19
14. Shane R. Computerized physician order entry: challenges and opportunities.Am J Health Syst Pharm.2002;59:286 –288 15. Sittig DF, Stead WW. Computer-based physician order entry:
the state of the art.J Am Med Inform Assoc.1994;1:108 –123 16. Johnson KB. Barriers that impede the adoption of pediatric
information technology. Arch Pediatr Adolesc Med. 2001;155: 1374 –1379
17. Johnson KB, Davison CL. Information technology: its impor-tance to child safety.Ambul Pediatr.2004;4:64 –72
18. Institute of Medicine, Committee on Quality in Healthcare in America.To Err Is Human: Building a Safer Health System. Wash-ington, DC: National Academy Press; 1999
19. Aspden P, Corrigan JM, Wolcott J, Erickson SM, eds.Patient Safety: Achieving a New Standard for Care. 1st ed. Washington, DC: National Academies Press; 2004
20. Bates DW, Teich JM, Lee J, et al. The impact of computerized physician order entry on medication error prevention.J Am Med Inform Assoc.1999;6:313–321
21. Potts AL, Barr FE, Gregory DF, Wright L, Patel NR. Comput-erized physician order entry and medication errors in a pedi-atric critical care unit.Pediatrics.2004;113:59 – 63
22. Bates DW, Kuperman GJ, Wang S, et al. Ten commandments for effective clinical decision support: making the practice of evidence-based medicine a reality.J Am Med Inform Assoc.2003; 10:523–530
23. McAlearney AS, Chisolm D, Veneris S, Rich D, Kelleher K. Utilization of evidence-based computerized order sets in pedi-atrics.Int J Med Inform.2005; In press
24. Aarts J, Berg M. A tale of two hospitals: a sociotechnical appraisal of the introduction of computerized physician order entry in two Dutch hospitals.Medinfo.2004;11:999 –1002
doi:10.1542/peds.2005-3163
In Reply.—
We appreciate the insightful comments of Longhurst et al, Jacobs et al, and Rosenbloom et al in response to our study findings. They all emphasize the critical impor-tance of careful planning and preparation for successful computerized provider order entry (CPOE) implementa-tion. We wholeheartedly agree. However, these authors also assert that our observations probably resulted from inadequate planning or foresight exercised by the CPOE project team at the Children’s Hospital of Pittsburgh (CHP). Although we cannot, ourselves, refute this asser-tion, we note that members of CHP’s CPOE project team seemed to have been well aware of the many potential
pitfalls that can impede successful CPOE implementa-tion.1The report by Upperman et al,1which described in great detail how CPOE was introduced at CHP, suggests that considerable thought and effort went into our in-stitution’s adoption of CPOE technology. This endeavor was rewarded, in fact, with a significant reduction in harmful adverse drug events.2In this regard we ask, had we not performed our independent investigation specif-ically examining mortality outcomes, would anyone have questioned the adequacy of the preparations made at CHP? Clearly, debating the potential merits and short-comings of various CPOE implementation strategies is important and necessary, but we submit that this debate must first proceed with a uniform definition of success. Because the mandate for widespread implementation of CPOE throughout all US hospitals stems from the prom-ise that this technology will save patient lives, we further submit that the realization of this promise should serve as the “gold-standard” definition of success.
Longhurst et al, Jacobs et al, and Rosenbloom et al also seem to make a clear distinction between CPOE and the implementation process, implying that our observation resulted entirely from faulty implemen-tation and not from CPOE technology, itself. Although we acknowledge this possibility in our article, we find it extremely difficult to separate one element from the other. A CPOE system does not operate in isolation, and its proper functioning requires its seamless inte-gration into a strong and dynamic health information technology (HIT) infrastructure. In its broadest sense, implementation encompasses not only solving poten-tial clinical workflow problems but also building this HIT infrastructure, resolving systems-integration glitches, and overcoming machine-human–interface obstacles. Al-though CPOE, itself, may not be to blame for undesired patient outcomes, a well-designed, well-programmed, user-friendly CPOE software architecture constructed “from the ground up” with a clear understanding of how clinicians think and operate in real-life situations can only serve to facilitate the implementation process. Conversely, a general, minimally modifiable, clinically awkward CPOE platform may require considerable creative efforts to “retrofit” this technology into operational form. Al-though CPOE, itself, may not be at fault for network problems related to ever-present threats from Internet worms and viruses, the vulnerability or resilience of a specific CPOE platform and its HIT infrastructure to properly function during periods of attack and after fire-walls and other network security upgrades have been installed seem to be considerations intrinsic to the adop-tion of this technology.
cau-tiously interpret our data. We remain unclear, however, how our results might be invalid, because we have per-formed a well-accepted method of regression analysis. Still, to appease their criticism of our approach, we have subsequently performed a propensity score analysis as requested. Propensity scores were generated with CPOE as the dependent variable and with all of the variables listed in Table 1 of our article, except for Pediatric Risk of Mortality (PRISM) score, as independent variables. The propensity scores were then recorded into deciles, and a logistic-regression model was fitted using CPOE, decile group, and the interaction between CPOE and decile group as predictors of mortality. The interaction be-tween CPOE and decile group was not found to be significant and was subsequently dropped from the model. Results then indicated that CPOE was signifi-cantly associated with increased odds of mortality (odds ratio [OR]: 2.420; 95% confidence interval [CI]: 1.508 – 3.883). The addition of PRISM score into this regression model continued to demonstrate increased odds of mor-tality (OR: 3.130; 95% CI: 1.848 –5.302).
We applaud Jacob et al for continuing to monitor mortality outcomes (a practice we strongly support) af-ter CPOE implementation at their institution. We find it very reassuring that they observed no increase in PICU mortality after using a different implementation strategy (and different CPOE platform). The CPOE project team at Cincinnati Children’s Hospital Medical Center should be commended for their diligence and their particular attention devoted to tackling the unique challenges of the PICU environment. On the other hand, if we are permitted to briefly play “devil’s advocate,” it seems that severity-of-illness–adjusted mortality rates did not sub-stantially improve after CPOE implementation, either. As fellow pediatric intensivists, Jacobs et al should be well aware of the excitement after the publication of the adult clinical trials of drotrecogin alpha for severe sepsis that showed significant improvements in survival.3 In contrast, the pediatric clinical trials were terminated re-cently because similar survival benefits could not be demonstrated among children. Although drotrecogin al-pha has received Food and Drug Administration ap-proval as a treatment for adults with severe sepsis, this approved indication has not been extended to the pedi-atric population.
We wish to make it clear that despite our unexpected study results, we continue to believe that CPOE holds great promise as a tool to improve patient care and save lives, and we reject any proposal that calls to abandon this important technology. At the same time, the excel-lent points made by Longhurst et al, Jacobs et al, and Rosenbloom et al reinforce our impression that the very complexity of CPOE technology and its implementation requires a thorough, organized, systems approach to adequately address their many concerns and may be best resolved through carefully designed, multicenter trials
supported by the National Institutes of Health and/or Food and Drug Administration. If the HIT community finds this suggestion unfeasible or unreasonable because of the unique nature of computer/software technology, another approach may be to examine how the aviation industry, long considered a model industry for maintain-ing high safety standards, has addressed the incorpora-tion of these technologies into its field. We note that Federal Aviation Administration oversight ensures the mission-critical reliability of any new computer-related device or software throughout all US airport control towers. Alternatively, an independent self-governance HIT body analogous to the American Board of Pediatrics, for example, might provide another means to certify the proper deployment of CPOE.
Yong Y. Han, MD
Department of Pediatrics and Communicable Diseases University of Michigan Medical School Ann Arbor, MI 48109
Joseph A. Carcillo, MD Shekhar T. Venkataraman, MD Robert S.B. Clark, MD R. Scott Watson, MD, MPH Hu¨lya Bayir, MD Richard A. Orr, MD
Departments of Critical Care Medicine and Pediatrics University of Pittsburgh School of Medicine Pittsburgh, PA 15213
Trung C. Nguyen, MD
Department of Pediatrics Baylor College of Medicine Houston, TX 77030
REFERENCES
1. Upperman JS, Staley P, Friend K, et al. The introduction of computerized physician order entry and change management in a tertiary pediatric hospital.Pediatrics.2005;116(5). Available at: www.pediatrics.org/cgi/content/full/116/5/e634
2. Upperman JS, Staley P, Friend K, et al. The impact of hospital-wide computerized physician order entry on medical errors in a pediatric hospital.J Pediatr Surg.2005;40:57–59
3. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis.N Engl J Med.2001;344:699 –709
doi:10.1542/peds.2006-0161
Failure to Thrive as Distinct From
Child Neglect
To the Editor.—
Com-mittee on Child Abuse and Neglect and ComCom-mittee on Nutrition.
Six aspects of the report are particularly troubling. First, the report does not make it sufficiently clear that FTT as a manifestation of child neglect represents a mi-nority of children with FTT. We agree that, in some cases, FTT may be a marker of neglect and that the diagnosis of abuse and/or neglect should be considered if there is any history of “intentional withholding of food from the child; strong beliefs in health and/or nutrition regimens that jeopardize a child’s well-being; and/or family that is resistant to recommended interventions despite multidisciplinary team approach.” Indeed, chil-dren with both FTT and neglect have lower cognitive skills initially and years after treatment2,3than children with either FTT or neglect alone.
A second concern is the conflation of poverty with child neglect. In any social class, family violence or al-cohol/substance abuse should raise protective concerns as possible impediments to adequate care of any child, FTT or not. Many of the other risk factors for neglect as a cause of FTT listed in the report are highly nonspecific and could be applied to many conditions associated with childhood poverty (eg, iron deficiency, overweight, and developmental delay). FTT in combination with neglect is a relatively rare occurrence, as illustrated by a Febru-ary 2004 report from the Massachusetts Department of Public Health covering 6 years of statewide experience with⬃1700 children referred to 6 multidisciplinary clin-ics for FTT.4 Almost 60% were from families below 200% of the federal poverty level, and 31% lived with single mothers. Most of these children would fulfill at least 1 of the criteria for neglect, as presented in the AAP report, yet the multidisciplinary clinicians providing care to the children judged abuse/neglect to be a relatively rare concern and referred only 7% of them to protective services.
Our third concern is that a casual reading of this nuanced report might lead some readers to the conclu-sion that the family of a child identified with FTT should be considered abusive and/or neglectful until proven otherwise. Too often neglect is assumed to be responsi-ble when organic causes are not found; neglect should not be a diagnosis by exclusion. Thus, the report might inadvertently serve to promote a punitive rather than a therapeutic approach to families, in effect “blaming the victim”5for failing to meet the needs of the child.
A fourth concern is that the report reflects the early literature on FTT, which was based on hospitalized sam-ples of children with almost no citation of important articles published in the last 10 years. Evidence supports the benefits of treating most children with FTT through sustained participation in multidisciplinary outpatient clinics and home visiting programs.6–9 Even when hos-pitalization is unavoidable in the short-term for the most
seriously malnourished, acutely infected, or otherwise acutely ill children, successful follow-up care requires a multidisciplinary team to work with the family in their home, wherever possible, which is an approach en-dorsed by the Committee on Nutrition in the Pediatric Nutrition Handbook.10 Follow-up studies based on chil-dren recruited from primary care or population-based surveys have shown that the long-term cognitive and academic consequences of early FTT are much less se-vere than reported previously from samples of hospital-ized children.11
The fifth concern is the vague approach to the med-ical evaluation of children with FTT. In the Massachu-setts sample, for example, nearly 30% of children with FTT were anemic. Although we agree with the recom-mendation to avoid unfocussed laboratory testing, it is often beneficial to test children for iron deficiency and other common but occult conditions impeding growth, such as lead poisoning.
Our final concern is that an unintended consequence of this statement may be to increase the number of children at risk of nutritional deprivation. Food insecu-rity is highly prevalent in the United States and has been increasing throughout the decade, affecting 42% of low-income households with children ⬍6 years of age in 2004.12 In the current climate, safety-net programs for the poor are often in jeopardy in response to budgetary pressures.13Despite evidence of nutritional benefits for low-income young children whose families participate in the Supplemental Nutrition Program for Women, In-fants, and Children (WIC) and Food Stamp programs,14,15a statement attributed to the AAP that nutritional growth failure is often a consequence of neglect could fuel ar-guments that child nutrition programs are unneces-sary.16
The AAP policy statements and clinical guidelines are rightfully accorded great significance, not only among clinicians but also among policy makers and the general public. We recommend that the statement be clarified to emphasize that the AAP recommends adequate nutri-tional resources for all low-income families with chil-dren, long-term multidisciplinary services for children with FTT, and referral to protective services for any children suspected of experiencing abuse/neglect, what-ever their growth.
Maureen M. Black, PhD Howard Dubowitz, MD, MS
Department of Pediatrics University of Maryland School of Medicine Baltimore, MD 21201
Patrick H. Casey, MD
Diana Cutts, MD
Department of Pediatrics Hennepin County Medical Center Minneapolis, MN 55415
Robert F. Drewett, DPhil
Department of Psychology University of Durham Durham DH1 3HP, United Kingdom
Dennis Drotar, PhD
Department of Pediatrics Case Western Reserve University School of Medicine Cleveland, OH 44106
Deborah A. Frank, MD
Department of Pediatrics Boston Medical Center Boston, MA 02118
Robert Karp, MD
Department of Pediatrics State University of New York Downstate Medical Center Brooklyn, NY 11203
Daniel B. Kessler, MD
Department of Pediatrics University of Arizona College of Medicine Tucson, AZ 85724
Children’s Health Center St Joseph’s Hospital Phoenix, AZ 85013
Alan F. Meyers, MD, MPH
Department of Pediatrics Boston University School of Medicine Boston, MA 02118
Charlotte M. Wright, MD
Department of Child Health Glasgow University Glasgow G12 8QQ, Scotland
REFERENCES
1. Block RW, Krebs NF; American Academy of Pediatrics, Com-mittee on Child Abuse and Neglect; American Academy of Pediatrics, Committee on Nutrition. Failure to thrive as a man-ifestation of child neglect.Pediatrics.2005;116:1234 –1237 2. Mackner LM, Starr RH Jr, Black MM. The cumulative effect of
neglect and failure to thrive on cognitive functioning. Child Abuse Negl.1997;21:691–700
3. Kerr M, Black MM, Krishnakumar A. Failure-to-thrive, mal-treatment and the behavior and development of 6-year-old children from low-income urban families. Child Abuse Negl.
2000;24:587–598
4. Nyambose J, Cunningham K. The Massachusetts Growth and Nutrition Program summary report FY 2003. Available at: www. mass.gov/dph/fch/growth/report㛭2003.pdf. Accessed December 1, 2005
5. Ryan W. Blaming the Victim. New York, NY: Random House; 1976
6. Bithoney WG, McJunkin J, Michalek J, Snyder J, Egan H, Epstein D. The effect of a multidisciplinary team approach on weight gain in nonorganic failure-to-thrive children. J Dev Behav Pediatr.1991;12:254 –258
7. Wright CM, Callum J, Birks E, Jarvis S. Effect of community based management in failure to thrive: randomised controlled trial.BMJ.1998;317:571–574
8. Kessler DB, Dawson P, eds. Failure to Thrive and Pediatric Undernutrition: A Transdisciplinary Approach. Baltimore, MD: Paul H. Brookes Publishing Co; 1999
9. Black MM, Feigelman S, Cureton PL. Evaluation and treatment of children with failure-to-thrive: an interdisciplinary perspec-tive.J Clin Outcomes Manage.1999;6:60 –73
10. Kleinman RE, ed. Failure to thrive (pediatric undernutrition). In:Pediatric Nutrition Handbook. 4th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2004:453
11. Corbett SS, Drewett RF. To what extent is failure to thrive in infancy associated with poorer cognitive development? A re-view and meta-analysis. J Child Psychol Psychiatry. 2004;45: 641–54
12. Nord M, Andrews M, Carlson S. Household food security in the United States, 2004: economic research report No. 11. Avail-able at: www.ers.usda.gov/publications/err11/err11.pdf. Ac-cessed December 1, 2005
13. Hulse C. Congress mounts drive for big budget cuts.New York Times. October 31, 2005:A15
14. Rose D, Habicht JP, Devaney B. Household participation in the Food Stamp and WIC programs increases nutrient intakes of preschool children.J Nutr.1998;128:548 –555
15. Black MM, Cutts DB, Frank DA, et al. WIC and infants’ growth, health, and food security: a multisite surveillance study. Pedi-atrics.2004;114:169 –176
16. Besharov D. We’re feeding the poor as if they’re starving.
Washington Post. December 8, 2002:B1
doi:10.1542/peds.2005-3043
In Reply.—
In reference to the letter from Black et al, we agree that failure to thrive (FTT) as a manifestation of child neglect represents a small but highly significant percentage of cases, that conflation of poverty with neglect is inappro-priate, that all etiologies and conditions associated with FTT should be considered during evaluations, that cur-rent literature on FTT is important, and that medical evaluations should be thorough and specific. Although the members of the Committee on Child Abuse and Neglect respect the work of the authors of the letter, we respectfully urge them and other readers to read the report with the same care that was used in its crafting. We share their concern that “a casual reading” of any medical literature can lead to inappropriate conclusions. We would remind all readers that a clinical report issued by the American Academy of Pediatrics (AAP) repre-sents “guidance for the clinician in rendering pediatric care,” rather than representing definitive policy. In this case, the statement provides guidance to enable physi-cians to consider neglect as a cause of FTT and to guide their thinking when this is the case. As the statement clearly states, it “serves as a guide for the assessment, management, and support of children with FTT as a manifestation of neglect.”
cause of FTT is nutritional deficiency,” followed by a long, but not all-inclusive, list of problems other than neglect associated with FTT. As the authors point out, “children with both FTT and neglect have lower cogni-tive skills….” For the general pediatrician who is not an expert in either FTT or child abuse and neglect, remem-bering to include neglect in the differential may save lives and improve brain development in some children. The report does not conflate poverty with child ne-glect. It states that “[p]overty is the greatest single risk factor for FTT,” not neglect, which is supported by 2 citations in the letter by Black et al. Although we agree that the listed risk factors in our report are nonspecific, we argue that being nonspecific does not equate with irrelevant. Risk factors are not “criteria,” as misperceived by Black et al.
It is unfortunate that Black et al continue to subscribe to the attitude that a report to a child protective services agency is only punitive, when it should be a valuable part of a multidisciplinary approach to supporting a fam-ily and/or protecting a child.
Our report does not contain references to more recent literature, which is a valid concern of Black et al, but it clearly refers readers to the very currentPediatric Nutri-tion Handbookfrom the AAP for an expanded discussion of FTT and references. The report was carefully coordi-nated with the Committee on Nutrition to ensure that any general statements were in line with the informa-tion in the handbook. The fifth concern in the letter refers to a “vague approach to the medical evaluation of children with FTT.” We chose to focus on the important components of history, physical examination, feeding observation, home visit, and thinking clearly about lab-oratory studies but did not include detail to preserve focus on the point of the report, clearly stated in the title, with reference again to thePediatric Nutrition Handbook for additional information.
Response to the authors’ last concern is difficult, be-cause it misquotes the report and is highly conjectural. Nowhere in the statement do we contend that “nutri-tional growth failure is often a consequence of neglect.” To speculate that policy makers might ignore child nu-trition programs because a small proportion of FTT cases are caused by severe neglect is a spurious argument. In sum, Committee on Child Abuse and Neglect members recommend a careful reading of all AAP statements, because many, like ours, focus on a specific issue and should not be inappropriately generalized.
Robert W. Block, MD, FAAP
Chairperson American Academy of Pediatrics, Committee on Child Abuse and Neglect
doi:10.1542/peds.2006-0041
The Folic Acid Debate Continues
To the Editor.—Brent and Oakley1argue that 2000 additional healthy infants in the United States, or 200 000 more healthy infants in the world, could be born each year by requir-ing flour millers to increase the amount of folic acid in “enriched” flour. There is unequivocal science-based ev-idence that if women have an adequate amount of folic acid in their diet before conception, then their infants largely will be spared a life of serious complications associated with neural tube defects.
The story of efforts to eliminate the harmful effects of inadequate dietary intake of iodine, another essential nutrient, may be instructive. In the 1820s, Coindent demonstrated before the Swiss Society of Natural Sci-ences that most cases of goiter would regress within weeks of administration of iodine. Scientists, during the next 100 years, decided that it was too dangerous to increase consumption of iodine. The correction of iodine deficiency in populations had to wait for Marine (1917) and Kimble (1924) to show the effectiveness of iodized water and iodized salt. The fear of Jod-Basedow phe-nomenon (hyperthyroidism in a few when given su-pralarge doses of iodine [coined by Kocker in 1910]) still causes some scientists today to advocate for inadequate amounts of iodine to be added to salt.2Understanding the benefit that adequate iodine intake had on protect-ing brain development in the fetus and cumulative im-pact of boosting national intelligence came later.3
Food-standard authorities around the world would act with lightning speed if there were a toxin (such as lead and mercury) or an infectious agent (such as the mad cow product) in the nation’s food supply. Action would be swift and decisive. Some have argued that it would be unethical to require fortification with folic acid based on hypothetical risk. The ethical issue here is not what harm the addition of a little extra folic acid in the food supply would be; rather, it’s the needless harm inflicted during each day of delay by food-regulatory bodies that have the authority to prevent it.
Glen F. Maberly, MB, BS, MD, FRACP
Global Health Rollins School of Public Health of Emory University Atlanta, GA 30322
REFERENCES
1. Brent RL, Oakley GP Jr. The Food and Drug Administration must require the addition of more folic acid in “enriched” flour and other grains.Pediatrics.2005;116:753–755
2. Matovinovic J, Ramalingaswami R. Therapy and prophylaxis of endemic goiter. In:Endemic Goiter. WHO Monograph Series 44. Geneva, Switzerland: World Health Organization: 1960:394 – 401 3. Hetzel BS. Historical development of the concept of the
brain-thyroid relationships. In: Stanbury JB, ed.The Damaged Brain of Iodine Deficiency. New York, NY: Cognizant Communication Corporation; 1994:1–7
In Reply.—
We appreciate the historical perspective that Maberly provides the readers of Pediatricsconcerning the obsta-cles that have confronted scientists when new ideas or discoveries have been proposed to the scientific and lay communities.1,2 Maberly reflects on the lengthy battle that lasted more than a century before iodination of salt was established as an acceptable public health practice; even after acceptance occurred, there were a few dooms-day predictors of catastrophe or poisonings.
We remember from our medical history the trauma and ridicule to which Ignatz Philipp Semmelweis, an Austrian obstetrician, was subjected when he suggested that child bed fever (puerperal fever) was a contagious disease that could be markedly reduced by having the physicians wash their hands before delivering their preg-nant patients. He met bitter opposition from the medical community that at times amounted to persecution, and he eventually became insane. The attempt to introduce chlorination and fluoridation of water supplies met with opposition from the lay and scientific community be-cause of concern about the hypothetical risk of cancer and other hypothetical adverse effects. Although both chlorination and fluoridation have markedly improved the health of numerous populations, there are still some lay groups and scientists who have suggested that hypo-thetical health risks outweigh their benefits.
No child should develop folic acid–preventable spina bifida anywhere in the world. It is tragic when prolonged scientific and policy debate about hypothetical risks pre-vents public health agencies from implementing public health programs that will be of significant benefit to men, women, and children. We have advocated, in 3 commentaries since 2000,3–5 that the Food and Drug Administration (FDA) should maximize the prevention of folic acid–preventable neural tube defects by increas-ing the concentration of folic acid in “enriched” cereal grains. We believe that scientists and policy makers should sit down soon and objectively discuss our recom-mendation without being unduly influenced by those who raise hypothetical risks. When this discussion is held, we think that the FDA will increase the concen-tration of folic acid that is required in enriched cereal grain products. George Santayana wrote: “Those who cannot remember the past are condemned to repeat it.”6 We hope that we do not repeat the iodine history with folic acid and thereby continue for many years to have children born with serious birth defects that could have been prevented by appropriate public health actions.
Robert L. Brent, MD, PhD
Departments of Pediatrics, Radiology, and Pathology Thomas Jefferson University A.I. duPont Hospital for Children Wilmington, DE 19899
Godfrey P. Oakley, Jr, MD, MSPM
Department of Epidemiology Rollins School of Public Health of Emory University Atlanta, GA 30322
REFERENCES
1. Matovinovic J, Ramalingaswami R. Therapy and prophylaxis of endemic goiter. In:Endemic Goiter. WHO Monograph Series 44. Geneva, Switzerland: World Health Organization: 1960:394 – 401 2. Hetzel BS. Historical development of the concept of the
brain-thyroid relationships. In: Stanbury JB, ed.The Damaged Brain of Iodine Deficiency. New York, NY: Cognizant Communication Corporation; 1994:1–7
3. Brent RL, Oakley GP Jr, Mattison DR. The unnecessary epidemic of folic acid–preventable spina bifida and anencephaly.Pediatrics.
2000;106:825– 827
4. Brent RL, Oakley GP Jr. The Food and Drug Administration must require the addition of more folic acid in “enriched” flour and other grains.Pediatrics.2005;116:753–755
5. Brent RL, Oakley GP Jr. Triumph and/or tragedy: the present Food and Drug Administration program of enriching grains with folic acid.Pediatrics.2006;117:930 –932
6. Santayana G.Reason in Common Sense: The Life of Reason. New York, NY: Charles Scribner’s Sons; 1905:284
doi:10.1542/peds.2006-0012
Is Appendicitis the Only Cause of
an Acute Abdominal Pain in
Childhood?
To the Editor.—
Assessment and management of acute abdominal pain in the pediatric emergency department is recognized to be suboptimal much of the time. Green at al1drew atten-tion to this and endeavored to dispel an old surgical legend that analgesia may hinder diagnosis in the child with acute abdominal pain.
Given the large throughput of their department (39 000 per annum), we were surprised that the number of patients eligible for the Green at al study was only 162. We see ⬃20 000 patients per year, and at least 1 patient per day will present with acute abdominal pain that merits a surgical consultation, an incidence of al-most 2 in 100 attendances. The case histories cited led us to further question enrollment of patients into this study; one criteria was⬍48 hours of pain, yet 2 of the 4 patients discussed had⬎48 hours of pain.
The authors looked at clinician confidence in diagno-sis. Diagnostic uncertainties are faced daily and are chal-lenging for both clinicians and patients. It is therefore of relevance to doctors whether an intervention might make diagnosis harder. Confidence depends on many variables including personality type and experience. The authors did not describe in detail how they assessed confidence and whether allowances were made for these potential confounding factors. Was a validated scoring tool used? We felt that the question of diagnostic cer-tainty was important but too complex to have been looked at as a secondary outcome.
It is unacceptable for a child to lie in pain; therefore, it is important to be able to give analgesia without im-peding diagnosis. Unfortunately, the power of this study was too low to allow for meaningful conclusions, but it is useful to have highlighted this topic inPediatrics.
Chloe T. Child, MBChB, MRCPCH Khalid Haque, MD, FCPCH, FRCP
Department of Pediatrics Queen Mary’s Hospital for Children Surrey SM5 1AA, United Kingdom
REFERENCE
1. Green R, Bulloch B, Kabani A, Hancock BJ, Tenenbein M. Early analgesia for children with acute abdominal pain. Pediatrics.
2005;116:978 –983
doi:10.1542/peds.2005-2759
In Reply.—
I am grateful for the interest of Child and Haque in our work. They indicate surprise that only 162 patients were eligible for our study and highlight their concern by sharing that they “see⬃20 000 patients per year, and at least 1 patient per day will present with acute abdominal pain that merits a surgical consultation.” The phrases “⬃20 000 per year” and “at least 1 patient per day” convey the perception that these are anecdotal impres-sions rather than systematically collected data.
Nevertheless, it would seem that there are real ences between our institutions. Some of these differ-ences are accounted for by our scientific design. As stated in our methods section, ours was a convenience study that was conducted during 2 of every 3 days. Furthermore, children who presented between midnight and 8:00AMwere not studied, and children⬍5 years old
were excluded. In addition to these design criteria, I suspect that a proportion of the apparent difference is a result of different patient populations seen at our insti-tution in Canada and theirs in the United Kingdom. Ours is a university-affiliated children’s hospital that serves a midsized city. The majority of our emergency department patients present for primary care rather than being referred for secondary or tertiary care. Perhaps there is a higher proportion of referred patients at their institution. Another potential reason for this apparent
difference of rate of consultation is a difference of prac-tice between the emergency pediatricians in my hospital and the attending physicians in their hospital.
I performed an audit to quantify the number of con-sultations to surgeons for abdominal pain requested by our emergency department in 1 year; there were 180. After applying our exclusion criteria, I found that there would have been 80 patients eligible for enrollment, a number remarkably similar to the 81 per year that we reported in our study.
Child and Haque also raise the issue of an apparent low incidence of nonsurgical diagnoses. In our study, we found that 54% of our patients had appendicitis. In the absence of this statistic from their hospital, it is difficult to know if our experience is different. If indeed there are differences, they could be a consequence of differences in the practice between the physicians in the 2 institu-tions, as I have speculated.
We regard mesenteric adenitis and self-resolving ab-dominal pain as equivalent diagnoses. Both are diag-noses of exclusion that can only truly be made in retro-spect. Indeed, they are convenient euphemisms.
As stated in our methods section, confidence in diag-nosis was declared by self-report using a scale of 0% to 100%. Because the emergency pediatrician was asked the question before and after study-medication admin-istration, she or he served as her or his own control for both the intervention and control groups. For the sur-geons, this comparison was performed between that in-tervention and control groups after the administration of the study medication. Thus, we are assessing for both differences within and between groups of patients. Be-cause these differences were assessed within the frame-work of a randomized clinical trial, concerns regarding confounding factors and validated scoring tools are min-imized.
We accept and indeed acknowledged the issue of power. Our posthoc calculation found that “⬎1000 chil-dren per group would need to be enrolled in a random-ized, controlled trial to attain a power of 0.80.” This cannot be accomplished within a single institution. In-deed, it would require the cooperation of 20 institutions of similar size to ours over a period of 2 years to accom-plish this task. I believe that most would regard this to be a goal that cannot be accomplished. The $500 000 to $1 000 000 needed to fund such a study would be diffi-cult to justify. Thus, as always, the clinician is left with the task of having to make an informed decision using the best available data.
Milton Tenenbein, MD
Department of Pediatrics and Child Health University of Manitoba Children’s Hospital Winnipeg, Manitoba, Canada R3A 1S1
