Placebo Effects in Infants, Toddlers, and Parents

Letters

The survey-weighted arithmetic means of energy intakes were estimated by FFR type. Trends were tested using surveyweighted linear regression. The proportions of children who were FFR consumers were evaluated using survey-weighted logistic regression. Amounts of energy consumed per FFR eating occasion defined by meal name and time were evaluated to distinguish between the number of FFR eating events and the amount of energy consumed. All analyses used Stata 13 (StataCorp), accounted for the complex survey design, and were representative of the US population 4 to 19 years of age. Results | Panel A in Figure 1 shows population-wide trends in children’s mean energy intakes by FFR type. Energy intakes from burger, pizza, and chicken FFRs decreased significantly while energy intakes from other FFRs remained constant (P > .15 for others). Panel B in Figure 1 shows that the percentage of children consuming fast food on a given day dropped from 38.8% in 2003 to 2004 to 32.6% in 2009 to 2010 (P = .008). The proportion of children eating at burger restaurants remained stable (P = .35) and a modest drop was observed for chicken restaurants (P = .01). The observed decrease in energy from pizza restaurants may have been driven in part by a decrease in the number of consumers. While 12.2% of children obtained food/beverages from pizza restaurants in 2003 to 2004, only 6.4% did so in 2009 to 2010. The percentage consuming the other FFR types remained constant (P ≥ .29). Median energy consumption per eating occasion declined (Figure 2) except for chicken and sandwich FFRs. Discussion | Analyses of nationally representative data by FFR type compared with menus can provide insights into the contribution of fast foods to children’s diets. Publicly available data can complement data obtained from consumer panels, which are costly, inaccessible to many public health stakeholders, and may not be representative of the US population, limiting their value to inform policy. The present results were consistent with published sales reports. The decline in total pizza sales from 2003 to 2010 has been noted by industry sources.6 Burger and pizza restaurants accounted for much of the reduction in energy intakes. No fast food market segment experienced a significant increase in energy during the 8-year study. Analyses of populationbased National Health and Nutrition Examination Survey dietary intakes data separated by FFR market segment should allow researchers to focus on children and other populations and can also be extended to monitor consumption for other dietary constituents of concern, including sodium, added sugars, and solid fats. Colin D. Rehm, PhD, MPH Adam Drewnowski, PhD Author Affiliations: Center for Public Health Nutrition, Department of Epidemiology, University of Washington, Seattle. Corresponding Author: Colin D. Rehm, PhD, MPH, Center for Public Health Nutrition, Department of Epidemiology, University of Washington, PO Box 353410, Seattle, WA 98195 ([email protected]). Published Online: March 30, 2015. doi:10.1001/jamapediatrics.2015.38. Author Contributions: Dr Rehm had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. 504

Study concept and design: Both authors. Acquisition, analysis, or interpretation of data: Both authors. Drafting of the manuscript: Both authors. Critical revision of the manuscript for important intellectual content: Both authors. Statistical analysis: Rehm. Obtained funding: Drewnowski. Administrative, technical, or material support: Drewnowski. Study supervision: Drewnowski. Conflict of Interest Disclosures: Dr Drewnowski advises McDonald’s Corporation on global issues related to public health nutrition. No other disclosures were reported. Funding/Support: This study was funded by a research grant from McDonald’s Corporation to the University of Washington. The University of Washington has received grants, donations, and contracts from both the public and the private sector. Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. 1. Fryer CD, Ervin RB. Caloric intake from fast food among adults: United States, 2007-2010. NCHS Data Brief. 2013;(114):1-8. 2. Bauer KW, Hearst MO, Earnest AA, French SA, Oakes JM, Harnack LJ. Energy content of U.S. fast-food restaurant offerings: 14-year trends. Am J Prev Med. 2012;43(5):490-497. 3. Drewnowski A, Rehm CD. Energy intakes of US children and adults by food purchase location and by specific food source. Nutr J. 2013;12:59. 4. National Health and Nutrition Examination Survey. MEC in-person dietary interviewers procedure manual. http://www.cdc.gov/nchs/data/nhanes/nhanes _09_10/DietaryInterviewers_Inperson.pdf. Accessed October 15, 2014. 5. Rehm CD, Drewnowski A. A new method to monitor the contribution of fast food restaurants to the diets of US children. PLoS One. 2014;9(7):e103543. 6. Smith DP. The fall of pizza. QSR magazine website. http://www.qsrmagazine .com/competition/fall-pizza. Accessed August 30, 2013.

COMMENT & RESPONSE

Placebo Effects in Infants, Toddlers, and Parents To the Editor Paul and colleagues1 studied 120 infants and toddlers aged 2 to 47 months with acute cough. Infants and toddlers were randomized to pasteurized agave nectar mixed with natural grape flavoring and other bulking agents, natural grapeflavored water with caramel color (placebo), and a no treatment group.1 Significant symptom improvements were found for the agave nectar and placebo groups compared with the no treatment group, with no significant differences for any outcome between the agave nectar and placebo groups.1 The study was designed with dose volume stratified by age, randomization performed by a statistician not affiliated with the study, and investigator blinding maintained, giving each study participant an opaque syringe filled with agave nectar, placebo, or no treatment. Although the lack of treatment in 1 of 3 groups can be viewed as a source of parental disappointment and assessment bias, the no treatment group permitted the detection of a true placebo effect excluding clinical changes owing to the natural history of the child’s acute illness. Parents were asked to report symptom improvements through a set of specific questions raising the question of whether the placebo effect was on the child, parent, or both. Parental reports can be flawed and in conflict with physicians’ reports and their expectations can interfere with perception of subjective outcomes. Placebos and otherwise

JAMA Pediatrics May 2015 Volume 169, Number 5 (Reprinted)

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/05/2015

jamapediatrics.com

Letters

equivalent interventions can positively influence parents’ and infants’ behaviors leading to clinical outcome improvements. Such findings spark debates on at least 3 important aspects. First, under what circumstances is it desirable to include a no treatment group? While the inclusion of no treatment groups provides evidence about the presence of a placebo effect excluding other confounding factors (eg, regression to the mean or false positive), the applicability of such a design seems to be limited to a very restricted number of illnesses and diseases. Second, are placebo effects in childhood greater than in adulthood? Placebo effects are likely to be greater in childhood because of pronounced learning skills, naive interpretations of events, and the peculiar dynamic of child-parentphysician interactions.2 Third, is it ethical to introduce placebos in pediatrics? If placebo effects in pediatrics are robust and clinically relevant,1-4 concerns surrounding the intentional use of placebos can be adequately addressed though transparent shared decisionmaking processes in which placebo use is openly deliberated among the parties.5 Luana Colloca, MD, PhD Author Affiliations: Department of Pain and Translational Symptom Science, School of Nursing, University of Maryland, Baltimore; Department of Anesthesiology, School of Medicine, University of Maryland, Baltimore. Corresponding Author: Luana Colloca, MD, PhD, University of Maryland Baltimore, 655 W Lombard St, Room 729, Baltimore, MD 21201 (colloca@son .umaryland.edu). Conflict of Interest Disclosures: None reported. 1. Paul IM, Beiler JS, Vallati JR, Duda LM, King TS. Placebo effect in the treatment of acute cough in infants and toddlers: a randomized clinical trial. JAMA Pediatr. 2014;168(12):1107-1113. 2. Simmons K, Ortiz R, Kossowsky J, et al. Pain and placebo in pediatrics: a comprehensive review of laboratory and clinical findings. Pain. 2014;155(11): 2229-2235. 3. Weimer K, Gulewitsch MD, Schlarb AA, Schwille-Kiuntke J, Klosterhalfen S, Enck P. Placebo effects in children: a review. Pediatr Res. 2013;74(1):96-102. 4. Faria V, Linnman C, Lebel A, Borsook D. Harnessing the placebo effect in pediatric migraine clinic. J Pediatr. 2014;165(4):659-665. 5. Brody H, Colloca L, Miller FG. The placebo phenomenon: implications for the ethics of shared decision-making. J Gen Intern Med. 2012;27(6):739-742.

To the Editor The phenomenon that Paul et al1 describe in their study of acute cough, where active interventions and placebo showed no difference yet were both superior to the no treatment group,1 may be an example of what has been described previously as placebo by proxy.2 The idea that family members and physicians may have an emotional response to a patient’s treatment and perceived improvement on subjective complaints2 and that a patient’s response to therapy may be affected by the behavior of other people who know the patient is undergoing therapy3 seem to be especially prominent in pediatric conditions.2 This phenomenon is similar to the common occurrence in placebo-controlled randomized trials of animals (eg, lameness from osteoarthritis in dogs4 and headshaking in horses5), where owners and physicians detect significant improvements in jamapediatrics.com

both active and placebo interventions but objective measures show no change. Joe Kossowsky, PhD Ted J. Kaptchuk, OMD Author Affiliations: Department of Anesthesiology, Perioperative, and Pain Medicine, Boston Children’s Hospital, Harvard Medical School, Boston, Massachusetts (Kossowsky); Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts (Kaptchuk). Corresponding Author: Joe Kossowsky, PhD, Department of Anesthesiology, Perioperative, and Pain Medicine, Boston Children's Hospital, Harvard Medical School, 333 Longwood Ave, Pain Treatment Service (5th Floor), Boston, MA 02115 ([email protected]). Conflict of Interest Disclosures: None reported. 1. Paul IM, Beiler JS, Vallati JR, Duda LM, King TS. Placebo effect in the treatment of acute cough in infants and toddlers: a randomized clinical trial. JAMA Pediatr. 2014;168(12):1107-1113. 2. Grelotti DJ, Kaptchuk TJ. Placebo by proxy. BMJ. 2011;343:d4345. 3. Whalley B, Hyland ME. Placebo by proxy: the effect of parents’ beliefs on therapy for children’s temper tantrums. J Behav Med. 2013;36(4):341-346. 4. Conzemius MG, Evans RB. Caregiver placebo effect for dogs with lameness from osteoarthritis. J Am Vet Med Assoc. 2012;241(10):1314-1319. 5. Talbot WA, Pinchbeck GL, Knottenbelt DC, Graham H, McKane SA. A randomised, blinded, crossover study to assess the efficacy of a feed supplement in alleviating the clinical signs of headshaking in 32 horses. Equine Vet J. 2013;45(3):293-297.

In Reply We very much appreciate the complimentary letters in response to our article. The Kossowsky and Kaptchuk letter commented that the emotional response to a patient’s treatment could influence subjective complaints. Indeed, parents and health care professionals alike have a difficult time watching an uncomfortable child, perhaps even more so, an uncomfortable infant. Gassiness, perceived formula intolerance, and gastroesophageal reflux are other examples where an uncomfortable infant may have office visits outside of routine care that result in medical interventions that may not be supported by evidence and may not result in objective improvement. Alternatively, intervening commonly satisfies parental and, in many instances, health care professional instincts to help an uncomfortable child. In our study, it was unknown whether infants themselves had a response based on parent behavior as suggested in the text of the Kossowsky and Kaptchuk letter. The correspondence by Colloca raised several interesting points regarding our use of a no treatment group in the study and the use of placebos for children in general. In our trial, we felt justified including a no treatment group as we have done previously1,2 for 2 main reasons. First, outside of comfort measures, no treatment is commonly recommended for treatment of these illnesses, especially during infancy. Thus, a no treatment group that is allowed to use analgesics and hydration measures represents clinical practice. Second, any potential treatment administered, even those that might provide short-term symptomatic relief, was not expected to influence the natural history of these self-resolving shortlived illnesses. Regarding the Colloca comment that placebo effects might be greater in childhood than in adulthood, the study popula(Reprinted) JAMA Pediatrics May 2015 Volume 169, Number 5



505

Letters

tion of infants and the subjective short-term nocturnal outcomes described in our study likely make the proposed mechanisms described in the letter unlikely. Rather, the effect is more likely an effect on the parents as described in the editorial by Taylor and Opel3 that accompanied our article. Ian M. Paul, MD, MSc Author Affiliation: Division of Academic General Pediatrics, Department of Pediatrics, Penn State College of Medicine, Hershey, Pennsylvania. Corresponding Author: Ian M. Paul, MD, MSc, Department of Pediatrics, Penn State College of Medicine, 500 University Dr, Long Lane Bldg, Room 111, Mail Code HS83, Hershey, PA 17033-0850 ([email protected]). Conflict of Interest Disclosures: None reported. 1. Paul IM, Beiler J, McMonagle A, Shaffer ML, Duda L, Berlin CM Jr. Effect of honey, dextromethorphan, and no treatment on nocturnal cough and sleep quality for coughing children and their parents. Arch Pediatr Adolesc Med. 2007; 161(12):1140-1146. 2. Paul IM, Beiler JS, King TS, Clapp ER, Vallati J, Berlin CM Jr. Vapor rub, petrolatum, and no treatment for children with nocturnal cough and cold symptoms. Pediatrics. 2010;126(6):1092-1099. 3. Taylor JA, Opel DJ. Using the placebo effect to treat cold symptoms in children. JAMA Pediatr. 2014;168(12):1091-1092.

Folic Acid Supplements During Pregnancy in Specific Clinical Settings: What Do We Know About Epilepsy? To the Editor We read with great interest the article by ValeraGran and colleagues1 published in JAMA Pediatrics. ValeraGran et al1 performed a multicenter prospective study with a mother-child cohort to evaluate the use of high doses of folic acid (FA) during pregnancy and the effects on child psychomotor development after the first year of life. The mean psychomotor scale score in children whose mothers were exposed to higher FA doses (>5000 μg/d) was significantly lower than those whose mothers consumed recommended doses (400-1000 μg/d). Moreover, higher FA doses appeared to be associated with a high risk of delayed psychomotor development, although this was lacking statistical significance. ValeraGran and colleagues1 concluded that while standard FA doses should be maintained for the periconceptional period, inappropriately high FA intakes should be avoided owing to possible detrimental infant psychomotor effects. The authors excluded 9 women affected by epilepsy as a medical condition that could lead physicians to prescribe FA doses higher than 5000 μg/d. In clinical practice, women with epilepsy are commonly advised to take higher periconceptional FA doses compared with the general population. This practice is prescribed by several national guidelines (ie, the American Epilepsy Society, American Academy of Neurology, UK National Institute for Health and Care Excellence, Scottish Intercollegiate Guidelines Network, and Italian League Against Epilepsy) because neural tube defects and major congenital malformations have been previously associated with exposure to antiepileptic drugs in utero. Such malformations are mostly associated with exposure to certain antiepileptic drugs (ie, valproate sodium, carbamazepine, and phenobarbital sodium) and, although the underlying mechanisms are unclear, alterations of FA metabolism have been hypothesized.2 506

The UK Epilepsy and Pregnancy Register failed to find any correlation between FA intake and major congenital malformation, weakening the hypothesis that an increased risk of major congenital malformation is linked to FA metabolism.3 In addition, a Cochrane Review of neurodevelopmental outcomes in children born to epileptic women showed an IQ reduction in the group exposed to valproate for whom higher FA doses were often administered.4 High FA exposures during and after gestation in rats have been found to induce significant reductions of seizure threshold.5 Findings from Valera-Gran et al1 should be taken into account in epileptic women because major congenital malformation seems to be unrelated to FA intake3 and negative neurodevelopmental outcomes are more frequent with valproate administration4 commonly associated with higher FA supplements. Although extrapolations from the general population to special groups, such as epileptic women treated with antiepileptic drugs, can be questionable, high doses of FA seem to exceed what is needed to reduce malformation risks, even in mothers with epilepsy. Larger studies are required to clarify the matter. Andrea Romigi, MD, PhD Fabio Placidi, PhD Francesca Izzi, PhD Author Affiliations: Neurophysiopathology Unit, Department of Systems Medicine, Sleep and Epilepsy Medicine Centre, University of Rome Tor Vergata Hospital, Rome, Italy (Romigi, Placidi, Izzi); IRCCS Neuromed, Pozzilli, Italy (Romigi). Corresponding Author: Andrea Romigi, MD, PhD, University of Rome Tor Vergata Hospital, Department of Neurophysiopathology, Sleep and Epilepsy Centre, Viale Oxford 81, 00133 Rome, Italy ([email protected]). Conflict of Interest Disclosures: None reported. 1. Valera-Gran D, García de la Hera M, Navarrete-Muñoz EM, et al; Infancia y Medio Ambiente (INMA) Project. Folic acid supplements during pregnancy and child psychomotor development after the first year of life. JAMA Pediatr. 2014;168(11):e142611. 2. Harden CL, Pennell PB, Koppel BS, et al; American Academy of Neurology; American Epilepsy Society. Management issues for women with epilepsy: focus on pregnancy (an evidence-based review): III, vitamin K, folic acid, blood levels, and breast-feeding: report of the Quality Standards Subcommittee and Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Epilepsia. 2009;50 (5):1247-1255. 3. Morrow JI, Hunt SJ, Russell AJ, et al. Folic acid use and major congenital malformations in offspring of women with epilepsy: a prospective study from the UK Epilepsy and Pregnancy Register. J Neurol Neurosurg Psychiatry. 2009; 80(5):506-511. 4. Bromley R, Weston J, Adab N, et al. Treatment for epilepsy in pregnancy: neurodevelopmental outcomes in the child. Cochrane Database Syst Rev. 2014; 10:CD010236. 5. Girotto F, Scott L, Avchalumov Y, et al. High dose folic acid supplementation of rats alters synaptic transmission and seizure susceptibility in offspring. Sci Rep. 2013;3:1465.

In Reply We appreciate the interest of Romigi and colleagues in our article recently published in JAMA Pediatrics. They clarify the effect of folic acid (FA) supplement doses higher than those recommended during pregnancy, particularly among women taking medical treatments for epilepsy.

JAMA Pediatrics May 2015 Volume 169, Number 5 (Reprinted)



jamapediatrics.com