- Email: [email protected]
Assessing parents' attitudes towards ketogenic dietary therapies
Epilepsy & Behavior 39 (2014) 1–5
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Assessing parents' attitudes towards ketogenic dietary therapies Natasha E. Schoeler a,b, Lindsay MacDonald c, Helena Champion d, J. Helen Cross b,e, Josemir W. Sander a,f,g, Sanjay M. Sisodiya a,f,⁎, Rob Horne c,⁎⁎ a
NIHR University College London Hospitals Biomedical Research Centre, Department of Clinical and Experimental Epilepsy, UCL Institute of Neurology, London WC1N 3BG, UK UCL Institute of Child Health, Great Ormond Street Hospital for Children, London WC1N 1EH, UK Centre for Behavioural Medicine, Department of Practice and Policy, UCL School of Pharmacy, London WC1H 9JP, UK d Department of Nutrition & Dietetics, Cambridge University Hospitals NHS Foundation Trust, Cambridge Biomedical Campus, Cambridge CB2 0QQ, UK e Young Epilepsy, Lingfield RH7 6PW, UK f Epilepsy Society, Chalfont St Peter SL9 0RJ, UK g Stichting Epilepsie Instellingen Nederland (SEIN), Heemstede, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 23 June 2014 Revised 9 July 2014 Accepted 14 July 2014 Available online xxxx Keywords: Low carbohydrate High fat Beliefs Adherence Epilepsy Ketogenic diet
a b s t r a c t We aimed to assess and quantify parental beliefs regarding ketogenic dietary therapies (KDTs). We also aimed to determine whether beliefs were related to response to KDTs. Adapted versions of the Beliefs about Medicine Questionnaire were completed by parents of children following KDTs for epilepsy. Demographic and clinical data were collected from hospital records. Ketogenic dietary therapy response was defined as ≥50% seizure reduction compared to baseline. Many parents had a positive perception of KDTs and were convinced of the necessity of KDTs for their children, although beliefs were wide-ranging. Over half of parents reported concerns about the potential long-term effects of KDTs. Parental beliefs about KDTs were significantly correlated with patient response. This was an attempt to quantify parents' beliefs regarding the use of KDTs for their child's epilepsy. The questionnaire may be used to identify individuals with a less positive attitude towards KDTs and who may be less likely to report a favorable response to KDTs. It is unknown whether people who have positive beliefs about KDTs engage in less nonadherent behavior or whether beliefs regarding KDTs simply reflect outcomes. The evidence behind the long-term side effects of KDTs should be emphasized when counseling patients and their families. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Ketogenic dietary therapies (KDTs) can be an effective treatment for drug-resistant epilepsy [1,2]. Various types of ketogenic diets have been designed to improve palatability or reduce adverse side effects, and there are a range of products and resources available to help people following KDTs. However, adherence to a dietary regime may not always be easy. Nonadherence to treatment may be unintentional, for example due to forgetfulness or limited resources, or intentional, where people make a conscious decision not to follow an agreed treatment plan [3]. Nonadherence is multifaceted and depends also on physician-related Abbreviations: KDTs, ketogenic dietary therapies; BMQ, Beliefs about Medicine Questionnaire; BMQ-KDTs, Beliefs about Medicine Questionnaire adapted for use with ketogenic dietary therapies. ⁎ Correspondence to: S.M. Sisodiya, Box 29, Queen Square, London WC1N 3BG, UK. Tel.: +44 20 3448 8612; fax: +44 20 3448 8615. ⁎⁎ Correspondence to: R. Horne, UCL School of Pharmacy, Mezzanine Floor, BMA House, Tavistock Square, London WC1H 9JP, UK. Tel.: +44 20 7874 1281; fax: +44 20 7387 5693. E-mail addresses: [email protected] (S.M. Sisodiya), [email protected] (R. Horne).
http://dx.doi.org/10.1016/j.yebeh.2014.07.020 1525-5050/© 2014 Elsevier Inc. All rights reserved.
and health-care system-related factors, but, ultimately, it is the patient who decides whether or not to adhere to treatment [4]. A key factor in understanding intentional nonadherence to treatment is the individual’s beliefs about treatment — specifically their perceived personal need for treatment and concerns or worries about treatment. This relationship is described in the Necessity–Concerns Framework and operationalized with the Beliefs about Medicine Questionnaire (BMQ) [5]. These necessity and concern beliefs are related to adherence to antiepileptic drugs (AEDs) and medicines prescribed for other long-term conditions [6,7]. In clinical practice, adherence to KDTs is generally assessed by monitoring blood and urine ketone levels. These measures are imperfect and do not consistently correlate with seizure reduction [8]. Previous studies have used biochemical measures [9] or personal communication [10] to assess adherence to KDTs. Accurate knowledge of food consumption may be obtained by patient surveillance and weighing of meals [11], but this would not be feasible on a large-scale outpatient basis. Previous studies have examined reasons for starting KDTs [12] and parental expectations and factors influencing diet duration [13], but no attempts have been made to formally quantify beliefs regarding KDTs using the Necessity–Concerns Framework.
2
N.E. Schoeler et al. / Epilepsy & Behavior 39 (2014) 1–5
We assessed parental beliefs regarding KDTs with an adapted BMQ. We also aimed to determine whether beliefs were related to response to KDTs. 2. Material and methods This project was reviewed and registered as an audit or service evaluation at each of the study sites: Great Ormond Street Hospital, Evelina London Children's Hospital, Young Epilepsy (Charity for Children and Young People with Epilepsy, http://youngepilepsy.org. uk/) in conjunction with Matthew's Friends charity (Dietary Treatments for Epilepsy: Information–Training–Research–Support, http://www. matthewsfriends.org/), Birmingham Children's Hospital, Cambridge University Hospital, and St George's Hospital. 2.1. Adaptation of the BMQ A version of the BMQ used to assess people's beliefs regarding AEDs and levels of adherence at our specialist NHS inpatient evaluation facility (unpublished data) was adapted for use by parents with children following KDTs. For example, ‘My tablets make my seizures better’ was changed to ‘The diet makes my child's seizures better’. The BMQ-KDTs (copyright, Prof. Rob Horne) consisted of a 4-item Necessity scale, which assessed parents' perceptions of the need for their child to follow KDTs, and a 7-item Concerns scale, which assessed their concerns about the potential adverse effects of following KDTs. All BMQ-KDT items were scored on a five-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = uncertain, 4 = agree, and 5 = strongly agree). Individual item scores were summed to give a Necessity and a Concerns score for each individual. Mean Necessity and Concerns scores were calculated by dividing the overall score for each individual by the number of items in the scale. A copy of the BMQ-KDTs is given in Supplementary material.
A Necessity–Concerns differential was calculated by subtracting Concerns scores from Necessity scores. A negative differential signifies that the individual rates his concerns about KDTs higher than his beliefs about the necessity of following KDTs; a positive differential signifies that the individual's belief in the necessity of following KDTs is stronger than his concerns about potential adverse effects from KDTs. Participants were categorized into attitudinal groups based on their beliefs about KDTs: ‘accepting’ of KDTs (high necessity, low concerns), ‘ambivalent’ (high necessity, high concerns), ‘skeptical’ (low necessity, high concerns), or ‘indifferent’ (low necessity, low concerns). Necessity and Concerns scores were classified as high or low by dichotomizing at the midpoint for each subscale: low necessity ≤3 and low concerns b3. One participant that had missing responses to many BMQ-KDT items could not be categorized. T-tests were used to explore putative correlations between mean BMQ-KDT scores and Necessity–Concerns differentials, with KDT response. A chi-square test was used to explore correlations between attitudinal groups and KDT response. 3. Results 3.1. Cohort demographics and KDT response
A cross-sectional design was adopted. Parents of children following KDTs for their epilepsy were asked to complete the BMQ-KDTs when attending specialist clinics at the six study sites. The minimum duration of KDTs was three months, as this is customarily the first time point at which response is assessed in the clinic. All children were participants in a previous ethically-approved research project, for which informed consent was obtained. All parents accepted the invitation to participate in this audit and complete the questionnaires.
Parents of 92 children completed the BMQ-KDTs. Of the cohort (n = 92), 47 (51%) children were female and 45 (49%) were male. Seven children had Dravet syndrome, 8 had epilepsy with myoclonic seizures (either myoclonic–atonic seizures and/or myoclonic absences or myoclonic epilepsy unspecified), 6 had West syndrome, 1 had juvenile absence epilepsy, 2 had late infantile Batten disease, 1 had Lennox–Gastaut syndrome, and 67 had no diagnosed epilepsy syndrome. The average age at KDT initiation was 5.7 years (range: 0.8– 16.2 years). At the time of completing the questionnaires, 66 (72%) children were following the classical ketogenic diet, 15 (16%) were following the medium-chain triglyceride diet, and 11 (12%) were following the modified ketogenic diet. The average duration of KDTs at the time that the questionnaires were completed was 14.8 months (range: 3–82 months). Fifty-seven (62%) individuals were classified as responders at the time that the questionnaires were completed; 29 (31.5%) were nonresponders; and KDT response was unknown for 6 (6.5%) individuals, as seizure frequency could not be accurately determined from hospital records at that particular time point.
2.3. Demographic and clinical data
3.2. Sensitivity analyses
Demographic and KDT response data had already been collected from hospital records and dietetic notes as part of the previous project. Baseline seizure frequency was determined from the last clinic letter prior to starting KDTs. Seizure frequency at the time that the BMQ was completed was determined from the clinic letter during the corresponding follow-up appointment. If an individual achieved ≥ 50% reduction in seizure frequency compared to baseline (as reported by parents in clinic and relayed in clinic letters), he/she was classified as a responder.
The internal consistency of the BMQ-KDT scales was deemed acceptable despite modifications to the original items. There are varying reports regarding what constitutes an acceptable alpha: values over 0.7 are often considered acceptable [14], although values below 0.7 may be appropriate, depending on what is being measured [15]. The BMQ-KDT—Necessity scale had an alpha of 0.93 and the BMQKDT—Concerns scale an alpha of 0.68. Removal of specific items did not substantially improve the alphas.
2.2. Participants
3.3. Beliefs regarding KDTs 2.4. Data analysis Statistical analysis was undertaken using IBM SPSS Statistics v22. Descriptive statistics were used to describe the sample and illustrate frequencies, mean, and distribution of responses to the BMQ-KDT scales. The internal consistency of the BMQ-KDT scales was tested using Cronbach's alpha. As several adaptations were made to the original items, it was necessary to check whether these items still adhered to the concept of the scale.
The mean necessity score for the cohort was 3.77 [standard deviation: 0.92], revealing that there was strong belief in the necessity of KDTs. Most parents indicated that it was necessary for their children to follow KDTs, although the proportion of parents strongly endorsing the necessity of KDTs varied between individual items (see Fig. 1). The mean Concerns score for the sample was 2.59 (standard deviation: 0.54). Examination of the individual Concerns items in the
N.E. Schoeler et al. / Epilepsy & Behavior 39 (2014) 1–5
Fig. 1. Percentage of parents expressing a view that it is necessary for their child to follow ketogenic dietary therapies, as indicated by responding ‘strongly agree’ or ‘agree’ to BMQKDT—Necessity items.
BMQ-KDTs showed that there was one overriding concern: over half (58%) of the parents expressed strong concerns regarding potential long-term effects of KDTs (see Fig. 2). The exact nature of these concerns was not documented as part of this study. Seventy-nine percent of the parents scored a positive Necessity– Concerns differential (mean differential = 1.18, standard deviation = 1.23), indicating that their beliefs regarding the necessity of following KDTs were stronger than their concerns regarding potential adverse effects from KDTs. Over half (55%) of the participants were found to be ‘accepting’ of KDTs (see Fig. 3). The rest of the cohort was approximately equally divided into those who were ‘ambivalent’ (17%), ‘skeptical’ (14%), or ‘indifferent’ (14%) towards KDTs.
3
Fig. 3. Attitudinal analysis of beliefs about ketogenic dietary therapies.
were lower (t(83) = −2.88; p = 0.005) (Fig. 4B) among those categorized as KDT responders. The Necessity–Concerns differential was higher in KDT responders than in nonresponders (t(83) = 5.57; p b 0.000), indicating a more negative perception of treatment in nonresponders. Attitudinal group was also significantly related to KDT response (p = 0.004). Those who were ‘accepting’ of KDTs had the highest proportion of responders (n = 37) compared to nonresponders (n = 9); the ‘indifferent’ and ‘skeptical’ groups had the lowest proportion of responders compared to nonresponders (4 responders and 7 nonresponders in the ‘indifferent’ group and 5 responders and 8 nonresponders in the ‘skeptical’ group). 4. Discussion
3.4. Relationship between beliefs and response to KDTs There was a significant relationship between parents' beliefs about KDTs and their children's response to KDTs. Necessity beliefs were higher (t(83) = 5.58; p b 0.000) (Fig. 4A) and concerns about KDTs
Fig. 2. Percentage of parents expressing concerns about their child following ketogenic dietary therapies, as indicated by responding ‘strongly agree’ or ‘agree’ to BMQ-KDT— Concerns items.
This is the first study, to our knowledge, to assess parents' beliefs regarding the use of KDTs for their child's epilepsy using the BMQ-KDTs. Many parents had a positive perception of KDTs and were convinced of the necessity of KDTs for their children, although others had doubts regarding the necessity of this treatment. Many reported concerns about the potential long-term effects of KDTs. Parental beliefs about KDTs were significantly correlated with patient response. The fact that long-term side effects were the most prevalent concern about KDTs is rational for individuals who have been following dietary treatment for long periods of time. Long-term effects of KDTs are outlined to parents and/or patients when they are considering KDTs, but this issue may deserve more focus during initial consultations in order to clarify disconnects between facts and beliefs. Poor linear growth, nephrolithiasis, fractures, and dyslipidemia have been reported by 82%, 25%, and 21% of 28 individuals following KDTs for N6 years [16], but other studies have reported such consequences to be much rarer or less severe than previously thought [17–19]. For example, although there is evidence for a negative effect on height and/or weight from KDTs, particularly when treatment is started at a young age [17, 20–24], some studies have reported little or no effect on growth parameters [17,25,26] and catch-up growth has been observed in children when KDTs are discontinued [27]. Lipid levels tend to return to baseline levels over a period of two years [16,28], can return to normal when the KDT is discontinued [9,20,29], or may even be improved by KDTs [30–32]. The most common adverse effects of KDTs, such as constipation, abdominal pain, or anorexia, tend to be nonsevere, transient, and managed with medication or dietary manipulation. Even so, this study has highlighted that long-term side effects are important not only to health professionals but also to parents. As with any treatment, the
4
N.E. Schoeler et al. / Epilepsy & Behavior 39 (2014) 1–5
Fig. 4. A — Mean [95% confidence interval] BMQ-KDT—Necessity score in responders and nonresponders. B — Mean [95% confidence interval] BMQ-KDT—Concerns score in responders and nonresponders.
benefits obtained from KDTs are constantly weighed up against the difficulties of following the diet and (potential) adverse effects. We showed that parental beliefs regarding KDTs were correlated with their child's response to KDTs. This may be because people who have positive beliefs about KDTs engage in less nonadherent behavior, which means that their children are more likely to respond, although this cannot be definitively established from this study. Improving adherence to treatment may have a greater effect than improvements in specific medical therapies [4], so it is important to identify people who are likely to have poor adherence. This is particularly relevant for KDTs, where blood and urine ketones, which are subject to large inter- and intra-individual variability and are affected by many different factors, are commonly used to indicate treatment adherence. There are important limitations to this study. This study is observational and cross-sectional, so it may be that those people who have had a good experience with KDTs (which may have nothing to do with their adherent or nonadherent behavior) are more likely to have a positive attitude towards the treatment. This needs exploration in a prospective study. Another limitation of this study is ascertainment bias: many questionnaires were completed by parents of children who had been following KDTs for a long time and parents with children who discontinued KDTs before the 3-month point due to, for example, increased seizure frequency or poor tolerability, and would not have had an opportunity to complete the BMQ-KDTs. Thus, although most parents' views regarding the necessity of KDTs outweighed any concerns regarding the potential harm of KDTs, the proportion of people who viewed the treatment as positive would probably be higher than if attitudes and behavior had been evaluated prospectively. The BMQ-KDTs would be of greater value if completed before starting KDTs. A further limitation is that we did not have a behavioral measure of adherence to KDTs. We used parental reports of efficacy of the diet in reducing seizure frequency as an indicator of an adherence-dependent outcome. We attempted to adapt the Medication Adherence Report Scale for use as a behavioral measure of adherence by substituting references to medication to KDTs. However, scores on the adapted scale were not related to perceptions of KDTs or to reports of seizure frequency. The moderate Cronbach's alpha correlation coefficient (for BMQKDT—Concerns) may have been affected by the small number of items in the questionnaire, the positive distribution of scores, or the small
sample size. A larger sample size would help approximate the nonnormal distribution towards normality and reduce the negative bias imposed on the alpha value [33]. This has been found to reduce bias to a greater extent than when the number of items/test length was increased. It would be of interest to see whether results differ when selfreported (as opposed to parental) beliefs are assessed, although only a small number of children following KDTs may be able to answer the BMQ-KDTs themselves. Collecting these data for adults who are following KDTs and perhaps comparing answers from people following different types of KDTs are areas for future study. 5. Conclusions In conclusion, we have found the BMQ-KDTs to be a useful tool to quantify parents' beliefs regarding the use of KDTs for their child's epilepsy and to highlight common areas of concern. Although this study has not assessed the relationship between beliefs and adherence, it is likely that the relationship between treatment beliefs and clinical outcome (seizure control) is mediated, at least in part, by adherence. According to the Necessity–Concerns Framework, a better understanding of individuals' perceptions of their treatment can help to identify those at risk of nonadherence. Interventions may then be tailored to the needs of each individual by addressing salient perceptual barriers to adherence. Further work is needed to develop KDT-specific items to assess adherence to KDTs. Acknowledgments NS is supported by a UCL Impact Studentship in conjunction with Epilepsy Society. JWS is supported by the Dr. Marvin Weil Epilepsy Research Fund. RH is supported by an NIHR Senior Investigator Award. The authors acknowledge Dr. Christin Eltze, Marian Sewell, Carla Fitzachary, and Georgiana Fitzsimmons from Great Ormond Street Hospital; Dr. Ruth Williams, Mary-Anne Leung, Tara Randall, Layla Alford, Amanda Tomalin, and Dr. Elaine Hughes from Evelina London Children's Hospital; Emma Williams, Elizabeth Neal, Julie Edwards, Teresa Stein, and Dr. Archana Desurkar from Matthew's Friends; Angela Mensah from Young Epilepsy; Dr. Sunny Philip, Bernie Concannon, and Vanessa Hopkins from Birmingham Children's Hospital; Dr. Alasdair Parker and Dr. Anna Maw from Cambridge University Hospital; and
N.E. Schoeler et al. / Epilepsy & Behavior 39 (2014) 1–5
Dr. Penny Fallon, Orla Stone, and Nicole Dos Santos from St George's Hospital. Disclosure of conflicts of interest Professor JH Cross has cosupervised a PhD student funded by Vitaflo. The remaining authors have no conflicts of interest in relation to this work. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2014.07.020. References [1] Levy R, Cooper P, Giri P. Ketogenic diet and other dietary treatments for epilepsy (review). Cochrane Database Syst Rev 2012;3:CD001903. [2] Neal EG, Chaffe H, Schwartz RH, Lawson MS, Edwards N, Fitzsimmons G, et al. The ketogenic diet for the treatment of childhood epilepsy: a randomised controlled trial. Lancet Neurol 2008;7:500–6. [3] Cerveny P, Bortlik M, Kubena A, Vlcek J, Lakatos PL, Lukas M. Nonadherence in inflammatory bowel disease: results of factor analysis. Inflamm Bowel Dis 2007; 13:1244–9. [4] Brown MT, Bussell JK. Medication adherence: WHO cares? Mayo Clin Proc 2011; 86:304–14. [5] Horne R, Weinman J, Hankins M. The beliefs about medicines questionnaire: the development and evaluation of a new method for assessing the cognitive representation of medication. Psychol Health 1999;14:1–24. [6] Chapman SC, Horne R, Chater A, Hukins D, Smithson WH. Patients' perspectives on antiepileptic medication: relationships between beliefs about medicines and adherence among patients with epilepsy in UK primary care. Epilepsy Behav 2014; 31:312–20. [7] Horne R, Chapman SC, Parham R, Freemantle N, Forbes A, Cooper V. Understanding patients' adherence-related beliefs about medicines prescribed for long-term conditions: a meta-analytic review of the Necessity–Concerns Framework. PLoS One 2013;8:e80633. [8] Schoeler NE, Cross JH, Sander JW, Sisodiya SM. Can we predict a favourable response to Ketogenic Diet Therapies for drug-resistant epilepsy? Epilepsy Res 2013;106:1–16. [9] Klein P, Janousek J, Barber A, Weissberger R. Ketogenic diet treatment in adults with refractory epilepsy. Epilepsy Behav 2010;19:575–9. [10] Coppola G, Veggiotti P, Cusmai R, Bertoli S, Cardinali S, Dionisi-vici C, et al. The ketogenic diet in children, adolescents and young adults with refractory epilepsy: an Italian multicentric experience. Epilepsy Res 2002;48:221–7. [11] Amari A, Grace NC, Fisher WW. Achieving and maintaining compliance with the ketogenic diet. J Appl Behav Anal 1995;28:341–2.
5
[12] Kossoff EH, Doerrer SS, Turner Z. How do parents find out about the ketogenic diet? Epilepsy Behav 2012;24:445–8. [13] Farasat S, Kossoff EH, Pillas DJ, Rubenstein JE, Vining EP, Freeman JM. The importance of parental expectations of cognitive improvement for their children with epilepsy prior to starting the ketogenic diet. Epilepsy Behav 2006;8:406–10. [14] Tavakol M, Dennick R. Making sense of Cronbach's alpha. Int J Med Educ 2011; 2:53–5. [15] Kline P. The handbook of psychological testing. 2nd ed. London: Routledge; 1999. [16] Groesbeck D, Bluml R, Kossoff EH. Long‐term use of the ketogenic diet in the treatment of epilepsy. Dev Med Child Neurol 2006;48:978–81. [17] Liu Y-MC, Williams S, Basualdo-Hammond C, Stephens D, Curtis R. A prospective study: growth and nutritional status of children treated with the ketogenic diet. J Am Diet Assoc 2003;103:707–12. [18] Furth SL, Casey JC, Pyzik PL, Neu AM, Docimo SG, Vining EP, et al. Risk factors for urolithiasis in children on the ketogenic diet. Pediatr Nephrol 2000;15:125–8. [19] Kang HC, Chung DE, Kim DW, Kim HD. Early- and late-onset complications of the ketogenic diet for intractable epilepsy. Epilepsia 2004;45:1116–23. [20] Patel A, Pyzik PL, Turner Z, Rubenstein JE, Kossoff EH. Long-term outcomes of children treated with the ketogenic diet in the past. Epilepsia 2010;51:1277–82. [21] Vining EPG, Pyzik P, McGrogan J, Hladky H, Anand A, Kriegler S, et al. Growth of children on the ketogenic diet. Dev Med Child Neurol 2002;44:796–802. [22] Williams S, Basualdo-Hammond C, Curtis R, Schuller R. Growth retardation in children with epilepsy on the ketogenic diet: a retrospective chart review. J Am Diet Assoc 2002:405–7. [23] Neal EG, Chaffe HM, Edwards N, Lawson MS, Schwartz RH, Cross JH. Growth of children on classical and medium-chain triglyceride ketogenic diets. Pediatrics 2008;122:e334–40. [24] Peterson SJ, Tangney CC, Pimentel-Zablah EM, Hjelmgren B, Booth G, Berry-Kravis E. Changes in growth and seizure reduction in children on the ketogenic diet as a treatment for intractable epilepsy. J Am Diet Assoc 2005;105:718–25. [25] Nordli DR, Kuroda MM, Carroll J, Koenigsberger DY, Hirsch LJ, Bruner HJ, et al. Experience with the ketogenic diet in infants. Pediatrics 2001;108:129–33. [26] Couch S, Schwarzman F. Growth and nutritional outcomes of children treated with the ketogenic diet. J Am Diet Assoc 1999;99:1573–5. [27] Kim JT, Kang HC, Song JE, Lee MJ, Lee YJ, Lee EJ, et al. Catch-up growth after long-term implementation and weaning from ketogenic diet in pediatric epileptic patients. Clin Nutr 2013;32:98–103. [28] Kwiterovich Jr PO, Vining EP, Pyzik P, Skolasky Jr R, Freeman JM. Effect of a high-fat ketogenic diet on plasma levels of lipids, lipoproteins, and apolipoproteins in children. JAMA 2003;290:912–20. [29] Mosek A, Natour H, Neufeld MY, Shiff Y, Vaisman N. Ketogenic diet treatment in adults with refractory epilepsy: a prospective pilot study. Seizure 2009;18:30–3. [30] Liu YM, Lowe H, Zak MM, Kobayashi J, Chan VW, Donner EJ. Can children with hyperlipidemia receive ketogenic diet for medication-resistant epilepsy? J Child Neurol 2013;28:479–83. [31] Sharman MJ, Kraemer WJ, Love DM, Avery NG, Gomez AL, Scheett TP, et al. A ketogenic diet favorably affects serum biomarkers for cardiovascular disease in normal-weight men. J Nutr 2002;132:1879–85. [32] Paoli A, Cenci L, Grimaldi KA. Effect of ketogenic Mediterranean diet with phytoextracts and low carbohydrates/high-protein meals on weight, cardiovascular risk factors, body composition and diet compliance in Italian council employees. Nutr J 2011;10:112. [33] Sheng Y, Sheng Z. Is coefficient alpha robust to non-normal data? Front Psychol 2012;3:34.
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Assessing parents' attitudes towards ketogenic dietary therapies Natasha E. Schoeler a,b, Lindsay MacDonald c, Helena Champion d, J. Helen Cross b,e, Josemir W. Sander a,f,g, Sanjay M. Sisodiya a,f,⁎, Rob Horne c,⁎⁎ a
NIHR University College London Hospitals Biomedical Research Centre, Department of Clinical and Experimental Epilepsy, UCL Institute of Neurology, London WC1N 3BG, UK UCL Institute of Child Health, Great Ormond Street Hospital for Children, London WC1N 1EH, UK Centre for Behavioural Medicine, Department of Practice and Policy, UCL School of Pharmacy, London WC1H 9JP, UK d Department of Nutrition & Dietetics, Cambridge University Hospitals NHS Foundation Trust, Cambridge Biomedical Campus, Cambridge CB2 0QQ, UK e Young Epilepsy, Lingfield RH7 6PW, UK f Epilepsy Society, Chalfont St Peter SL9 0RJ, UK g Stichting Epilepsie Instellingen Nederland (SEIN), Heemstede, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 23 June 2014 Revised 9 July 2014 Accepted 14 July 2014 Available online xxxx Keywords: Low carbohydrate High fat Beliefs Adherence Epilepsy Ketogenic diet
a b s t r a c t We aimed to assess and quantify parental beliefs regarding ketogenic dietary therapies (KDTs). We also aimed to determine whether beliefs were related to response to KDTs. Adapted versions of the Beliefs about Medicine Questionnaire were completed by parents of children following KDTs for epilepsy. Demographic and clinical data were collected from hospital records. Ketogenic dietary therapy response was defined as ≥50% seizure reduction compared to baseline. Many parents had a positive perception of KDTs and were convinced of the necessity of KDTs for their children, although beliefs were wide-ranging. Over half of parents reported concerns about the potential long-term effects of KDTs. Parental beliefs about KDTs were significantly correlated with patient response. This was an attempt to quantify parents' beliefs regarding the use of KDTs for their child's epilepsy. The questionnaire may be used to identify individuals with a less positive attitude towards KDTs and who may be less likely to report a favorable response to KDTs. It is unknown whether people who have positive beliefs about KDTs engage in less nonadherent behavior or whether beliefs regarding KDTs simply reflect outcomes. The evidence behind the long-term side effects of KDTs should be emphasized when counseling patients and their families. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Ketogenic dietary therapies (KDTs) can be an effective treatment for drug-resistant epilepsy [1,2]. Various types of ketogenic diets have been designed to improve palatability or reduce adverse side effects, and there are a range of products and resources available to help people following KDTs. However, adherence to a dietary regime may not always be easy. Nonadherence to treatment may be unintentional, for example due to forgetfulness or limited resources, or intentional, where people make a conscious decision not to follow an agreed treatment plan [3]. Nonadherence is multifaceted and depends also on physician-related Abbreviations: KDTs, ketogenic dietary therapies; BMQ, Beliefs about Medicine Questionnaire; BMQ-KDTs, Beliefs about Medicine Questionnaire adapted for use with ketogenic dietary therapies. ⁎ Correspondence to: S.M. Sisodiya, Box 29, Queen Square, London WC1N 3BG, UK. Tel.: +44 20 3448 8612; fax: +44 20 3448 8615. ⁎⁎ Correspondence to: R. Horne, UCL School of Pharmacy, Mezzanine Floor, BMA House, Tavistock Square, London WC1H 9JP, UK. Tel.: +44 20 7874 1281; fax: +44 20 7387 5693. E-mail addresses: [email protected] (S.M. Sisodiya), [email protected] (R. Horne).
http://dx.doi.org/10.1016/j.yebeh.2014.07.020 1525-5050/© 2014 Elsevier Inc. All rights reserved.
and health-care system-related factors, but, ultimately, it is the patient who decides whether or not to adhere to treatment [4]. A key factor in understanding intentional nonadherence to treatment is the individual’s beliefs about treatment — specifically their perceived personal need for treatment and concerns or worries about treatment. This relationship is described in the Necessity–Concerns Framework and operationalized with the Beliefs about Medicine Questionnaire (BMQ) [5]. These necessity and concern beliefs are related to adherence to antiepileptic drugs (AEDs) and medicines prescribed for other long-term conditions [6,7]. In clinical practice, adherence to KDTs is generally assessed by monitoring blood and urine ketone levels. These measures are imperfect and do not consistently correlate with seizure reduction [8]. Previous studies have used biochemical measures [9] or personal communication [10] to assess adherence to KDTs. Accurate knowledge of food consumption may be obtained by patient surveillance and weighing of meals [11], but this would not be feasible on a large-scale outpatient basis. Previous studies have examined reasons for starting KDTs [12] and parental expectations and factors influencing diet duration [13], but no attempts have been made to formally quantify beliefs regarding KDTs using the Necessity–Concerns Framework.
2
N.E. Schoeler et al. / Epilepsy & Behavior 39 (2014) 1–5
We assessed parental beliefs regarding KDTs with an adapted BMQ. We also aimed to determine whether beliefs were related to response to KDTs. 2. Material and methods This project was reviewed and registered as an audit or service evaluation at each of the study sites: Great Ormond Street Hospital, Evelina London Children's Hospital, Young Epilepsy (Charity for Children and Young People with Epilepsy, http://youngepilepsy.org. uk/) in conjunction with Matthew's Friends charity (Dietary Treatments for Epilepsy: Information–Training–Research–Support, http://www. matthewsfriends.org/), Birmingham Children's Hospital, Cambridge University Hospital, and St George's Hospital. 2.1. Adaptation of the BMQ A version of the BMQ used to assess people's beliefs regarding AEDs and levels of adherence at our specialist NHS inpatient evaluation facility (unpublished data) was adapted for use by parents with children following KDTs. For example, ‘My tablets make my seizures better’ was changed to ‘The diet makes my child's seizures better’. The BMQ-KDTs (copyright, Prof. Rob Horne) consisted of a 4-item Necessity scale, which assessed parents' perceptions of the need for their child to follow KDTs, and a 7-item Concerns scale, which assessed their concerns about the potential adverse effects of following KDTs. All BMQ-KDT items were scored on a five-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = uncertain, 4 = agree, and 5 = strongly agree). Individual item scores were summed to give a Necessity and a Concerns score for each individual. Mean Necessity and Concerns scores were calculated by dividing the overall score for each individual by the number of items in the scale. A copy of the BMQ-KDTs is given in Supplementary material.
A Necessity–Concerns differential was calculated by subtracting Concerns scores from Necessity scores. A negative differential signifies that the individual rates his concerns about KDTs higher than his beliefs about the necessity of following KDTs; a positive differential signifies that the individual's belief in the necessity of following KDTs is stronger than his concerns about potential adverse effects from KDTs. Participants were categorized into attitudinal groups based on their beliefs about KDTs: ‘accepting’ of KDTs (high necessity, low concerns), ‘ambivalent’ (high necessity, high concerns), ‘skeptical’ (low necessity, high concerns), or ‘indifferent’ (low necessity, low concerns). Necessity and Concerns scores were classified as high or low by dichotomizing at the midpoint for each subscale: low necessity ≤3 and low concerns b3. One participant that had missing responses to many BMQ-KDT items could not be categorized. T-tests were used to explore putative correlations between mean BMQ-KDT scores and Necessity–Concerns differentials, with KDT response. A chi-square test was used to explore correlations between attitudinal groups and KDT response. 3. Results 3.1. Cohort demographics and KDT response
A cross-sectional design was adopted. Parents of children following KDTs for their epilepsy were asked to complete the BMQ-KDTs when attending specialist clinics at the six study sites. The minimum duration of KDTs was three months, as this is customarily the first time point at which response is assessed in the clinic. All children were participants in a previous ethically-approved research project, for which informed consent was obtained. All parents accepted the invitation to participate in this audit and complete the questionnaires.
Parents of 92 children completed the BMQ-KDTs. Of the cohort (n = 92), 47 (51%) children were female and 45 (49%) were male. Seven children had Dravet syndrome, 8 had epilepsy with myoclonic seizures (either myoclonic–atonic seizures and/or myoclonic absences or myoclonic epilepsy unspecified), 6 had West syndrome, 1 had juvenile absence epilepsy, 2 had late infantile Batten disease, 1 had Lennox–Gastaut syndrome, and 67 had no diagnosed epilepsy syndrome. The average age at KDT initiation was 5.7 years (range: 0.8– 16.2 years). At the time of completing the questionnaires, 66 (72%) children were following the classical ketogenic diet, 15 (16%) were following the medium-chain triglyceride diet, and 11 (12%) were following the modified ketogenic diet. The average duration of KDTs at the time that the questionnaires were completed was 14.8 months (range: 3–82 months). Fifty-seven (62%) individuals were classified as responders at the time that the questionnaires were completed; 29 (31.5%) were nonresponders; and KDT response was unknown for 6 (6.5%) individuals, as seizure frequency could not be accurately determined from hospital records at that particular time point.
2.3. Demographic and clinical data
3.2. Sensitivity analyses
Demographic and KDT response data had already been collected from hospital records and dietetic notes as part of the previous project. Baseline seizure frequency was determined from the last clinic letter prior to starting KDTs. Seizure frequency at the time that the BMQ was completed was determined from the clinic letter during the corresponding follow-up appointment. If an individual achieved ≥ 50% reduction in seizure frequency compared to baseline (as reported by parents in clinic and relayed in clinic letters), he/she was classified as a responder.
The internal consistency of the BMQ-KDT scales was deemed acceptable despite modifications to the original items. There are varying reports regarding what constitutes an acceptable alpha: values over 0.7 are often considered acceptable [14], although values below 0.7 may be appropriate, depending on what is being measured [15]. The BMQ-KDT—Necessity scale had an alpha of 0.93 and the BMQKDT—Concerns scale an alpha of 0.68. Removal of specific items did not substantially improve the alphas.
2.2. Participants
3.3. Beliefs regarding KDTs 2.4. Data analysis Statistical analysis was undertaken using IBM SPSS Statistics v22. Descriptive statistics were used to describe the sample and illustrate frequencies, mean, and distribution of responses to the BMQ-KDT scales. The internal consistency of the BMQ-KDT scales was tested using Cronbach's alpha. As several adaptations were made to the original items, it was necessary to check whether these items still adhered to the concept of the scale.
The mean necessity score for the cohort was 3.77 [standard deviation: 0.92], revealing that there was strong belief in the necessity of KDTs. Most parents indicated that it was necessary for their children to follow KDTs, although the proportion of parents strongly endorsing the necessity of KDTs varied between individual items (see Fig. 1). The mean Concerns score for the sample was 2.59 (standard deviation: 0.54). Examination of the individual Concerns items in the
N.E. Schoeler et al. / Epilepsy & Behavior 39 (2014) 1–5
Fig. 1. Percentage of parents expressing a view that it is necessary for their child to follow ketogenic dietary therapies, as indicated by responding ‘strongly agree’ or ‘agree’ to BMQKDT—Necessity items.
BMQ-KDTs showed that there was one overriding concern: over half (58%) of the parents expressed strong concerns regarding potential long-term effects of KDTs (see Fig. 2). The exact nature of these concerns was not documented as part of this study. Seventy-nine percent of the parents scored a positive Necessity– Concerns differential (mean differential = 1.18, standard deviation = 1.23), indicating that their beliefs regarding the necessity of following KDTs were stronger than their concerns regarding potential adverse effects from KDTs. Over half (55%) of the participants were found to be ‘accepting’ of KDTs (see Fig. 3). The rest of the cohort was approximately equally divided into those who were ‘ambivalent’ (17%), ‘skeptical’ (14%), or ‘indifferent’ (14%) towards KDTs.
3
Fig. 3. Attitudinal analysis of beliefs about ketogenic dietary therapies.
were lower (t(83) = −2.88; p = 0.005) (Fig. 4B) among those categorized as KDT responders. The Necessity–Concerns differential was higher in KDT responders than in nonresponders (t(83) = 5.57; p b 0.000), indicating a more negative perception of treatment in nonresponders. Attitudinal group was also significantly related to KDT response (p = 0.004). Those who were ‘accepting’ of KDTs had the highest proportion of responders (n = 37) compared to nonresponders (n = 9); the ‘indifferent’ and ‘skeptical’ groups had the lowest proportion of responders compared to nonresponders (4 responders and 7 nonresponders in the ‘indifferent’ group and 5 responders and 8 nonresponders in the ‘skeptical’ group). 4. Discussion
3.4. Relationship between beliefs and response to KDTs There was a significant relationship between parents' beliefs about KDTs and their children's response to KDTs. Necessity beliefs were higher (t(83) = 5.58; p b 0.000) (Fig. 4A) and concerns about KDTs
Fig. 2. Percentage of parents expressing concerns about their child following ketogenic dietary therapies, as indicated by responding ‘strongly agree’ or ‘agree’ to BMQ-KDT— Concerns items.
This is the first study, to our knowledge, to assess parents' beliefs regarding the use of KDTs for their child's epilepsy using the BMQ-KDTs. Many parents had a positive perception of KDTs and were convinced of the necessity of KDTs for their children, although others had doubts regarding the necessity of this treatment. Many reported concerns about the potential long-term effects of KDTs. Parental beliefs about KDTs were significantly correlated with patient response. The fact that long-term side effects were the most prevalent concern about KDTs is rational for individuals who have been following dietary treatment for long periods of time. Long-term effects of KDTs are outlined to parents and/or patients when they are considering KDTs, but this issue may deserve more focus during initial consultations in order to clarify disconnects between facts and beliefs. Poor linear growth, nephrolithiasis, fractures, and dyslipidemia have been reported by 82%, 25%, and 21% of 28 individuals following KDTs for N6 years [16], but other studies have reported such consequences to be much rarer or less severe than previously thought [17–19]. For example, although there is evidence for a negative effect on height and/or weight from KDTs, particularly when treatment is started at a young age [17, 20–24], some studies have reported little or no effect on growth parameters [17,25,26] and catch-up growth has been observed in children when KDTs are discontinued [27]. Lipid levels tend to return to baseline levels over a period of two years [16,28], can return to normal when the KDT is discontinued [9,20,29], or may even be improved by KDTs [30–32]. The most common adverse effects of KDTs, such as constipation, abdominal pain, or anorexia, tend to be nonsevere, transient, and managed with medication or dietary manipulation. Even so, this study has highlighted that long-term side effects are important not only to health professionals but also to parents. As with any treatment, the
4
N.E. Schoeler et al. / Epilepsy & Behavior 39 (2014) 1–5
Fig. 4. A — Mean [95% confidence interval] BMQ-KDT—Necessity score in responders and nonresponders. B — Mean [95% confidence interval] BMQ-KDT—Concerns score in responders and nonresponders.
benefits obtained from KDTs are constantly weighed up against the difficulties of following the diet and (potential) adverse effects. We showed that parental beliefs regarding KDTs were correlated with their child's response to KDTs. This may be because people who have positive beliefs about KDTs engage in less nonadherent behavior, which means that their children are more likely to respond, although this cannot be definitively established from this study. Improving adherence to treatment may have a greater effect than improvements in specific medical therapies [4], so it is important to identify people who are likely to have poor adherence. This is particularly relevant for KDTs, where blood and urine ketones, which are subject to large inter- and intra-individual variability and are affected by many different factors, are commonly used to indicate treatment adherence. There are important limitations to this study. This study is observational and cross-sectional, so it may be that those people who have had a good experience with KDTs (which may have nothing to do with their adherent or nonadherent behavior) are more likely to have a positive attitude towards the treatment. This needs exploration in a prospective study. Another limitation of this study is ascertainment bias: many questionnaires were completed by parents of children who had been following KDTs for a long time and parents with children who discontinued KDTs before the 3-month point due to, for example, increased seizure frequency or poor tolerability, and would not have had an opportunity to complete the BMQ-KDTs. Thus, although most parents' views regarding the necessity of KDTs outweighed any concerns regarding the potential harm of KDTs, the proportion of people who viewed the treatment as positive would probably be higher than if attitudes and behavior had been evaluated prospectively. The BMQ-KDTs would be of greater value if completed before starting KDTs. A further limitation is that we did not have a behavioral measure of adherence to KDTs. We used parental reports of efficacy of the diet in reducing seizure frequency as an indicator of an adherence-dependent outcome. We attempted to adapt the Medication Adherence Report Scale for use as a behavioral measure of adherence by substituting references to medication to KDTs. However, scores on the adapted scale were not related to perceptions of KDTs or to reports of seizure frequency. The moderate Cronbach's alpha correlation coefficient (for BMQKDT—Concerns) may have been affected by the small number of items in the questionnaire, the positive distribution of scores, or the small
sample size. A larger sample size would help approximate the nonnormal distribution towards normality and reduce the negative bias imposed on the alpha value [33]. This has been found to reduce bias to a greater extent than when the number of items/test length was increased. It would be of interest to see whether results differ when selfreported (as opposed to parental) beliefs are assessed, although only a small number of children following KDTs may be able to answer the BMQ-KDTs themselves. Collecting these data for adults who are following KDTs and perhaps comparing answers from people following different types of KDTs are areas for future study. 5. Conclusions In conclusion, we have found the BMQ-KDTs to be a useful tool to quantify parents' beliefs regarding the use of KDTs for their child's epilepsy and to highlight common areas of concern. Although this study has not assessed the relationship between beliefs and adherence, it is likely that the relationship between treatment beliefs and clinical outcome (seizure control) is mediated, at least in part, by adherence. According to the Necessity–Concerns Framework, a better understanding of individuals' perceptions of their treatment can help to identify those at risk of nonadherence. Interventions may then be tailored to the needs of each individual by addressing salient perceptual barriers to adherence. Further work is needed to develop KDT-specific items to assess adherence to KDTs. Acknowledgments NS is supported by a UCL Impact Studentship in conjunction with Epilepsy Society. JWS is supported by the Dr. Marvin Weil Epilepsy Research Fund. RH is supported by an NIHR Senior Investigator Award. The authors acknowledge Dr. Christin Eltze, Marian Sewell, Carla Fitzachary, and Georgiana Fitzsimmons from Great Ormond Street Hospital; Dr. Ruth Williams, Mary-Anne Leung, Tara Randall, Layla Alford, Amanda Tomalin, and Dr. Elaine Hughes from Evelina London Children's Hospital; Emma Williams, Elizabeth Neal, Julie Edwards, Teresa Stein, and Dr. Archana Desurkar from Matthew's Friends; Angela Mensah from Young Epilepsy; Dr. Sunny Philip, Bernie Concannon, and Vanessa Hopkins from Birmingham Children's Hospital; Dr. Alasdair Parker and Dr. Anna Maw from Cambridge University Hospital; and
N.E. Schoeler et al. / Epilepsy & Behavior 39 (2014) 1–5
Dr. Penny Fallon, Orla Stone, and Nicole Dos Santos from St George's Hospital. Disclosure of conflicts of interest Professor JH Cross has cosupervised a PhD student funded by Vitaflo. The remaining authors have no conflicts of interest in relation to this work. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2014.07.020. References [1] Levy R, Cooper P, Giri P. Ketogenic diet and other dietary treatments for epilepsy (review). Cochrane Database Syst Rev 2012;3:CD001903. [2] Neal EG, Chaffe H, Schwartz RH, Lawson MS, Edwards N, Fitzsimmons G, et al. The ketogenic diet for the treatment of childhood epilepsy: a randomised controlled trial. Lancet Neurol 2008;7:500–6. [3] Cerveny P, Bortlik M, Kubena A, Vlcek J, Lakatos PL, Lukas M. Nonadherence in inflammatory bowel disease: results of factor analysis. Inflamm Bowel Dis 2007; 13:1244–9. [4] Brown MT, Bussell JK. Medication adherence: WHO cares? Mayo Clin Proc 2011; 86:304–14. [5] Horne R, Weinman J, Hankins M. The beliefs about medicines questionnaire: the development and evaluation of a new method for assessing the cognitive representation of medication. Psychol Health 1999;14:1–24. [6] Chapman SC, Horne R, Chater A, Hukins D, Smithson WH. Patients' perspectives on antiepileptic medication: relationships between beliefs about medicines and adherence among patients with epilepsy in UK primary care. Epilepsy Behav 2014; 31:312–20. [7] Horne R, Chapman SC, Parham R, Freemantle N, Forbes A, Cooper V. Understanding patients' adherence-related beliefs about medicines prescribed for long-term conditions: a meta-analytic review of the Necessity–Concerns Framework. PLoS One 2013;8:e80633. [8] Schoeler NE, Cross JH, Sander JW, Sisodiya SM. Can we predict a favourable response to Ketogenic Diet Therapies for drug-resistant epilepsy? Epilepsy Res 2013;106:1–16. [9] Klein P, Janousek J, Barber A, Weissberger R. Ketogenic diet treatment in adults with refractory epilepsy. Epilepsy Behav 2010;19:575–9. [10] Coppola G, Veggiotti P, Cusmai R, Bertoli S, Cardinali S, Dionisi-vici C, et al. The ketogenic diet in children, adolescents and young adults with refractory epilepsy: an Italian multicentric experience. Epilepsy Res 2002;48:221–7. [11] Amari A, Grace NC, Fisher WW. Achieving and maintaining compliance with the ketogenic diet. J Appl Behav Anal 1995;28:341–2.
5
[12] Kossoff EH, Doerrer SS, Turner Z. How do parents find out about the ketogenic diet? Epilepsy Behav 2012;24:445–8. [13] Farasat S, Kossoff EH, Pillas DJ, Rubenstein JE, Vining EP, Freeman JM. The importance of parental expectations of cognitive improvement for their children with epilepsy prior to starting the ketogenic diet. Epilepsy Behav 2006;8:406–10. [14] Tavakol M, Dennick R. Making sense of Cronbach's alpha. Int J Med Educ 2011; 2:53–5. [15] Kline P. The handbook of psychological testing. 2nd ed. London: Routledge; 1999. [16] Groesbeck D, Bluml R, Kossoff EH. Long‐term use of the ketogenic diet in the treatment of epilepsy. Dev Med Child Neurol 2006;48:978–81. [17] Liu Y-MC, Williams S, Basualdo-Hammond C, Stephens D, Curtis R. A prospective study: growth and nutritional status of children treated with the ketogenic diet. J Am Diet Assoc 2003;103:707–12. [18] Furth SL, Casey JC, Pyzik PL, Neu AM, Docimo SG, Vining EP, et al. Risk factors for urolithiasis in children on the ketogenic diet. Pediatr Nephrol 2000;15:125–8. [19] Kang HC, Chung DE, Kim DW, Kim HD. Early- and late-onset complications of the ketogenic diet for intractable epilepsy. Epilepsia 2004;45:1116–23. [20] Patel A, Pyzik PL, Turner Z, Rubenstein JE, Kossoff EH. Long-term outcomes of children treated with the ketogenic diet in the past. Epilepsia 2010;51:1277–82. [21] Vining EPG, Pyzik P, McGrogan J, Hladky H, Anand A, Kriegler S, et al. Growth of children on the ketogenic diet. Dev Med Child Neurol 2002;44:796–802. [22] Williams S, Basualdo-Hammond C, Curtis R, Schuller R. Growth retardation in children with epilepsy on the ketogenic diet: a retrospective chart review. J Am Diet Assoc 2002:405–7. [23] Neal EG, Chaffe HM, Edwards N, Lawson MS, Schwartz RH, Cross JH. Growth of children on classical and medium-chain triglyceride ketogenic diets. Pediatrics 2008;122:e334–40. [24] Peterson SJ, Tangney CC, Pimentel-Zablah EM, Hjelmgren B, Booth G, Berry-Kravis E. Changes in growth and seizure reduction in children on the ketogenic diet as a treatment for intractable epilepsy. J Am Diet Assoc 2005;105:718–25. [25] Nordli DR, Kuroda MM, Carroll J, Koenigsberger DY, Hirsch LJ, Bruner HJ, et al. Experience with the ketogenic diet in infants. Pediatrics 2001;108:129–33. [26] Couch S, Schwarzman F. Growth and nutritional outcomes of children treated with the ketogenic diet. J Am Diet Assoc 1999;99:1573–5. [27] Kim JT, Kang HC, Song JE, Lee MJ, Lee YJ, Lee EJ, et al. Catch-up growth after long-term implementation and weaning from ketogenic diet in pediatric epileptic patients. Clin Nutr 2013;32:98–103. [28] Kwiterovich Jr PO, Vining EP, Pyzik P, Skolasky Jr R, Freeman JM. Effect of a high-fat ketogenic diet on plasma levels of lipids, lipoproteins, and apolipoproteins in children. JAMA 2003;290:912–20. [29] Mosek A, Natour H, Neufeld MY, Shiff Y, Vaisman N. Ketogenic diet treatment in adults with refractory epilepsy: a prospective pilot study. Seizure 2009;18:30–3. [30] Liu YM, Lowe H, Zak MM, Kobayashi J, Chan VW, Donner EJ. Can children with hyperlipidemia receive ketogenic diet for medication-resistant epilepsy? J Child Neurol 2013;28:479–83. [31] Sharman MJ, Kraemer WJ, Love DM, Avery NG, Gomez AL, Scheett TP, et al. A ketogenic diet favorably affects serum biomarkers for cardiovascular disease in normal-weight men. J Nutr 2002;132:1879–85. [32] Paoli A, Cenci L, Grimaldi KA. Effect of ketogenic Mediterranean diet with phytoextracts and low carbohydrates/high-protein meals on weight, cardiovascular risk factors, body composition and diet compliance in Italian council employees. Nutr J 2011;10:112. [33] Sheng Y, Sheng Z. Is coefficient alpha robust to non-normal data? Front Psychol 2012;3:34.