Comparing the effectiveness of Hydrastis canadensis mother tincture against constitutional homoeopathic medicines in treatment of pediatric constipation: An open-label, randomized, exploratory, equivalence, pilot trial
Kapil Kumar Shaw*1, Gautam Ash2, Rajib Purkait3
- Postgraduate Trainee, Dept. of Pediatrics, National Institute of Homoeopathy, Govt. of India
- Professor and Head, Dept. of Pediatrics, National Institute of Homoeopathy, Govt. of India
- Postgraduate Trainee, Dept. of Homoeopathic Pharmacy, National Institute of Homoeopathy, Govt. of India
Background: Constipation is a common problem in childhood that affects the child’s activity and behavior. Homoeopathy is claimed to have a promising role in the pediatric constipation. Homoeopathic medicine Hydrastis canadensis (HC) mother tincture (ϕ) has its probable effect in pediatric constipation, but its action has not yet been evaluated systematically. We aimed to compare the effectiveness of pre-decided HC ϕ against constitutional homoeopathic medicines (CH) in treatment of pediatric constipation.
Method: An open-label, randomized (1:1), two parallel arms, exploratory, pilot pragmatic trial was conducted on 40 children suffering from constipation, as defined by the Rome III diagnostic criteria for functional constipation. Constipation severity transformed percentage scores and physician plus parent assessed global status (‘improved’ or ‘not improved’) were the primary and secondary outcomes respectively, assessed at baseline, and 3 and 6 months after treatment. Comparative analysis was done on the protocol-compliant sample to detect group differences.
Results: Out of 84 screened, 44 were enrolled and randomized, 2 dropped out in each group, 40 completed the trial. Groups were comparable at baseline (all P > 0.01). After 3 and 6 months of intervention, both the outcome measures changed significantly in both the groups (all P < 0.01) without any significant group differences (all P > 0.01).
Conclusion: HC ϕ and CH – both the therapies seemed to be equally effective in treatment of pediatric constipation. Further rigorous independent replications are warranted.
Keywords: Homoeopathy; Hydrastis canadensis; Constipation; Pediatrics; Randomized controlled trial
Constipation is one of the commonest digestive complaints in children, which has recently grown to quite an alarming height. Like many other functional disorders, its etiology, patho-physiology is ill-understood. This results in strongly-held, believes-driven and self-introduced management strategies, which are blended with the culture of the country, sometimes even harmful to children. However, body of the scientific knowledge has grown both in depth and width during last decade . Estimates for prevalence of constipation in children vary widely in different series and in different geographical areas. The prevalence of constipation in world-wide general population ranged from 0.75-79% and prevalence of constipation in worldwide pediatric population ranged from 0.7-29.6%. In India, childhood constipation accounts for 3% visit to general pediatric clinics and 30% of visits to pediatric gastro-enterologists . A normal child might have a soft stool only every 2nd and 3rd day without difficulty, and this is not constipation. A hard stool passed with difficulty every 3rd day should be treated as constipation. The normal stool frequency decreases from 4 or more per day during infancy to once per day at 4 years of age. A stool frequency of < 2/ weeks is considered abnormal for all ages . Signs and symptoms may vary according to the age of the child. Infants may present with clinical features, such as straining, turning red in face and crying. Toddlers may present with passing painful and hard stool that may be associated with bleeding per rectum, secondary to a small tear in the anal canal which leads to further withholding. Parents may note that the child resists the urge to defecate . Patients may also present with retentive fecal soiling secondary to withholding that can be mistaken as diarrhea and will force parents to seek medical care. Other manifestations include abdominal pain, distention, and feeling of incomplete disimpaction causing nausea and decreasing appetite. Some patients also present with enuresis and urinary tract infections, because stool masses press on the urinary tract and block normal urinary flow . Several risk factors have been identified in association with pediatric constipation including low consumption of dietary fibers and school-related stressful events . The patho-physiology of constipation in children is multi-factorial and is associated with interactions of many risk factors. Many organic diseases cause constipation. However, the majority of patients with constipation secondary to organic conditions usually have other clinical features suggestive of the relevant underlying organic disease. Organic diseases presenting as isolated constipation are rather uncommon . Constipation can be divided into two major groups: Functional and organic. The functional group causes of 85% of the patients and only 15% have the organic etiology . Treatment depends on the type of constipation. If it is functional, then behavioral training, dietary changes along with medicine can cure the case. Nutritious diets fruits, vegetables and adequate fluid are recommended. Behavioral training involves establishing a positive routine of sitting on toilet for passing stool after meals regularly. Embarrassment and punishment should be avoided. Effective education of the parents and children with regards to constipation is crucial in chronic behavioural pattern .
Although it is claimed that homoeopathic treatment is effective in cases of pediatric constipation, systematic research evidences remain sparse. We hypothesized that pre-decided Hydrastis canadensis (HC) mother tincture (ϕ) could produce equivalent treatment effect to constitutional homoeopathic medicines (CH), based on characteristic symptoms in pediatric constipation.
Trial design: This open-label, randomized, prospective, exploratory, pilot, pragmatic, two parallel arms clinical trial was conducted at the pediatric out-patient departments of National Institute of Homoeopathy (NIH). Institutional Ethical Committee (IEC) approved the protocol [Ref. No. 5-023/NIH/PG/Ethical Comm. 2009/Vol. III/1976 (A/S); dated March 27, 2017]. The trial protocol (unpublished) and full dissertation was submitted as the postgraduate thesis of the corresponding author to The West Bengal University of Health Sciences.
Participants: Inclusion criteria were the children suffering from constipation as diagnosed by the Rome III diagnostic criteria of functional constipation (ICD 10 code K59.00) and symptomatic for at least last month, both sexes, age 1-18 years, and guardians willing to make their children participate in the study and giving written informed consent. Exclusion criteria were any congenital abnormalities and cases of extensive morbid pathologies, where surgical intervention were indicated or required, children who were too unwell to take part, inability to read patient information sheets and/or denied consent to take part, diagnosed cases of developmental disorders, uncontrolled hypothyroidism, unstable psychiatric complaints or other systemic diseases affecting quality of life, currently receiving homoeopathic treatment for any chronic condition(s), self-reported immune compromised conditions, and substance abuse and/or dependence.
Intervention: Each standardized dose was consisted of 15 drops of the HC ϕ in 20 ml of distilled water. The medicine was obtained from a Good Manufacturing Practice (GMP) certified firm – SBL Pvt. Ltd. The medicine was directed to be taken twice daily orally on clean tongue with empty stomach for 7 consecutive days followed by placebo and repeated as and when necessary. Such intervention continued for 6 months. The participants were assessed by the two homoeopaths every month in every follow-up. Irrespective of the allocated codes, provision was kept for prescribing ‘acute medicines’ based on ‘acute totality’ to meet adverse or serious adverse events as per homoeopathic principles.
Control: Control was planned as administering individualized homoeopathic medicines in centesimal scale potencies as detailed above and in individualized dosage, as decided appropriate to the case or condition. As in the experimental arm, instructions were given to take each dose orally on clean tongue with empty stomach. Duration of treatment was 6 months. On each occasion, single individualized medicine was prescribed considering symptom totality, clinical history details, constitutional traits, miasmatic features, repertorization as and when required, consultation with Materia Medica, and agreement among the two prescribing homoeopaths. Individualized dose was used based on the two homoeopaths’ judgment of susceptibility and agreement. Subsequent prescriptions were produced in accord with Kent’s observations and Hering’s law. One of the prescribers possessed masters’ degree in homoeopathy and had been practicing ‘classical’ homoeopathy for more than 30 years. The corresponding author was a postgraduate trainee at NIH with 3 years of experience. The two homoeopaths involved were affiliated with respective state councils.
Outcomes: The primary outcome measure was the constipation severity scores, derived from the items mentioned in Rome III diagnostic criteria. Presence of each symptom was ascribed with a score of 1 or otherwise as 0, then the total score was calculated by summing up, and then the raw score was transformed into percentage scores. The main reason for developing this scoring system was the absence of any pre-validated outcome scales on pediatric constipation. The psychometric validity and reliability of this newly developed instrument remained to be addressed formally in future studies, but beyond the scope of this paper. The secondary outcome measures were physician-rated and parent-rated global health status as ‘improved’ or not improved’. Both the outcome measures were assessed at baseline, and after 3 and 6 months.
Sample size: Formal effect size calculation was not possible on account of absence of any open randomized trial on pediatric constipation comparing effectiveness of HC ϕ against CH. As the designed trial was equivalence or non-inferior in type, and HC ϕ was considered as one of the recommended homoeopathic medicines for the said condition, we presumed very small difference and small effect size in favor of either of the therapies and considered both the therapeutic options as equally effective (H0). Hence assuming a small effect size (d) of 0.1, and fixing α at 0.01, power (1 – β) = 80% (minimum recommended) and allocation ratio of 1.1, in order to identify a significant difference between two independent means of constipation severity scores (primary outcome measure) of the two groups by unpaired t test (two tailed), we would have required 3142 patients, i.e. 1571 in each arm. However, keeping in mind the exploratory nature of the trial, feasibility issues, and time constraints, we targeted to achieve 44 patients, i.e. 22 in each arm. Given β/α ratio of 4, effect size of 0.1 and sample size of 22 in each arm, post hoc analysis revealed a compromised power (1-β) of 22%. The final analysis was restricted on the protocol-compliant sample (n=40) only; thus further compromising the study power up to 21.7%. This indicated the need for cautious interpretation while extrapolating the results to general population.
Randomization: Intervention (HC ϕ) or comparator (CH) was implemented as per the random number chart created by using the random number generator at StatTrek. The chart was generated using 4 blocks of size restricted to 10 (4 × 10 = 40) plus another block of size 4 to maintain alike allocation between groups and 1:1 ratio easily; thus the same numbers of patients were randomized to either code 1 or 2, either of intervention or control.
Blinding: The prescribing homoeopaths and the patients – both were aware (i.e. open-label) of the generated allocation codes all through the study. Allocated codes were maintained till the end of the trial. This open list of random numbers with allocated codes was made available with the pharmacist, and he was instructed for dispensing of either HC ϕ or CH to the patients according to the chart.
Statistical methods: The protocol-compliant sample was analyzed in the end. Both descriptive and inferential statistics were applied. Baseline comparability of the groups was tested using chi-square (for categorical data) or unpaired t test (for continuous data). Outcome data obtained at baseline and after 3 and 6 months were compared by paired t test and repeated measure ANOVA. Mean group differences after 3 and 6 months were compared by unpaired t test. P values for two-tailed tests were set at less than 0.01 as statistically significant. No interim and subgroup analyses were planned. Statistical Package for the Social Sciences, version 20 (IBM Corp., IBM SPSS Statistics for Windows, Armonk, NY: USA) was used for statistical analysis of data. Reporting adhered to the CONSORT  and RedHot  guidelines for reporting trials, Mathie’s criteria for evaluating the model validity of homoeopathic treatment [7, 8] (MVHT) and Saha’s criteria for reporting quality of individualization in clinical trials of homoeopathy .
Study flow: As per the pre-specified inclusion and exclusion criteria, 84 children suffering from constipation were screened; 40 were excluded on account of various reasons; 44 met the eligibility criteria and were enrolled into the trial. Following that, baseline socio-demographic and outcome data were obtained and were randomized to either HC ϕ (n=22) or CH (n=22). After 3 and 6 months of intervention, outcome data were recorded again. During course of treatment, 4 dropped out (2 in each group); 40 completed the trial and entered into the analysis. (Figure 1)
Figure 1: Study flow diagram
Recruitment: The study continued for 14 months; started in April 2017 and continued up to May 2018.
Baseline data: Distribution of socio-demographics features including age group, gender, residence, religion, and body mass index were similar between the HC ϕ and CH group (all P > 0.01, Table 1) ensuring comparability of the two groups.
Distribution of the constipation severity score percentages at baseline were also similar between the two groups with no significant differences (HC ϕ group 57.5 ± 11.5 vs. CH group 60.8 ± 12.4; mean difference 3.3 ± 3.8, 95% CI -4.3 to 10. 9, t score = 0.881, P = 0.383, unpaired t test). It implied that the groups were comparable at baseline.
Table 1: Socio-demographic features of the participants at baseline (N=40)
|Characteristics||HC ϕ group (n=20); n (%)||CH group (n=20); n (%)||P value|
|Age groups (yrs):
§ 1 – 9
§ 10 – 18
|Body mass index
Chi-square test; P values two-tailed less than 0.01 considered as statistically significant
Numbers analyzed: Outcomes from 20/22 from each group was complete respectively; rest dropped out of the trial. Only the protocol-compliant patients (i.e. 20 in each group) entered into the final analyses.
Outcomes and estimation:
- Within group changes of constipation severity score over 3 and 6 months in the HC ϕ group: After 3 months of treatment, the reduction was statistically significant [pre 57.5 ± 11.5 vs. post 41.7 ± 17.5, P < 0.001, paired t test]. The reduction was also significant after 6 months [pre 57.5 ± 11.5 vs. post 24.2 ± 19.1, P < 0.001, paired t test]. Repeated measure ANOVA (0 – 3 mo – 6 mo) comparing baseline value with those obtained after 3 and 6 months showed statistically significant improvement [Wilks’ lambda = 0.262, partial eta-squared = 0.738, F2, 18 = 25.355, P < 0.001]. (Figure 2)
- Within group changes of constipation severity score over 3 and 6 months in the CH group: After 3 months of treatment, the reduction was statistically significant [pre 60.8 ± 12.4 vs. post 39.9 ± 12.6, P < 0.001, paired t test]. The reduction was also significant after 6 months [pre 60.8 ± 12.4 vs. post 25.0 ± 21.3, P < 0.001, paired t test]. Repeated measure ANOVA (0 – 3 mo – 6 mo) comparing baseline value with those obtained after 3 and 6 months showed statistically significant improvement [Wilks’ lambda = 0.248, partial eta-squared = 0.752, F2, 18 = 27.293, P < 0.001]. (Figure 2)
- Group differences were non-significant, both after 3 months [mean difference -5.0, SE 4.7, CI –14.6 to 4.6, t score = 1.057, P = 0.296, unpaired t test] and after 6 months [mean difference –5.0, SE 3, CI –17.9 to 7.9, t score 0.787, P = 0.436, unpaired t test] of intervention. (Figure 2)
- Physician assessed status over 3 and 6 months: There were no significant differences in the proportion of improved cases between the two groups, both after 3 and 6 months [HC ϕ group 14/20 vs. CH group 17/20, chi square = 1.290, P = 0.256].
- Parents assessed status over 3 and 6 months: There were no significant differences in the proportion of improved cases between the two groups, both after 3 months [HC ϕ group 15/20 vs. CH group 14/20, chi square = 0.125, P = 0.723] and 6 months [HC ϕ group 16/20 vs. CH group 12/20, chi square = 1.905, P = 0.168].
- Medicines used in the CH group: The most frequently indicated medicines in the CH group were Calcarea carbonica (n=7, 35%), Graphites (n=5, 25%), Magnesium muriaticum and Sulphur (n=4 each, 20%).
- Adverse events: No harms, unintended effects, serious adverse events, and untoward homoeopathic aggravations were reported from either group during the study period.
Figure 2: Changes in constipation severity scoring in the two groups over time
Empirically selected HC ϕ seemed to be non-inferior and equally effective to CH in treatment of pediatric constipation, as evidenced from constipation severity score and physician and parent assessed wellbeing score; however, the trial being underpowered, cautious interpretation is warranted. Further exploration is necessary.
One of the major strength of the study was randomization; i.e. each number had the same chance of being allocated to either of the two arms. It was a “gold standard” for proof of efficacy/effectiveness. The advantages of randomization included equipoise across groups, enhanced internal validity due to homogenous study groups, minimized selection bias, unbiased allocation, generation of two comparable groups, balancing among possible confounders, facilitating simple analysis, and to declare evidently that the difference was because of the treatment. Prospective study design allowed causal inference to be drawn. The two trial arms had no significant difference at baseline ensuring comparability. No placebo control was used; rather all the randomized patients received medicine – either HC ϕ or CH. Hence, ethically the study was less vulnerable. In the study design, we kept provision for pair-wise comparisons using rigorous parametric t tests and repeated measure ANOVA.
The study was the very first of its kind in homoeopathy; hence exploratory in nature. Effect size could not be calculated on account of absence of any study of similar design; hence sample size was based on assumption only. The trial was open; no blinding was used. Thus, patient selection bias, treatment assignment bias, patient evaluation bias, and data analysis bias may have affected the outcomes of the study. Besides, our analysis was restricted to protocol-compliant sample only. Imputation of missing values using appropriate statistical techniques and an intention-to-treat analysis would have generated more proper results than this. On account of the study being underpowered, much caution needs to be taken while generalizing the results to the population. Further studies are suggested on similar or more rigorous design with validated clinical scoring system and questionnaire focusing on quality of life issues.
This prospective, open, randomized, two parallel arms, exploratory, pilot, pragmatic trial was carried out at National Institute of Homoeopathy, India on 40 children suffering from constipation. Over 3 and 6 months of intervention, symptom severity and health status improved significantly in both the HC ϕ group and CH group without any significant differences in between, suggesting HC ϕ as an equally effective treatment option as that of CH in pediatric constipation. However, being an underpowered trial, results should be interpreted cautiously and further independent research in multi-centric design is required on larger sample size before making any firm recommendations.
Conflict of interest statement: The authors declare that they have no competing interests. The trial was carried out as the postgraduate thesis of the corresponding author under guidance of GA, who was a permanent teaching faculty of the institute.
Funding: We received no funding for the study. Funding required for the project was borne by the authors themselves.
Authors’ contributions: KKS conceived and designed the study. KKS and GA collected data. RP ran the statistical analyses and interpreted the data. KKS prepared the manuscript. All the authors reviewed and approved the final manuscript.
Acknowledgements: We thank Dr. Subhranil Saha, Independent Researcher for his valuable inputs in statistical analysis and drafting of the manuscript. The authors are grateful to the institutional heads of both academic and hospital section for their cooperation. We thank the fellow postgraduate trainees, staff, pharmacists, and the patients for their sincere participation in the study.
- Rajindrajith S, Devanarayana NM. Constipation in children: novel insight into epidemiology, pathophysiology and management. J Neurogastroenterol Motil. 2011;17(1):35.
- Mugie SM, Benninga MA, Di Lorenzo C. Epidemiology of constipation in children and adults: A systematic review. Best Pract Res Clin Gastroenterol. 2011;25(1):3-18.
- Ghai OP. Essential pediatrics. CBS Publishers and distributors Pvt. Limited; 2010, pp. 283-286.
- Xinias I, Mavroudi A. Constipation in Childhood. An update on evaluation and management. Hippokratia. 2015:19(1):11.
- Schulz KF, Altman DG, Moher D, for the CONSORT group. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c332.
- Dean ME, Coulter MK, Fisher P, Jobst K, Walach H. Reporting data on homeopathic treatments (RedHot): a supplement to CONSORT. Homeopathy 2007;96:42-45.
- Mathie RT, Van Wassenhoven M, Jacobs J, Oberbaum M, Roniger H, Frye J, et al. Model validity of randomised placebo-controlled trials of individualised homeopathic treatment. Homeopathy 2017;106:194-202.
- Mathie RT, Van Wassenhoven M, Jacobs J, Oberbaum M, Frye J, Manchanda RK, et al. Model validity and risk of bias in randomised placebo-controlled trials of individualised homeopathic treatment. Complement Ther Med. 2016;25:120-125.
- Saha S, Koley M, Ganguly S, Rath P, Roy Chowdhury P, Hossain SI. Developing the criteria for evaluating quality of individualization in homeopathic clinical trial reporting: a preliminary study. J Integr Med. 2014;12:13-19.
|Cite this article as: Shaw KK, Ash G, Purkait R. Comparing the effectiveness of Hydrastis canadensis mother tincture against constitutional homoeopathic medicines in treatment of pediatric constipation: An open-label, randomized, exploratory, equivalence, pilot trial. National Homoeo Recorder 2018;14(4):25-33.
Download: Comparing the effectiveness of Hydrastis canadensis mother tincture against constitutional homoeopathic medicines in treatment of pediatric constipation: An open-label, randomized, exploratory, equivalence, pilot trial