In childhood, functional constipation is a common complaint. In the general population prevalence is reported to vary from 0.7% to 29.6% (1). This large variation may be because of lack of a generally used definition to classify constipation. Consensus is hampered by the fact that clinical presentation is diverse and pathophysiology is multifactorial. Even though several internationally accepted guidelines such as those of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (2), the Paris Consensus on Childhood Constipation Terminology (3), and the Rome III criteria (4) have been developed to provide criteria for constipation; none of them have been implemented worldwide in research or clinical practice yet.
There are different concepts on the clinical course of constipation in children. Some authors suggest that constipation is a constitutional condition that gradually disappears (5). Others find that despite intensive therapy 30% to 50% of the children persist having severe symptoms even after 5 years of follow-up (6,7).
Knowledge regarding factors influencing the clinical course of functional constipation in children is important to enable general practitioners and paediatricians to give accurate patient information, to weigh treatment strategies, and to identify children with high risk for unfavourable outcome. However, no overview of these prognostic factors exists in the literature. Therefore, our aim was to investigate and summarise the quantity and quality of the existing evidence on the course of constipation in children with and without treatment, and determinants that predict this course.
The MEDLINE database was searched from 1965 to March 2009, and Embase from 1980 to March 2009. The key words (medical subheadings [MeSH] and text words) used to describe constipation were “constipation,” “obstipation,” “coprostasis,” “encopresis,” and “soiling.” The study population was identified by the words “child,” “infant,” and “adolescent.” For MEDLINE the following query was added: (incidence [MeSH] or follow-up studies [MeSH] or prognosis or predict*[Text Word] or course*[Text Word] or epidemiologic studies). For Embase we combined the search with the strategy for detecting prognostic studies recommended by Wilczynski et al (8). In addition, reference lists of review articles and included studies were searched. No language restriction was applied.
Two reviewers independently screened all of the abstracts of identified published articles for eligibility. For this purpose, 4 inclusion criteria were used: study population consisted of children between 0 and 18 years of age; a prospective observational study design; one of the aims of the study was to evaluate the prognosis of functional constipation expressed as duration or recurrence of functional constipation and determinants that influence prognosis; and follow-up was at least 8 weeks.
Excluded were articles concerning children with a developmental delay or mental health problems (ie, eating disorders), studies investigating children with organic causes of constipation, and children with functional nonretentive faecal incontinence.
All potentially relevant studies, as well as the studies of which the abstracts did not provide sufficient information for inclusion or exclusion, were retrieved as full-text articles. Any disagreements regarding the inclusion of articles were resolved through consensus when possible or by arbitration of a third person.
To assess methodological quality of the included studies we developed a standardised list (Table 1). We modified an established criteria list used in systematic reviews of prognostic studies (9–12). Two reviewers independently rated the methodological quality with the 15 items of the quality score list. Each of the items had an answer option of “yes”/“no”/“unclear” (ie, insufficient information). A score of 1 point was given only to a criterion that is assessed with “yes.” Equal weights were applied to all items, resulting in a maximum score of 15 points. Disagreements were resolved through consensus or by arbitration of a third person.
Two reviewers independently performed a structured data extraction from the original reports. Extracted information included (if available) items referring to setting, participants (diagnosis, age, sex, and withdrawal/dropouts), interventions, and outcome measures. Disagreements were resolved by consensus or by arbitration of a third person.
Interassessor reliability on methodological quality was calculated using κ-scores (13). Our primary outcomes were the following: recovery from constipation or no constipation as defined by the authors of the included articles, and no laxative use, at the end of follow-up and successful outcome or no constipation as defined by the authors of the included articles, regardless of laxative use at the end of follow-up. Large clinical diversity among the included studies with regard to participants, disease definitions, and definition of outcomes existed. Furthermore, different statistical approaches and adjustments for different variables were used. Nevertheless results on prognosis of constipation were pooled using stratification to overcome large differences in duration of follow-up, study quality, and setting.
We refrained from statistical pooling of results with regard to prognostic factors (11), but carried out a best evidence synthesis for associations with recovery from constipation. Using the methodological quality list, quality scores were calculated as a percentage of the maximum score. High quality is defined as a score of 60% or more of the maximum score (ie, a score of ≥9 points).
In the best evidence synthesis (14,15), evidence was divided into the following levels: Strong–consistent findings (≥75% of the studies report consistent findings) in at least 2 high-quality studies; moderate–consistent findings in 1 high-quality cohort and at least 2 low-quality studies; limited–findings of 1 high-quality cohort or consistent findings in at least 3 low-quality studies; conflicting–inconsistent findings (<75% of the studies report consistent findings); and insufficient–no high-quality studies and <3 low-quality studies available. The level of evidence was based on the results in high-quality studies only in case of ≥2 high-quality studies available. Only statistically significant associations are considered as associated prognostic factors in this synthesis.
The search strategy resulted in a total of 2882 abstracts. After eligibility, screening of 20 publications was judged potentially relevant (5–7,16–33). After reading the full-text articles, 6 studies were excluded (5,19–21,29,30) because the study design was not a prospective observational study but based on a retrospective chart review (5,20,21,29,30) or a cross-sectional survey (19). Only 3 studies were not published in English but in Polish (16) and Spanish (28,32). These articles were translated. Full characteristics of the included studies are described in Table 2.
Of the 14 included studies, 6 were conducted in a general paediatric department (18,22–25,27), 7 in a paediatric gastroenterology department (16,17,26,28,31,32), and in 1 article no setting was stated (6). None of the included studies were conducted in a primary care centre.
All of the studies described a composed definition of recovery, resolution, or successful treatment of constipation as positive outcome, except for 2 studies that did not specify their outcome (18,27). Definitions of outcome measures varied strongly, nevertheless all of the studies took into account defecation frequency: in 1 study, having more than 4 bowel movements per week was a requirement for success (6); all of the other studies applied the criterion of at least 3 bowel movements per week. Frequency of faecal incontinence was included in the success definitions of 8 studies: in 5 studies (7,22,23,25,32) the success definition required <2 episodes per month, 1 study (17) allowed 1 episode per 2 weeks, and in 2 studies (16,24) children were not allowed to have any faecal incontinence. No abdominal pain or no pain with defecation was included in the success definition of 3 studies (7,25,32). All but 3 of the studies (27,28,33) took laxative use into account in their definition of success. Six studies (7,16–18,25,31) presented recovery rates of children taken off laxatives, as well as success rates including children still using laxatives.
In total, 1752 children participated in the included studies. Age was reported in mean (16,18,22–28) (mean 72.2 ± 37.6 months, range 21.0 to 118.8 months) or median (6,17,31,32) (66.0 ± 39.3 months, range 3.5 to 100.8 months) values. Distribution of sex of all randomised children was reported in all but 2 of the studies (18,23): 51.2% of participants were boys (M/F: 822/784). Duration of constipation before start of the study was reported in 12 articles (expressed in mean or median duration before start of the study) (7,16,18,31), mean or median age of onset (17,24,26–28), or defined in the inclusion criteria (6,22,32). Two studies did not report duration (23,25). Numbers on severity of constipation at baseline (eg, defecation frequency, frequency of faecal incontinence episodes, presence of abdominal pain) were not presented in 4 studies (18,25–27). All of the other included studies used symptoms of constipation to express severity of constipation.
Most articles expressed follow-up time in mean number of months (mean 32.5 ± 34.7 months). Two studies presented follow-up time in median number of months (7,31), and 1 study presented a range (22).
The 3 reviewers (M.A.M.P., M.E.J.B., and M.Y.B.) initially agreed on 85.6% of the items of the methodological quality list. Interobserver reliability of methodological quality assessment (0.71) was high. Overall methodological quality score of all of the included studies ranged from 4 to 10 out of a maximum of 15 points, with a mean score of 6.8. Only 21% of the included studies (n = 3) (7,17,31) were considered of high quality. Most prevalent shortcomings of the studies were the following: outcome measurement was not independent of prognostic factors (100%, n = 14), unstandardised and dependent measurement of prognostic factors (both 85.7%, n = 12), no presentation of association measures and measures of variance, or no multivariate analysis performed (both 78.6%, n = 11).
Of all of the children followed for 6 to 12 months (7,16,17,22–25,31), 49.3% ± 11.8% (mean ± SD) recovered and were taken off laxatives at the time of follow-up. The percentage of children who were free from complaints, regardless of laxative use, after 6 to 12 months was 60.6 ± 19.2 (7,16,17,22–25,27,28,31,32).
Children followed for 1 to 2 years showed a recovery rate of 58.0% ± 14.1% (7,16,18), and 69.3% ± 19.0% were recovered with or without laxative use (2,4,18). After a follow-up period of 5 to 10 years, 56.0% ± 11.3% of the children recovered and were no longer taking laxatives (6,7); 56.3% ± 10.4% of the children had successful outcomes regardless of laxative use (6,7,26).
A total of 74.2% ± 14.5% of the children included in a paediatric gastroenterology department (n = 979) had successful outcomes regardless of laxative use at the time of follow-up (mean follow-up: 28.8 ± 35.8 months) (7,16,17,26,28,31,32). Studies performed in general paediatric departments (18,22–25,27) (n = 643) showed a lower recovery rate of 57.8% ± 19.5% (P = 0.11) (mean follow-up: 13.0 ± 2.5 months, P = 0.31).
Studies of high methodological quality (17,31,32) (n = 634) showed a success rate of 70.9% ± 5.3%, compared with 63.8% ± 20.4% (P = 0.57) in low-quality studies (6,16,18,22–28,32) (n = 1118).
In the present review a total of 22 prognostic factors were analysed in 63 associations with recovery. In 44 evaluations no significant association was found, whereas in 19 evaluations a statistically significant association was reported. The mean sample size of studies that reported “no association” was not significantly different from the mean sample size of studies that reported a significant association (mean sample size: 130.9 ± 94.5 vs 145.6 ± 115.3, P = 0.60). All of the results of the best evidence syntheses on the prognostic factors are presented in Tables 3 and 4.
Seven studies reported on the association between sex and recovery from constipation. Two studies of high methodological quality (7,31) found no statistically significant association, and 1 high-quality study (17) found that male sex was negatively associated with recovery. In addition, 4 studies of low methodological quality (6,23,24,26) found no significant association. In conclusion, there is conflicting evidence that sex influences recovery rate.
Age at Intake
Based on 4 low-quality studies reporting no significant association between age and recovery, limited evidence for no association was found (6,22–24).
Age of Onset/Duration
One high-quality (31) and 4 low-quality studies (22–24,26) concluded that age of onset of constipation and recovery are not statistically significantly associated. In contrast, a high-quality study (7) showed that onset between 1 and 4 years of age is not significantly associated with recovery, but onset at the age of 4 years or older gives a higher recovery rate rather than onset before the age of 1 year. A low-quality study (6) supports the finding that an older age at onset was associated with a higher recovery rate. Based on these 7 articles, we conclude that evidence is conflicting.
Two high-quality studies (7,31) found no significant association between a positive family history for childhood constipation and recovery. In contrast, 1 low-quality study (6) found a negative association. Based on these 3 studies, we found strong evidence for no association.
Defecation Frequency at Intake
Two high-quality studies (17,31) found no statistically significant association between defecation frequency and recovery. We found 3 low-quality studies (6,23,24) that support this finding. In conclusion, these studies provide strong evidence that there is no association between defecation frequency and recovery.
Presence of Faecal Incontinence at Intake
Two low-quality studies reported on the association between the presence of faecal incontinence and recovery from constipation. One of them (25) found a positive association; the other (6) found no significant association. Therefore, evidence for this association is insufficient.
Frequency of Faecal Incontinence
We included 1 high-quality (17) and 1 low-quality (22) study that found no significant association between the frequency of episodes of faecal incontinence and recovery. In contrast, a high-quality (7) and a low-quality (23) study showed that in recovered children the frequency of episodes of faecal incontinence was significantly lower at baseline than in children who did not recover during follow-up (negative association). Overall, this provides conflicting evidence.
We found 3 low-quality studies reporting on history of abdominal pain or abdominal pain at presentation (19,21,22). All 3 show the same results, together providing limited evidence for no association between abdominal pain and recovery from constipation.
Urinary Tract Infection
Both of the 2 low-quality studies (23,24) found no significant association between previous urinary tract infections and recovery from constipation. Evidence for this association is insufficient.
Palpable Rectal or Abdominal Mass
Two high-quality studies (7,17) evaluated the relation between absence of a rectal or abdominal mass on physical examination and recovery from constipation. One study (7) found no statistically significant association, whereas the other study (17) found absence of a palpable rectal mass to be positively associated but absence of palpable abdominal mass not significantly associated with recovery. These findings provided conflicting evidence for an association. In addition, 1 low-quality study found no significant association between absence of an abdominal or rectal mass and recovery (24), another low-quality study found a negative association (23), and yet another low-quality study (22) investigated the association between presence of an abdominal mass and treatment failure, and found a negative association.
On the association between the ability to defecate a rectal balloon and recovery, 2 low-quality studies (22,25) reported a positive association that provides insufficient evidence. Another low-quality study (23) that reported on the association between the disability to defecate a rectal balloon and treatment failure found a positive association as well.
Relaxation of External Anal Sphincter
One low-quality study (22) found a positive association between the ability to relax the external sphincter and recovery. Another low-quality study (23) investigated the association between an abnormal contraction of the external sphincter and treatment failure, and also found a positive association. This provides insufficient evidence.
Colonic Transit Time/Total Gastrointestinal Transit Time
We found 2 studies reporting on the association between colonic transit time or total gastrointestinal transit time and recovery. One high-quality study (17) found a negative association, and 1 low-quality study (6) found no significant association. In conclusion, we found limited evidence that children with a longer transit time have a lower recovery rate.
In addition, there are several prognostic factors that were investigated in 1 high-quality study. We found limited evidence that premature birth (31), delayed passage of meconium (31), and production of large stools (17) are not associated with recovery. There is also limited evidence that children with duration of symptoms of <3 months before presentation and children with treatment duration of <2 months have a higher recovery rate than children with longer treatment or symptom duration (31).
To our knowledge, no previous reviews on prognosis or prognostic factors of childhood constipation have been performed. In the present systematic review, only 14 articles concerning the course of childhood constipation and its determinants could be included. The majority of these studies showed poor methodological quality. Furthermore, studies were heterogeneous, encompassing different definitions, populations, outcome measures, and follow-up periods. Without regard to these differences, 60.6% ± 19.2% of the children who were diagnosed with constipation are free of symptoms after 6 to 12 months. In addition, 49.3% ± 11.8% of the children studied for 6 to 12 months recovered and were taken off laxatives. After a follow-up period of 1 to 2 years, 58.0% ± 14.1% of the children recovered and were taken off laxatives, and after 5 to 10 years, this percentage does not rise any further, being 56.0% ± 11.3%. Children included in a specialist setting show a higher success rate (74.2% ± 14.5%) than children included in general paediatric departments (57.8% ± 19.5%).
Based on the present literature, there is substantial evidence that defecation frequency and a positive family history are not associated with recovery of constipation. With limited level of evidence, a short duration of symptoms and treatment before presentation results in better prognosis, whereas studies evaluating other factors in the medical history showed no relation or were insufficient to draw firm conclusions. Conflicting evidence exists on the prognostic value of sex, age of onset, and faecal incontinence. Furthermore, there is insufficient evidence available to determine the role of prognosis for one third of the prognostic factors described in literature.
A potential shortcoming of this systematic review is the literature search. To minimise the risk of missing relevant publications as much as possible, we performed an extensive and sensitive literature search without language restrictions.
Various outcome measures have been used in the included studies. A definition of recovery, resolution, or success was described by every author except for 2 (4,12), but no uniformity among these definitions existed. Of all of the symptoms that may occur with constipation, only defecation frequency was consistently included in the recovery definitions, but not all of the studies applied the same limiting value.
Because there were large variations in presentation of outcome and prognostic factors between the studies, it was impossible to perform a true meta-analysis on prognostic factors. In the best evidence synthesis we present a summary of the studies reporting on a prognostic factor. We assessed the methodological quality of the included studies for the best evidence synthesis using a standardised list. Because of misclassification of items, bias may occur. However, of the 14 included studies, only 3 scored as having high quality. Because of the low-quality scores of most studies, misclassification of 1 item would not change the classification into a high methodological quality. Therefore we assume that the effect of a possible bias on the results is minimal. Nevertheless, published results per prognostic factor were scarce. Therefore, the results of the evidence-based synthesis should be interpreted with caution.
In the best evidence synthesis, we only considered statistically significant associations as associated prognostic factors. We included several studies with a small sample size (implying low statistical power), of which misclassification could have occurred because their results did not reach statistical significance. Statistically pooling of data would have been a solution to this problem, but it was not possible because of the large clinical heterogeneity. However, we found that small sample size did not influence the results of our best evidence synthesis.
Despite differences between the studies included, we statistically pooled data to summarise results on prognosis of constipation. Although overall 6- to 12-month success rate of constipation in children was found to be 60.6%, a large variation (range 36.0%–98.4%) (22,28) among included studies was found. Interpretation of these pooled recovery rates is biased because studies were heterogeneous with regard to study populations and definitions of constipation and outcome measures used.
The finding that prognosis is more favourable for children in specialised centres than for children in general paediatric departments is somewhat surprising. Previous research showed that prognosis is better the earlier the treatment starts after the onset of constipation (1). Because children usually first present to their general practitioner or paediatrician before being referred to a more specialised centre, one would expect better prognosis in first- and second-line settings. Furthermore, children with constipation referred to a tertiary centre are most likely children with more severe symptoms of constipation unresponsive to conventional treatment. Children seen in specialist settings may receive more advanced or more aggressive treatment than children in general paediatric settings. Because of the large diversity between the studies, it was impossible to make a valid comparison of study populations. This also counts for the treatment regimens applied, although the evidence for an effect of treatment of functional constipation is not sufficiently proven (34).
The present literature shows strong evidence that a positive family history is not associated with recovery of constipation. Strong evidence also exists that defecation frequency is not an influence on prognosis of constipation. This finding supports the idea that functional constipation is a disease entity that has more aspects than defecation frequency only, as described in the Rome III criteria (4).
The present review does not provide insight into the prognostic value of faecal incontinence. A negative influence would be expected based on experience, but at the moment evidence is conflicting because of a lack of studies evaluating its role.
The results of our review show that further research by means of large follow-up studies on prognosis of childhood constipation and factors of influence on prognosis is necessary. We recommend using clear definitions for both the diagnosis of constipation and the recovery from constipation. This definition preferably is a uniform definition used worldwide, taking into account all aspects of constipation, such as the Rome III criteria (4). It is important to investigate prognosis not only in children seen in specialised settings but also in a more general population to gain insight into possible differences between these settings and the prognosis of constipation of recent onset. In addition, more detailed registration of symptom severity and treatment regimens applied is needed.
The few studies published on prognosis of childhood functional constipation and its predictive factors showed large clinical diversity and poor methodological quality. Overall 6- to 12-month recovery rate of constipation in children was found to be 60.6% regardless of laxative use, but large variation ranging from 36.0% to 98.4% among the included studies was found. Children included in a specialist setting show a higher recovery rate than children included in general paediatric departments. Recovery rate showed no relation to defecation frequency or positive family history. Based on the present literature, we are not able to identify a group of children at risk for poor outcome.
1. van den Berg MM, Benninga MA, DiLorenzo C. Epidemiology of childhood constipation: a systematic review. Am J Gastroenterol 2006; 101:2401–2409.
2. Baker SS, Liptak GS, Colletti RB, et al. Constipation in infants and children: evaluation and treatment. A medical position statement of the North American Society for Pediatric Gastroenterology and Nutrition. J Pediatr Gastroenterol Nutr 1999; 29:612–626.
3. Benninga M, Candy DC, Catto-Smith AG, et al. The Paris Consensus on Childhood Constipation Terminology (PACCT) Group. J Pediatr Gastroenterol Nutr 2005; 40:273–275.
4. Rasquin A, DiLorenzo C, Forbes D, et al. Childhood functional gastrointestinal disorders: child/adolescent. Gastroenterology 2006; 130:1527–1537.
5. Abrahamian FP, Lloyd-Still JD. Chronic constipation in childhood: a longitudinal study of 186 patients. J Pediatr Gastroenterol Nutr 1984; 3:460–467.
6. Staiano A, Andreotti MR, Greco L, et al. Long term follow up of children with chronic idiopathic constipation. Dig Dis Sci 1994; 39:561–564.
7. van Ginkel R, Reitsma JB, Buller HA, et al. Childhood constipation: longitudinal follow up beyond puberty. Gastroenterology 2003; 125:357–363.
8. Wilczynski NL, Haynes RB. Optimal search strategies for detecting clinically sound prognostic studies in EMBASE: an analytic survey. J Am Med Inform Assoc 2005; 12:481–485.
9. Borghouts JA, Koes BW, Bouter LM. The clinical course and prognostic factors of non-specific neck pain: a systematic review. Pain 1998; 77:1–13.
10. Hudak PL, Cole DC, Frank JW. Perspectives on prognosis of soft tissue musculoskeletal disorders. Int J Rehabil Res 1998; 21:29–40.
11. Altman DG. Systematic reviews of evaluations of prognostic variables. BMJ 2001; 323:224–228.
12. Hayden JA, Coté P, Bombardier C. Evaluation of the quality of prognosis studies in systematic reviews. Ann Intern Med 2006; 144:427–437.
13. Landis JR, Koch GG. An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics 1977; 33:363–374.
14. van Tulder MW, Furlan A, Bombarbier C, et al, Editorial Board of the Cochrane Collaboration Back Review Group. Updated method guidelines for systematic reviews in the Cochrane Collaboration Back review Group. Spine 2003; 28:1290–1299.
15. Sackett DL, Straus SE, Richardson WS, et al. Evidence-based Medicine: How to Practice and Teach EBM. Edinburgh: Churchill Livingstone; 2000.
16. Banaszkiewicz A, Bibik A, Szajewska H, et al. Functional constipation in children; a follow up study [Polish]. Pediatria Wspolczesna 2006; 8:21–23.
17. de Lorijn F, van Wijk MP, Reitsma JB, et al. Prognosis of constipation: clinical factors and colonic transit time. Arch Dis Child 2004; 89:723–727.
18. Elshimy N, Gallagher B, West D, et al. Outcome in children under 5 years of age with constipation: a prospective follow-up study. Int J Clin Pract 2000; 54:25–27.
19. Gallagher B, West D, et al. Characteristics of children under 5 referred to hospital with constipation: a one year prospective study. Int J Clin Pract 1998; 52:165–167.
20. Khan S, Campo J, Bridge JA, et al. Long term outcome of functional childhood constipation. Dig Dis Sci 2007; 52:64–69.
21. Keuzenkamp-Jansen CW, Fijnvandraat CJ, Kneepkens CM, et al. Diagnostic dilemmas and results of treatment for chronic constipation. Arch Dis Child 1996; 75:36–41.
22. Loening-Baucke VA. Factors responsible for persistance of childhood constipation. J Pediatr Gastroenterol Nutr 1987; 6:915–922.
23. Loening-Baucke V. Factors determining outcome in children with chronic constipation and faecal soiling. Gut 1989; 30:999–1006.
24. Loening-Baucke V. Constipation in early childhood: patient characteristics, treatment and longterm follow-up. Gut 1993; 34:1400–1404.
25. Loening-Baucke V. Balloon defecation as a predictor of outcome in children with functional constipation and encopresis. J Pediatr 1996; 128:336–340.
26. Michaud L, Lamblin M, et al. Outcome of functional constipation in childhood: a 10-year follow-up study. Clin Pediatr 2009; 48:26–31.
27. Miele E, Simeone D, Marino A, et al. Functional gastrointestinal disorders in children: an Italian prospective survey. Pediatrics 2004; 114:73–78.
28. Polanco I, Abarca L, et al. Longitudinal study of constipation symptoms and dietetic habits in children. FREI study. [Spanish]. Pediatrika 2004; 24:9–16.
29. Procter E, Loader P. A 6-year follow-up study of chronic constipation and soiling in a specialist paediatric service. Child Care Health Dev 2003; 29:103–109.
30. Sutphen JL, Borowitz SM, Hutchison RL, et al. Long-term follow-up of medically treated childhood constipation. Clin Pediatr 1995; 34:576–580.
31. van den Berg MM, van Rossum CH, de Lorijn F, et al. Functional constipation in infants: a follow-up study. J Pediatr 2005; 147:700–704.
32. Martinez-Costa C, Palao-Ortuño MJ, et al. Functional constipation: prospective study and treatment response. [Spanish]. Anales de Pediatria 2005; 63:418–425.
33. Benninga MA. Children with constipation: what happens to them when they grow up? Scand J Gastroenterol 2004; 39(Suppl 241):23–26.
34. Pijpers MAM, Tabbers M, Benninga MA, et al. Currently recommended treatments of childhood constipation are not evidence based. A systematic literature review on the effect of laxative treatment and dietary measures. Arch Dis Child 2009; 94:117–131.
© 2010 Lippincott Williams & Wilkins, Inc.