Acute otitis media (AOM) is the most common illness in children for which physicians are consulted, with an estimated 24.5 million visits per year accounting for 12.4% of all pediatric consultations. 1, 2 Review articles have reported that the 10-day course of antibiotics routinely used in the United States for AOM provides only marginal benefit and that 70 to 90% of all AOM episodes resolve spontaneously. 3–5 The rise in antibiotic-resistant bacteria may be partially a result of overuse of antibiotics for AOM. 6 A recent survey found that 55.5% of all US children had received antibiotics by 200 days of life. The most common indication was AOM. 7 The “watch and wait” method of treating with antibiotics only those cases of AOM with symptoms persisting for >3 days has become standard practice in the Netherlands, and questions remain as to whether antibiotics are indicated for most episodes of AOM. 6, 8
Homeopathic treatment is based on the principle of similars, whereby a substance that can cause certain symptoms when given in large doses to a healthy person is prescribed in very small doses to cure those same symptoms in someone who is sick. An example is the homeopathic medicine Allium cepa, made from the common red onion. Because the symptoms of a person exposed to the fumes of an onion (sneezing, lacrimation and clear nasal discharge) are similar to those of the common cold, Allium cepa is prescribed frequently to treat this illness. One of the basic tenets of homeopathic treatment is individualization, whereby two or more patients with the same medical diagnosis might receive different medicines, based on the specific symptoms of illness in each patient. For example a child with acute otitis media may be prescribed one of several different medicines commonly used for this illness based on such factors as the child’s mood, type of pain, amount of thirst and the time aggravation of symptoms (Table 1).
An estimated 3.4% of the US population used homeopathy in 1997, a 5-fold increase since 1990. 10 Homeopathy is one of the most controversial types of alternative medicine, largely because of the high dilution of the medicines used, which some critics say could be nothing more than placebo. 11, 12 Despite this several clinical trials of homeopathy for various conditions have been published reporting positive results, as well as some demonstrating no difference from placebo. 13–20 A recent metaanalysis of 89 homeopathic clinical trials found an odds ratio of 2.45 (95% confidence interval, 2.05 to 2.93) in favor of homeopathy when compared with placebo, although the authors found insufficient evidence that homeopathy is clearly efficacious for any single clinical condition. 21
Anecdotal reports during the past 100 years have claimed that homeopathy is an effective treatment for symptoms of AOM. 22–24 A recent nonrandomized open trial in Germany found a faster resolution of pain and fewer recurrences in children treated with homeopathy rather than conventional treatment, 25 whereas another open study found that 92% of patients treated with homeopathy for AOM did not require antibiotics. 26 A recent survey reported that otitis media is the third most common diagnosis seen by homeopathic physicians. 27
If homeopathy were found to provide symptomatic relief in the treatment of AOM, it might be an attractive treatment option for providers and parents during the “watch and wait” period before prescribing antibiotics. If homeopathy were found to decrease the need for antibiotic therapy, it also would be of value by reducing the risk of antibiotic resistance to common pathogens. Alternately if homeopathy were found to be ineffective, it would discourage the use of an inappropriate therapy.
In this pilot study we began the evaluation of the safety and efficacy of homeopathy in the treatment of AOM using a double blind randomized placebo-controlled trial. We wanted to determine which outcome measures seemed most appropriate and which homeopathic medicines would be most commonly prescribed in the study population, and we wanted to gather data for sample size calculations for a larger study.
Children ages 18 months to 6 years who presented at a private group pediatric practice in Seattle, WA, during January, 1996, through January, 1997, with a diagnosis of acute otitis media were considered for enrollment into the study. Diagnosis of acute otitis media was made by one of the three pediatricians or one nurse practitioner in the group, none of whom was a validated otoscopist. AOM was diagnosed when there was a middle ear effusion, along with one or both of the ear pain characterized as moderate or severe and fever of >38.0°C orally. Middle ear effusion was determined by pneumatic otoscopy, according to a clinical research form based on clinical signs from >10 000 cases found to have a predictive value of 0.80 or greater (Table 2). 28 Patients with a history of ear pain for >36 h or those who had received antibiotics within the past week or homeopathic medications within the previous 72 h were excluded from the study. Also excluded were children who had previous tonsillectomy, adenoidectomy or tympanostomy tubes as well as those with a perforated tympanic membrane and/or a discharge from the ear. Children on concurrent medication for another acute or chronic illness were also excluded from the study, as were those with a cleft palate or Down’s syndrome.
Once a diagnosis of AOM was made by the pediatrician, the parent was asked to provide informed consent by signing a form that had been approved by the Human Subjects Committee of the University of Washington. Each child was interviewed by an experienced homeopathic practitioner and an individualized homeopathic medicine was prescribed. Homeopathic practitioners included two medical doctors (JJ, DC), a physician’s assistant (LV) and a naturopathic physician (RU). Tympanometry was done at the initial visit by the homeopathic practitioners using the Welch/Allyn hand-held tympanometer (MicroTemp).
In a double blind randomized fashion, one half of the children were given active homeopathic medicine, and the other half received placebo. Parents were instructed to give 3 to 5 pellets of the study medication three times daily for 5 days, or until improvement occurred, whichever occurred first. Medication was in the form of easily dissolvable pellets, which were placed directly on the child’s tongue. Use of other medications, except for analgesics, was discouraged.
Daily symptom diaries, which included information on pain, fever, irritability, appetite, energy level, sleep, concurrent upper respiratory tract symptoms and number of doses of study medications given, were completed three times daily for the first 3 days by the parent. A diary symptom score, consisting of numerical values for each of the seven factors in the symptom diary, was totaled for each of the 8-h intervals in the diary. The maximum possible score was 9 (Table 3).
The occurrence of treatment failure during the first 5 days was determined by objective criteria that were established before initiation of the study. They were as follows: ear pain and/or a fever of greater than 38.0°C orally at any time after the first 48 h of treatment; or severe ear pain (crying from pain) and/or a fever of greater than 39.0°C orally after the first 24 h. Occurrence of treatment failure was ascertained by a daily phone call by a study assistant for the first 5 days, and children meeting these criteria were referred back to the clinic immediately for standard treatment. Follow-up visits were made by an otolaryngology resident (LD), who was blinded as to treatment allocation, at 2 and 6 weeks after the initial evaluation. An interval history was taken at these visits to determine whether a treatment failure (relapse or recurrence of AOM) had occurred, the presence or absence of middle ear effusion was determined by pneumatic otoscopy and tympanometry was done.
Because this was a preliminary study, several outcome measures were evaluated. These included the proportion of cases in each group with treatment failure after 5 days, 2 weeks and 6 weeks of treatment. Other outcome measures included comparison of symptom scores on the daily diaries during the first 3 days of treatment and the presence of middle ear effusion at the 2- and 6-week follow-up.
An intention-to treat analysis was performed, and all data analyses were completed before the concealment code was broken. The Yates corrected chi square statistic of two-by-two tables was used to compare the proportion of treatment failures between the two groups at 5 days, as well as treatment failures and the presence of middle ear effusion at the 2- and 6-week follow-up. Relative risks with confidence intervals were also calculated for these results. Student’s t test (two tailed) was used to compare the mean symptom scores between the two groups as reported in the parent diaries at various time points. Sample size calculations were made with the statistical program Epi-Info (USC, Inc., Stone Mountain, GA).
Study medications were randomized into coded bottles by a pharmacist at the Standard Homeopathic Company in Los Angeles, who held the code until the study was completed. For each of the 16 individual homeopathic medicines that are most commonly used for AOM, a box was prepared of coded bottles that had been randomized to contain either active medication or placebo using a random number generator and pattern blocks of 4 and 6. When a specific homeopathic medicine was chosen for a patient, the next coded bottle in sequence in that box of randomized active or placebo medicine was given to the parent for administration.
Medications were prepared on No. 38 lactose pellets that had been impregnated and tumbled dry with an identical amount of either a homeopathic medication in the 30C potency, prepared in accordance with the Homeopathic Pharmacopoeia of the United States, 29 or placebo. A 30C (C = 100) potency means that the active substance was diluted 1/100 in an 87% water/alcohol solution 30 successive times for a final concentration of 1 × 10−60. Placebo was prepared with the water/alcohol solution lacking the active substance. The alcohol in the original solutions evaporates during the 30-min tumbling process, and the pellets are packaged dry, without any remaining alcohol taste or smell. Both active and placebo medications had the same sweet taste of lactose pellets and were easily accepted by the children. There were no detectable differences in taste, odor or color between the treatment medication and placebo, and they were packaged in identical tubes that were sealed at the laboratory and remained unopened until delivery to the patient. As such patients and providers were blind as to the treatment group.
Because concern was expressed about possible contamination of homeopathic medicines by a pharmacologically active substance in a previous study, 12 random samples of both active and placebo bottles were submitted for independent laboratory analysis by the Departments of Medicinal Chemistry and Laboratory Medicine at the University of Washington. Antibacterial activity was assessed by dissolving pellets of study medication (weighing ∼45 mg) in 1 ml of sterile water. To 170 μl of Haemophilus Test Media were added 20 μl of this solution, after which the well was inoculated with 10 μl of an overnight culture of Haemophilus influenzae or Streptococcus pneumoniae. Gas chromatography with a mass selective detector was used to screen for 167 different pharmacologic agents, including common analgesics, antihistamines and decongestants.
Independent assessment of tympanograms was done by a clinical audiologist who categorized the likelihood of middle ear effusion as probable, unlikely or equivocal, using standardized protocols.
Seventy-five children were enrolled into the study, 36 in the group receiving active medication and 39 in the placebo group, of which 72 completed all follow-up visits (Fig. 1). There were no significant differences between the two groups in descriptive characteristics at entry into the study (Table 4). Children were predominantly white and from households where >90% of parents had a post-high school education.
Of the 75 children who were initially randomized, there were 19 treatment failures within the first 5 days of treatment, 12 in the placebo group and 7 of the children receiving homeopathic medication, which was not statistically significant (Table 5). There were fewer treatment failures in the homeopathy group at all time points, while the presence of middle ear effusion (MEE) was higher in the control group at 2 weeks and in the treatment group 6 weeks after treatment. None of these differences was statistically significant. There were no adverse effects reported in either of the 2 treatment groups, and compliance was comparable in both groups (>90%) as recorded in the symptom diary and during the follow-up daily phone calls. Parents reported the use of analgesics in 10 children in the placebo group and in 5 children receiving homeopathic medicines.
A comparison of diary symptom scores of the 69 patients for whom forms were returned (36 in the homeopathic group, 33 in the placebo group) showed a decreased symptom score at all points in time for the group receiving homeopathic medication, with a statistically significant improvement (P < 0.05) in the group receiving homeopathy after 24 and 64 h of treatment (Fig. 2).
Sample size calculations (two tailed) based on the treatment failure rates of the 2 groups at 5 days indicate that a sample size of 243 children in each of 2 treatment groups would be necessary to show a significant difference with an alpha of 0.05 and a power of 80% (beta error of 0.20).
Although there were 8 different homeopathic medicines prescribed in the study, the 4 most common medicines were prescribed in 88% of the cases. These included Pulsatilla nigrans (62.7%), Chamomilla (10.7%), Sulphur (9.3%) and Calcarea carbonica (5.3%). Indications for these four medicines can be found in Table 1. 9 Laboratory analysis of 4 randomly chosen study medications, both active and placebo, found no evidence of antibacterial activity or of any of the 167 pharmacologic agents for which screening was done.
A total of 376 tympanograms were recorded, of which 335 (89.1%) were considered interpretable. However, only 54 (72%) children had interpretable tympanograms at entry into the study, with evidence of middle ear effusion in 39 of these (52%). Of these, 19 were in the group receiving homeopathy and 20 in the placebo group. A subgroup analysis of these children with objective evidence of effusion at the first visit found 1 treatment failure in the homeopathy group and 4 in the placebo group after five days, 4 failures in the homeopathy group and 10 in the placebo group at 2 weeks and 5 failures in the treatment group with 12 in the placebo group after 6 weeks. Middle ear effusion at 2 weeks was present in 17 subjects in both the homeopathy and placebo groups; at 6 weeks 9 children in the homeopathy group had MEE compared with 8 in the group receiving placebo.
It is impossible to draw conclusions from a preliminary study such as this, because it was not intended to be large enough for adequate statistical power. However, because AOM is the most common presenting diagnosis in children and the routine use of antibiotics is being questioned more vigorously, it seems timely to report the findings of this study, which follows the natural history of the illness in a cohort of children not initially treated with antibiotics.
Various authors have reported 75 to 95% spontaneous recovery rates in children with AOM, using the method of treating with antibiotics only those children with symptoms remaining after 3 to 7 days of observation. 30, 31 Because of requirements from the human subjects committee, the children in this study were considered treatment failures at any time after 48 h if symptoms of AOM were present. Our lower rates of spontaneous recovery (70 to 80%) are likely the result of this more stringent definition of treatment failure, which includes a shorter observation period of time than that reported in the literature. The rates of MEE at 2 and 6 weeks are also high in this cohort, but comparable in both groups with objective evidence of MEE at enrollment.
There are several limitations to this study. The definition of AOM and the algorithm for MEE were such that children with a middle ear effusion and fever from some other illness could have been entered into the study, as could children with a red, bulging tympanic membrane that was freely mobile. Any misclassification because of these factors should be equally distributed in both treatment groups as a result of the randomization process and would be unlikely to affect the overall results. Future studies should be designed that require more specific symptoms of AOM as entry criteria, such as ear pain or unaccustomed rubbing or tugging at the ear, marked tympanic membrane redness and distinct bulging of the TM. 32
The low percentage of typanometrically confirmed middle ear effusions at the initial visit is also a cause for concern and is thought to be due to the inexperience of the homeopathic practitioners administering this procedure. Pediatricians skilled in tympanometry should be utilized for this in future studies. The failure rates of the subgroup of 39 children with objective evidence of MEE at entry into the study were remarkably similar to the group as a whole and suggest that any inappropriately enrolled subjects were randomly distributed between the two groups.
The generalizability of these results should be questioned, because the patients in the study were predominantly white and from educated families. Future studies should be conducted with a more diverse population. An inherent methodologic problem of any homeopathic clinical trial is the use of more than one treatment medication. Because homeopathic treatment requires individualization of the medicine to the patient, giving the same medicine to all patients would not be a valid test of homeopathy. In this study we wanted to evaluate the use of the homeopathic system of individual homeopathic prescribing, not the effects of specific medicines.
Because of the preliminary nature of this study and its small sample size, definitive results were not found. However, we did find consistently fewer treatment failures in the group receiving homeopathy after 5 days, 2 weeks and 6 weeks of follow-up compared with placebo. We also found evidence for symptomatic relief using homeopathy during the first 3 days with a statistically significant reduction in symptoms after 24 h. These results suggest that a positive treatment effect of homeopathy when compared with placebo in uncomplicated AOM cannot be excluded and that a larger study is justified.
We thank the following persons for their contributions to this work: Andrea Corage Baden, Lianne De Serres, M.D., Lucy Vaughters, P.A., C.C.H., Robert Ullman N.D., D.H.A.N.P., Carol Doroshow, M.D., Christine Caldwell, M.D., Sandra Jolley, N.P., Mark S. Phillips, Pharm.D., Gary Elmer, Ph.D and Susan Norton, Ph.D., C.C.C.-A. Funding for this study was made possible by a grant from the Standard Homeopathic Company.
1. Stephenson M. Observation may be as effective as antibiotic therapy for treating otitis media. Infect Dis Child 1997;Feb:10.
2. Froom J, Culpepper L, Grob P, et al. Diagnosis and antibiotic treatment of acute otitis media
: report from International Primary Care Network. Br Med J 1990; 300: 582–6.
3. Rosenfeld RM. What to expect from medical treatment of otitis media. Pediatr Infect Dis J 1995; 14: 731–8.
4. Lehnert T. Acute otitis media
in children: role of antibiotic therapy. Can Fam Physician 1993; 39: 2157–62.
5. Rosenfeld RM, Vertrees JE, Carr J, et al. Clinical efficacy of antimicrobial drugs for acute otitis media
: metaanalysis of 5400 children from thirty-three randomized trials. J Pediatr 1994; 124: 355–67.
6. Culpepper L, Froom J. Routine antimicrobial treatment of acute otitis media
: is it necessary? JAMA 1997; 278: 1643–5.
7. Bergus GR, Levy BT, Levy SM, Slager SL, Diritsy MC. Antibiotic use during the first 200 days of life. Arch Fam Med 1996; 5: 523–6.
8. Rosenfeld RM. An evidence-based approach to treating otitis media. Pediatr Clin North Am 1996; 43: 1165–81.
9. Boericke W. Pocket manual of homeopathic materia medica. 9th ed. New Delhi: B. Jain Publishers, 1978.
10. Eisenberg DM, Davis RB, Ettner LS, et al. Trends in alternative medicine
use in the United States, 1990–1997. JAMA 1998; 280: 1569–75.
11. Anonymous. When to believe the unbelievable. Nature 1988; 333: 787.
12. Sampson W, London W. Homeopathic treatment of childhood diarrhea. Pediatrics 1995; 96: 961–4.
13. Reilly DT, Taylor MA, McSharry C, Aitchison T. Is homeopathy
a placebo response? Controlled trial of homeopathic potency, with pollen in hayfever as model. Lancet 1986; 2: 881–5.
14. Fisher P, Greenwood A, Huskisson EC, Turner P, Belon P. Effect of homeopathic treatment on fibrositis (primary fibromyalgia). Br Med J 1989; 299: 365–6.
15. Ferley JP, Smirou D, D’Adhemar D, Balducci F. A controlled evaluation of a homeopathic preparation in the treatment of influenza-like syndromes. Br J Clin Pharmacol 1989; 27: 329–35.
16. Shipley M, Berry H, Broster G, et al. Controlled trial of homeopathic treatment of osteoarthritis. Lancet 1983; 1: 97–8.
17. Jacobs J, Jiminez LM, Gloyd S, Gale J, Crothers D. Treatment of acute childhood diarrhea with homeopathic medicine: a randomized clinical trial in Nicaragua. Pediatrics 1994; 93: 719–25.
18. Reilly D, Taylor MA, Beattie NGM, et al. Is evidence for homeopathy
reproducible? Lancet 1994; 344: 1601–6.
19. Lökken P, Straumsheim PA, Tveiten D, Skjelbred P, Borchgrevink CF. Effect of homeopathy
on pain and other events after acute trauma: placebo controlled trial with bilateral oral surgery. Br Med J 1995; 310: 1439–42.
20. Weiser M, Strosser W, Klein P. Homeopathic vs.
conventional treatment of vertigo: a randomized double-blind controlled clinical study. Arch Otolaryngol Head Neck Surg 1998; 124: 879–85.
21. Linde K, Clausius N, Ramirez G, et al. Are the clinical effects of homeopathy
placebo effects? A meta-analysis of placebo-controlled trials. Lancet 1997; 350: 834–43.
22. Houghton HC. Catarrhal inflammation of the middle ear (93 cases). Homeopathic Times 1875; 1: 24.
23. Neustaedter R. Management of otitis media with effusion in homeopathic practice. J Am Inst Homeopathy
1986; 79: 133–40.
24. Messer S. Homeopathic treatment of pediatric otitis media. J Am Inst Homeopathy
1987; 80: 15–21.
25. Friese KH, Kruse S, Moeller H. Acute otitis media
in children: comparison between conventional and homeopathic therapy. HNO 1996; 44: 462–6.
26. Barnett ED, Levatin JL, Chapman EH, et al. Challenges of evaluating homeopathic treatment of acute otitis media
. Pediatr Infect Dis J 2000; 19: 273–5.
27. Jacobs J, Chapman EH, Crothers D. Patient characteristics and practice patterns of physicians using homeopathy
. Arch Fam Med 1998; 7: 537–40.
28. Yankelowitz S, Gravel J, Wallace I, et al. A clinical research form for use in the documentation of middle ear effusion. Ear Hearing 1991; 12: 296–8.
29. Homeopathic pharmacopoeia of the United States. Washington, DC: The Homeopathic Pharmacopoeia Convention of the United States, 1988.
30. Burke P, Bain J, Robinson D, Dunleavy J. Acute red ear in children: controlled trial of non-antibiotic treatment in general practice. BMJ 1991; 303: 558–62.
31. van Buchem FL, Peeters MF, van’t Hof MA. Acute otitis media
: a new treatment strategy. Br Med J 1985; 290: 1033–7.
32. Hoberman A, Paradise JL, Burch DJ, et al. Equivalent efficacy and reduced occurrence of diarrhea from a new formulation of amoxicillin/clavulanate potassium (Augmentin) for treatment of acute otitis media
in children. Pediatr Infect Dis J 1997; 16: 463–70.