Ear Infection: A Retrospective Study Examining Improvement from Chiropractic Care and Analyzing for Influencing Factors

INTRODUCTION

Objective: The aims of this study were to determine (a) if the patients improved while under chiropractic care: (b) how many treatments were needed to reach improvement; and (c) which factors were associated with early improvement.

Design: Cohort, nonrandomized retrospective study.

Setting: Private chiropractic practice in a Minneapolis suburb.

Participants: Forty-six children aged 5 yr and under.

Intervention: All treatments were done by a single chiropractor, who adjusted the subluxations found and paid particular attention to the cervical vertebrae and occiput. Sacral Occipital Technique-style pelvic blocking and the doctor's own modified applied kinesiology were also used. Typical treatment regimen was three treatments per week for 1 wk, then two treatments per week for 1 wk, then one treatment per week. However, treatment regimen was terminated when there was improvement.

Outcome Measure: Improvement was based on parental decision (they stated that the child had no fever, no signs of ear pain, and was totally asymptomatic), and/or the child seemed to be asymptomatic to the treating DC and/or the parent stated that the child's MD judged the child to be improved. A data abstraction form was used to determine number of treatments used and presence of factors possibly associated with early improvement.

Results: 93% of all episodes improved, 75% in 10 days or fewer and 43% with only one or two treatments. Young age, no history of antibiotic use, initial episode (vs. recurrent) and designation of an episode as discomfort rather than ear infection were factors associated with improvement with the fewest treatments.

Conclusion: Although there were several limitations to this study (mostly because of its retrospection but also, significantly, because very little data was found regarding the natural course of ear infections), this study's data indicate that limitation of medical intervention and the addition of chiropractic care may decrease the symptoms of ear infection in young children. (J Manipulative Physiol Ther 1996: 19:169-77).

Key Indexing Terms: Otitis Media, Chiropractic, Antibiotics, Earache.

By the age of 3, more than two thirds of all children have had one or more episodes of acute otitis media, including 33% who have had three or more episodes ( 1). Although quantifiable objective signs of otitis media are not usually available for diagnosis and the exact definition of otitis media varies among health practitioners, the signs and symptoms of ear pain, lever, swollen neck lymph glands and red eardrum are commonly attributed to otitis media ( 2).

In 1985, more than $100 million was spent in the U.S. for prescriptive antibiotics for acute otitis media ( 2): currently, $2 million per year is spent on its diagnosis and treatment ( 3). Prescribed treatment is typically a course of antibiotics (there is much world-wide variation regarding the type of antibiotic and the length of treatment) and, not infrequently, tympanostomy tubes or myringotomy ( 2).

Problems with antibiotic use include cost, allergic reactions and cross-reactions, GI upset, destruction of the gut's natural flora leading to yeast proliferation (one example being thrush) and the increased resistance of pathogens to antibiotics, thereby necessitating stronger antibiotics and the costly search for new and more effective antibiotics.

Tympanostomy tubes, used to prevent recurrence, mastoiditis and hearing loss, are frequently used to treat chronic otitis media ( 4, 5). However, almost all (98%) of the children undergoing the procedure have a recurrence of effusion within two months ( 6, 7). Twenty-five percent of tympanostomy tube patients have total hearing loss 7-10 yr later ( 8). Also, there are anesthetic risks, postsurgical infections and psychological trauma ( 1).

Complications of acute otitis media are few but not unimportant. Hearing loss, important because it relates to language comprehension, development of speech, and social interaction, is a frequent concern as a sequela of acute otitis media. Hearing loss may be caused by decreased sound conduction or decreased nerve transmission. Conduction loss may occur when the eardrum is immobile, either because of fluid in the middle ear, infection and/or swelling of the Eustachian tube, which prevents its normal opening and closing. If the Eustachian tube cannot open and close properly, pressure on both sides of the eardrum is not equal and therefore the eardrum cannot vibrate properly. Decreased eardrum vibration causes decreased ear ossicle movement and decreased stimulation to cranial nerve VIII. Although it is less common, the infection of chronic otitis media may also directly damage the inner ear structures, causing decreased nerve transmission.

Studies that examined complications of ear infections and their possible association with type of treatment and also included a control group were difficult to find. In one of them, a double-blind study of 171 children (239 affected ears) diagnosed with acute otitis media, patients were treated by four different methods: neither antibiotics nor myringotomy; myringotomy only; antibiotics only or both antibiotics and myringotomy ( 9). There were no significant differences among the four groups with respect to otoscopic findings, fever, relapses, complications and hearing loss. After 1 month, approximately 31% of all treatment groups had substantial hearing loss and after 2 months, 19% still had substantial hearing loss. Therefore, hearing loss may be caused by the ear infection itself, the treatment for the ear infection or some other unknown factor.

Mastoiditis is often cited as a complication of acute otitis media. The incidence of mastoiditis in patients treated with antibiotics is very low; however, the incidence of mastoiditis in patients treated without antibiotics is also very similarly low, about 0.2-2% ( 10, 11).

There has been much anecdotal evidence but very little research regarding chiropractic treatment for acute otitis media. Because the Eustachian tubes' cilia are so important in transporting the constantly-accumulating exudates out of the middle ear, chiropractic treatment is based upon normalizing Eustachian tube and immune system functions ( 12). Eustachian tube function depends upon appropriate contraction of the tensor veli palatini muscle, which in turn depends upon proper function of the mandibular branch of cranial nerve V, which in turn is dependent upon proper alignment and movement of the occiput and upper cervical vertebrae as the nucleus of the spinal trigeminal tract extends down to the level of the upper cervical segments. Eustachian tube function may also be impaired mechanically if there is swelling of the neck lymphatic tissue ( 12, 13). The immune system is influenced by many things, including spasm of the muscles surrounding the lymph tissue in the neck ( 12). Muscle spasm frequently accompanies vertebral subluxation ( 14).

This retrospective study examines the role of chiropractic treatment in the care of acute otitis media. The aims of this study were to determine (a) if the patients improved while under chiropractic care, (b) how many treatments were needed to reach improvement and (c) which factors were associated with early improvement.

The specific research questions addressed by this study were:

Is chiropractic care associated with clinical improvement in the course of ear discomfort/infection?
Is there an association between age and treatment outcome? Although three studies determined the age distribution of otitis media, studies showing a correlation between age and treatment outcome were not found ( 2, 9, 15).
Is there a prevalence of one sex presenting with ear discomfort/infection and is one sex more likely to show early improvement? vanBuchem, Dunk, and van't Hof determined from their study that the frequency of acute otitis media did not differ significantly between the sexes ( 9). Four other studies found that male gender was associated with increased risk for acute otitis media ( 2, 15-17). None of the studies addressed whether one sex improved faster with or without treatment.
Is a past history of antibiotic use associated with a different treatment outcome than no history of antibiotic use? Although two studies addressed the question of whether antibiotics improved the course of the illness, no studies were found that correlated past history of antibiotic use with treatment outcome of current episodes ( 2, 9).
Is there a difference in treatment outcome between initial vs. recurrent episodes? No studies were found that addressed this. Initial episodes might be expected to fare better than recurrent episodes on the basis that this group of children would be less likely to be inherently predisposed to ear infection than children who had a history of ear infections; recurrent episodes might be expected to fare better, if the patient received chiropractic care for all episodes of ear discomfort/infection, on the basis that Eustachian tube and immune system functions were becoming increasingly normalized. However, because this study was retrospective, there was no control over whether there was treatment, or the type of treatment, for other episodes of ear discomfort/ infection.
Did those episodes designated as ear discomfort improve with fewer treatments than those episodes designated as ear infection? The expectation is that the ear discomfort episodes would typically be less severe and therefore would be more likely to improve with fewer treatments.
METHODS
Enrollment
Charts of all children 0-5 yr old treated in a private practice in a Minneapolis suburb for ear discomfort or ear infection from August 1987 through November 1992 (with most of the cases from 1990-1992) were examined. Data recorded from each chart included: age, sex, history of antibiotic use, history of tympanostomy tubes, number of previous episodes of ear infection/discomfort, ear discomfort vs. ear infection, treatment dates, whether antibiotics were used during this episode and result (designated as worse, unchanged, uncertain if improvement, or improved). Recurrent episodes treated were noted, and the information was updated each time regarding whether antibiotics were used during a particular episode, treatment dates and result. Two patients (four episodes) were excluded because of insufficient data. Forty-six patients remained, with a total of ninety-five episodes (including recurrences). A data abstraction form developed for this study was used (Appendix 1).

Because the subjects in this study constituted a sample of convenience, there is no way to know if the sample population was representative of the target population, and no inference regarding extrapolation to the target or general population may be drawn.

Definitions
Ear discomfort/ear infection. Although the signs and symptoms of acute otitis media (ear pain, fever, swollen neck lymph glands, possible GI distress, red and bulging eardrum, possibly with purulent discharge) are well-known, clinicians do not agree on the specific criteria for diagnosis ( 2, 18). Myringotomy for culture is rarely done, and in one study, over 30% of specimens from ears of children who had clinically diagnosed acute otitis media were sterile ( 19). For this study, both ear discomfort and ear infection were defined as ear pain, probable fever and probable swollen neck lymph glands, with ear infections designated additionally by a positive otoscopic exam by the treating DC or MD (or MD's nurse). Diagnostic certainty in one large study was 58% in children aged 0-12 months and up to 73% in children ? 31 months old ( 2). It seems that the two designations are along a continuum; therefore, patients with ear discomfort were not separated from patients with ear infection except in one correlation.

History, of antibiotic use. Considered positive if use occurred at any time in the child's life.

History of tympanostomy tubes. Considered positive if use occurred at any time in the child's life.

History, of previous episodes. This is as judged by the parents (or as told to the parents by a health practitioner) and noted to be zero, one, two or more previous episodes of similar signs/symptoms.

Recurrence. This is another episode of ear discomfort or infection after the child had a symptom-free period and at least 1 month after the previous episode ends. Recurrence was only known if the patient returned for treatment.

Improved. This is based on parental decision (they stated that the child had no fever, no signs of ear pain and was totally asymptomatic) and/or the child seemed to be asymptomatic (no fever, no signs of ear pain, no swollen neck lymph glands and, if done initially, a negative otoscopic exam) and/or the parent stated that the child's MD judged the child to be improved. Improvement after only one visit was determined by a follow-up phone call to the child's parent(s) or by questioning the parent(s) at a later visit.

Treatment
All treatments were performed in a private practice in Minnesota. Treatment consisted of adjusting the subluxations found (usually with an Activator), paying particular attention to the cervical vertebrae and occiput. Blocking and modified applied kinesiology were also used. Typical treatment regimen was three treatments per week for 1 wk, then two treatments per week for 1 wk, then one treatment per week. However, the treatment regimen was terminated when there was improvement.

Data Analysis
Data were coded and entered using SPSS for Windows, Version 6.0. Data were analyzed using X² to detect association between number of treatments and the independent variables of: age, sex, history of antibiotic use, prior history of ear discomfort/infection episodes and classification of episodes as ear infection vs. ear discomfort. The critical level of significance was .05.

RESULTS
There were 46 patients with 95 episodes of ear discomfort/ infection. All were aged 5 yr or younger, with the majority, 82% (78), aged 2 yr or younger. Forty-five percent were girls and 55% were boys. Sixty-six percent (62) had a history of antibiotic use, although only 3 of the 95 episodes had antibiotics during the course of chiropractic treatment. The majority, 61% (58), had a history of two or more prior ear discomfort/ infection episodes, whereas 24% (23) had a history of 1 prior episode and 15% ( 14) had no prior history of ear discomfort/ infection. Sixty percent were deemed to have ear discomfort and 40% were diagnosed with ear infection. Only three of the patients had a history of tympanostomy tubes.

Treatment outcome, based on clinical signs and symptoms regarding ear pain, fever, neck lymph gland swelling and possible otoscopic exam, was determined to be improved, uncertain if improved, unchanged or worse. Because the great majority, 93% (88) of the 95 episodes, were determined to be improved and because this study focused on analyzing the number of treatments to reach improvement and which factors were associated with early improvement, only those 88 episodes that improved were analyzed in depth.

The remaining seven episodes, in which it was uncertain if there was improvement, are discussed briefly to better understand the make-up of all of the episodes presenting. Of those seven episodes, characteristics were similar to all of the episodes with respect to age, history of antibiotic use, history of prior episodes and ear discomfort vs. ear infection. However, five of the seven episodes were in girls, and there were fewer treatments than overall; five of the episodes had one treatment only, one had two treatments, and one had three treatments.

Treatment outcome was divided into two groups: those episodes that improved with only one or two treatments (early improvement) and those episodes needing three or more treatments to reach improvement.

Treatment Outcome
In 93% (88) of the episodes there was improvement. In 7% ( 7) of the episodes, it was uncertain if there was improvement. None of the 95 episodes" results were judged to be worse or unchanged.

Regarding the treatment outcomes of the improved episodes, almost half, 43% 138), improved with only one or two treatments, whereas 57% (50) needed three or more treatments (Figure 1). Analyzing the overall treatment outcome more specifically, 27% (24) improved with one treatment only, whereas 66% (58) improved with one to three treatments. Most of the episodes, 75% (66), needed ? 10 days to improve. There was an average of 3.09 treatments per episode in those that improved, with a median of 2.50 treatments per improved episode.

Age Distribution and Treatment Outcome by Age
Age distribution of this study is roughly similar to that of the large Greater Boston study, which found a peak incidence at 6-12 months of age and a steadily declining incidence from the first yr to the seventh yr ( 15). In this study, 82% (78) of all of the episodes as well as 82% (72) of the improved episodes were in children aged 2 yr or younger with a declining incidence from the 0-<1-yr-old age category to the 3-5-yr-old age category.

The younger the children were, the more likely they were to have early improvement of their episodes of ear discomfort/ infection (Figure 2). However, when age was tested for association with number of treatments needed to reach improvement, no statistical significance was achieved (p = .41).

Statistical analysis by X² test did not show an association between age and number of treatments needed (p = .413: Table 1).

Sex Distribution and Treatment Outcome by Sex
Two large studies have shown that being male increases the likelihood of having otitis media ( 15, 17). In this study as well, slightly more boys than girls presented with ear discomfort/ infection: 55% (52) boys and 45% 143) girls. Eighty-eight percent (38) of all girls and 96% (50) of all boys improved.

Of the improved episodes, boys and girls were very similar with respect to the number of treatments needed to reach improvement. Forty-two percent ( 16) of the girls and 44% (22) of the boys only needed one or two treatments to reach improvement: 58% (22) of the girls and 56% (28) of the boys needed three or more treatments (Figure 3). The only difference between boys and girls in the treatment outcomes was that of the seven episodes in which it was uncertain if there was improvement, five occurred in girls.

Analyzing the data between sex of the patients and number of treatments needed did not show an association (p = .859) (Table 2).

Past History of Antibiotic Use and Treatment Outcome by Past History of Antibiotic Use
The object was to determine it there was an association between past history of antibiotic use and the number of treatments needed to reach improvement. Two thirds of all of the episodes that presented and also of the episodes that improved had a past history of antibiotic use. Past history of antibiotic use was clearly associated with needing more treatments to reach improvement.

Of the episodes in which there was a past history of antibiotic use, only 35% improved in one or two treatments, whereas 60%, of the episodes in which there was no past history of antibiotic use improved in one or two treatments (Figure 4).

Analysis by X² test did show a direct association between past history of antibiotic use and number of treatments needed for improvement (p = .025) (Table 3).

Prior History of Ear Discomfort/Infection and Treatment Outcome
Because some children may be inherently predisposed to ear infections based on anatomic, physiologic or immunologic factors, and because these same factors might hamper improvement, this study investigated the relationship between initial vs. recurrent episodes and compared the number of treatments needed to reach improvement for each group.

The greatest likelihood for early improvement was seen in those episodes that were first-time, whereas the least likelihood for early improvement was seen in those episodes occurring in children who had a history of two or more previous episodes of ear discomfort/infection.

With no history of prior ear discomfort/infection, 62% (8 of 13) of the episodes needed only one or two treatments to reach improvement and 38% ( 5) needed three or more treatments. With a history of one prior episode, 52% ( 11) needed only one or two treatments and 48% ( 10) needed three or more treatments. With a history of two or more prior episodes, 35% (19 of 54) needed only one or two treatments, whereas 35 episodes, 65%, needed three or more treatments to improve/Figure 5).

Because data analysis showed a .201 probability of association, there was not a significant level of association between prior history of ear discomfort/infection and number of treatments needed to reach improvement (Table 4).

Ear Discomfort vs. Ear Infection by Treatment Outcome
Although in this study ear discomfort and ear infection were regarded as essentially the same process, they generally differ in the degree to which signs and symptoms manifest. The expectation is that the episodes designated as ear discomfort would improve with fewer treatments than those episodes designated as ear infection.

Episodes designated as ear discomfort were three times as likely to improve with only one or two treatments as those episodes designated as ear infection: 60% (31) of the ear discomfort episodes improved with one or two treatments, whereas only 19% ( 7) of the ear infection episodes improved with as many treatments (Figure 6). Additional factors that may have influenced the number of treatments needed for improvement were analyzed. Neither age, sex nor prior history of ear discomfort/infection episodes differed significantly between ear discomfort and ear infection. Past history of antibiotic use (which in this study is associated with needing more treatments to reach improvement) was more prevalent in the episodes of ear infection than all of the improved episodes. Sixty-six percent of all of the improved episodes had a past history of antibiotic use, whereas 83% of the ear infection episodes had a past history of antibiotic use.

As expected, there was a significant association between ear infection episodes and number of treatments needed (p = .00018; Table 5).

Because only 3 of the 46 patients (6 episodes) had a history of tympanostomy tubes and in only 3 of the 95 episodes was there antibiotic use during the chiropractic care for the ear discomfort/infection, this data was not analyzed with respect to chiropractic care.

Regarding the six episodes in which there was a history of tympanostomy tubes, five improved: of those five, there was an average of 4.2 treatments per episode (there was an average of 3.09 treatments per episode of the 88 episodes that improved). For one episode, it was uncertain if there was improvement. Four had ear infection and two had ear discomfort.

In those three episodes in which the patient used antibiotics during the chiropractic care, all improved, with an average of 4.0 treatments per episode. Two were deemed to have ear infection and one had ear discomfort.

The treatment schedules were very similar overall and any variations were similarly distributed throughout the treatment outcomes. Some of the episodes did have a somewhat more concentrated treatment schedule: these episodes, however, did not have a greater percentage of early improvement than the other episodes.

DISCUSSION
Addressing the fourth research question (regarding past history of antibiotic use), the most striking finding of this study was the degree to which past history of antibiotic use was associated with increased number of treatments to reach improvement. Sixty percent of the episodes with no history of antibiotic use required only one or two treatments to improve, whereas only 35% of those episodes in which there was a past history of antibiotic use needed only one or two treatments. It should be noted that 20% of the episodes that occurred in children with no past history of antibiotic use were deemed to be ear infection, whereas 51% of the episodes that occurred in children with a past history of antibiotic use were deemed to be ear infection. The severity of their illness, then, that necessitated previous antibiotics, may have influenced the outcome, in that the ear infection episodes were less likely to improve with only one or two treatments than the ear discomfort episodes.

Regarding clinical improvement, the first specific research topic of this study, the great majority (93%) of the episodes of ear discomfort/infection improved. Two thirds of the episodes improved with only one to three treatments and virtually all did so without the standard medical treatment of antibiotics. Forty-three percent needed only one or two treatments to improve. Seventy-five percent of the episodes improved in 10 days or fewer.

Looking at the second question, regarding the age factor, age distribution in this study is one of a decline from ages 0 through 5, with 65% (57) of the improved episodes in patients aged 1 yr or younger. In comparing outcome of episodes in children by year from ages 0-3, the younger the children were, the more likely they were to get better in only one or two treatments. Fifty-five percent of the episodes in children aged 0-12 months and 31% of the episodes in 3-5-yr-olds needed only one or two treatments to improve.

This study had 20% more boys than girls presenting. Al though there were some slight differences between the sexes with respect to treatment outcome (more girls' episodes were uncertain if improved), sex of the patients was not a major factor in treatment outcome.

Research question number five, regarding whether there was a difference in treatment outcome between initial vs. recurrent episodes, was examined comparing no history of prior episodes with one and two or more prior episodes. The improved episodes in patients with no history of prior ear discomfort/ infection were much more likely to show early improvement; 62% of the initial episodes needed only one or two treatments, 52% of the episodes in children with a history of one prior episode needed only one or two treatments and only 35% of the episodes in children with a history of two or more prior episodes improved with only one or two treatments. Possibly the anatomic, physiologic and immunologic factors that predisposed the child to otitis media also hampered his or her recovery.

The sixth and last question of the study, whether ear discomfort episodes improved with fewer treatments than ear infection episodes, was pertinent to this study only. Ear discomfort and ear infection are defined identically in this study except that ear infections had a positive otoscopic exam by the treating DC or MD (or MD's nurse). Because diagnostic certainty of otoscopic exams is not high, especially in children less than 3 yr old ( 2), they were not performend routinely in this study. Therefore, many episodes designated as ear discomfort could potentially have been ear infections. The fact remains, however, that three times as many episodes of ear discomfort, 60%, needed only one or two treatments, as did episodes of ear infection; 19% of ear infection episodes needed only one or two treatments to improve. One factor that might have been influential was past history of antibiotic use. The ear discomfort episodes were less likely to have a prior history of antibiotic use than the ear infection episodes. Of the ear discomfort episodes, 54% had a history of antibiotic use, whereas 81% of the ear infection episodes had a history of antibiotic use.

In studying chiropractic treatment outcome, it is helpful to also examine and compare it with the typical U.S. medical treatment outcome and with the 'natural' course of acute otitis media. The typical U.S. medical treatment, given to 98% of the patients, is an 8-10 day course of antibiotics. Myringotomy is also frequently done ( 9).

Regarding the natural course of acute otitis media, only one study was found that had as its control group a nonpreselected sample that received symptomatic care only ( 9). This group, then, would roughly parallel the natural course of acute otitis media; the symptomatic care consisted of decongestant nose-drops and analgesics in the form of suppositories. This study of 171 children concluded that there was no significant difference in clinical course between those episodes receiving myringotomy and/or antibiotics and those episodes receiving symptomatic care only. Seven days after being seen, 90% of the control group was pain-free and 93% of the group receiving antibiotics was pain-free. All of the children in the study had no fever at 7 days. The outcomes of the patients in this chiropractic retrospective study were very similar to the above study's control group, which could be said to approximate the natural course of acute otitis media.

Although this study did not control for treatment by other health practitioners in the interim between initial and recurrent episodes nor before the beginning of this study, recurrent episodes that were treated for the second, third or fourth recurrence were 70% more likely to improve with only one or two treatments than the episodes that were treated initially in this study. This might suggest an ongoing process of normalizing Eustachian tube and immune system functions. The role of chiropractic may be greater in the prevention of ear infections than in the treatment of ear infections.

There were several limitations of this study, mostly because of its retrospective nature. The sample size was fairly small, such that percentages in some of the sub-groups were easily changed by a substantial amount by the results of only a few of the episodes. Although the objectivity of the diagnostic criteria is a problem in any study of otitis media in children, it is not to be discounted and, ideally, temperature and otoscopic exams would have been done on every patient. There was a similar objectivity problem with outcome measures; that is, the treating chiropractor usually determined improvement. In some cases, improvement was determined solely by a follow-up phone call to the child's parent(s). Lack of a control group makes it difficult to draw conclusions about the efficacy of chiropractic treatment.

The above criticisms are also the basis for recommendations for future studies. Larger sample size, more objective data with respect to diagnostic criteria and outcome measures and a control group are necessary for a more thorough investigation of chiropractic care of acute otitis media.

CONCLUSION
In a retrospective study of 46 children aged 5 yr or younger with 95 episodes of ear discomfort/infection, chiropractic treatment was used. Overall clinical improvement was studied and age, sex, history of antibiotic use, history of prior episodes and ear discomfort vs. ear infection were investigated to determine which, if any, were factors in improvement with the fewest treatments. Virtually all (93%) improved, 75% in 10 days or fewer and 43% with only one or two treatments. Young age, no history of antibiotic use, initial episode (vs. recurrent) and designation of an episode as discomfort rather than ear infection were factors associated with improvement with the fewest treatments; no history of antibiotic use and designation of an episode as discomfort rather than infection were statistically significant factors.

ACKNOWLEDGMENTS
The author would like to acknowledge the assistance of Bill Jose, Ph.D., for his input into the text, Kassem Kassak, Ph.D., for his input into statistical analysis and J. Joseph Abeler, D.C., for the initial data.

Paper submitted December 19, 1994; in revised form August 7, 1995.

Table 1. Age and treatment outcome: ?² analysis to detect significant association
1-2 ? 3
Age treatments treatments Row total

? 1 yr 16 13 29
55.2% 44.8% 33.0%

1 yr 11 17 28
39.3% 60.7% 31.8%

2 yr 6 9 15
40.0% 60.0% 17.0%

3-5 yr 5 11 16
31.3% 68.8% 18.2%

Column total 38 50 88
43.2% 56.8% 100%

?² Value DF Significance

Pearson 2.86293 3 0.41

Likelihood ratio 2.87518 3 0.41

Mantel-Haenszel 2.33365 1 0.13
test for linear
association

Minimum expected frequency: 6.477
Number of missing observations: 0
Table 2. Sex and treatment outcome: ?² analysis to detect significant association
1-2 ?3
Sex treatments treatments Row total

Girls 16 22 38
42.1% 57.9% 43.2%
Boys 22 28 50
44.0% 56.0% 56.8%
Column total 38 50 88
43.2% 56.8% 100%

?² Value DF Significance

Pearson .03159 1 .85893

Continuity correction .00000 1 1.00000

Likelihood ratio .03161 1 .85889

Mantel-Haenszel test
for linear association .03123 1 .85972

Minimum expected frequency: 16.409
Number of missing observations: 0
Table 3. Past history of antibiotic use and treatment outcome: ?² analysis to detect significant association
1-2 ? 3
History treatments treatments Row total

No past history of 18 12 30
antibiotic use 60.0% 40.0% 34.5%

Past history of 20 37 57
antibiotic use 35.1% 64.9% 65.5%

Column total 38 49 87
43.7% 56.3% 100%

?² Value DF Significance

Pearson 4.95864 1 .02596

Continuity correction 3.99766 1 .04556

Likelihood ratio 4.96175 1 .02591

Mantel-Haenszel test 4.90165 1 .02683
for linear association

Minimum expected frequency: 13.103
Number of missing observations: 1
Table 4. Prior history of ear discomfort/infection episode and treatment outcome: ?² analysis to detect significant association
1-2 ? 3
Episodes treatments treatments Row total

No prior episodes 8 5 13
61.5% 38.5% 14.8%

1 prior episode 11 10 21
52.4% 47.6% 23.8%

? 2 prior episodes 19 35 54
35.2% 64.8% 61.4%

Column total 38 50 88
43.2% 56.8% 100%

?² Value DF Significance

Pearson 3.20648 2 .20124

Likelihood ratio 3.20031 2 .20187

Mantel-Haenszel test 3.16403 1 .07528
for linear association

Minimum expected frequency: 5.514
Number of missing, observations: 0
Table 5. Designation of episodes as ear discomfort vs. ear injection and treatment outcome: ?² analysis to detect significant association
1-2 ? 3
Designation treatments treatments Row total

Ear discomfort 31 21 52
59.6% 40.4% 59.1%

Ear infection 7 29 36
19.4% 80.6% 40.9%

Column total 38 50 88
43.2% 56.8% 100.0%

?² Value DF Significance

Pearson 13.99133 1 .00018

Continuity correction 12.40194 1 .00043

Likelihood ratio 14.73277 1 .00012

Mantel-Haenszel test 13.83233 1 .00020
for linear association

Minimum expected frequency: 15.545
Number of missing observations: 0
Fig. 1 Percent of improved episodes by number of treatments to reach improvement (n = 88).
1-2 Treatments 43%

3 or more Treatments 57%
Fig. 2 Percent of improved episodes by age and treatment outcome. 0-<1 yr. n = 29:1 yr, n = 28:2 yr, n = 15; 3-5 yr, n = 16.
1-2 Treatments 3 or More Treatments

0-<1 year olds 55% 45%

1 year olds 39% 61%

2 year olds 40% 60%

3-5 year olds 31% 69%
Fig. 3 Percent of improved episodes by sex and treatment outcome. Girls, n = 38; boys, n = 50.
1-2 Treatments 3 or More Treatments

Females 42% 58%

Males 44% 56%
Fig. 4 Percent of improved episodes by past history of antibiotic use and treatment outcome. No history of antibiotic use. n = 30; history of antibiotic use, n = 57.
1-2 Treatments 3 or More Treatments

No past history 60% 40%

Past history 35% 65%
Fig. 5
Percent of improved episodes by history of ear discomfort/ infection episodes and treatment outcome. No history of discomfort/ infection, n = 13; one prior episode of discomfort/infection, n = 21. greater than two prior episodes of discomfort/infection, n = 54.

1-2 Treatments 3 or More Treatments

No prior episodes 62% 38%

One prior episodes 52% 48%

>= Two prior episodes 35% 65%
Fig. 6
Percent of improved episodes by designation as ear discomfort vs. ear infection by treatment outcome. Ear discomfort, n = 52; ear infection, n = 36.

1-2 Treatments 3 or More Treatments

Ear Discomfort 60% 40%

Ear Infection 19% 81%
REFERENCES
(1.) Schmidt MA. Childhood ear infections. Berkeley: North Atlantic Books and Homeopathic Editorial Services, 1990.

(2.) Froom J, Culpepper L, Grob P, et al. Diagnosis and antibiotic treatment of acute otitis media: a report from International Primary Care Network. Br Med J 1990: 300:582-6.

(3.) Bluestone CD. Update on antimicrobial therapy for otitis media and sinusitis in children. Cutis 1985; 36:7-12.

(4.) Stool SE. Myringotomy: an office procedure. Clin Pediatr 1968; 7:470-3.

(5.) Schwartz RH, Rodriguez WJ, Schwartz DM. Office myringotomy for acute otitis media: its value in preventing middle ear effusion. Laryngoscope 1981; 91:616-9.

(6.) Van Cauwenberg P. The long term results of the treatment with transtympanic ventilation tubes in children with chronic secretory otitis media. Int J Pediatr Otorhinolaryngol 1979; 1:109-16.

(7.) Mackinnon DM. The sequel to myringotomy for exudative otitis media. J Laryngol Otol 1971: 85:773-94.

(8.) Stangerup SE, Tos M. Etiologic role of suppurative otitis media in chronic secretory otitis. Am J Otol 1985: 6:126-31.

(9.) vanBuchem FL, Dunk JHM, van'T Hof MA. Therapy of acute otitis media: myringotomy, antibiotics or neither? Lancet 1981; Oct 24, 883-7.

(10.) Diamant M, Diamant, B. Abuse and timing of use of antibiotics in acute otitis media. Arch Otolaryngol 1974: 100:226-3.

(11.) Medical Research Council. Acute otitis media in general practice. Lancet 1957: 2:510-14.

(12.) Hendricks CL, Larkin-Thier SM. Otitis media in young children. Chiro J Chiro Res Stud Invest 1989: 2:9-13.

(13.) Harker LA. Middle ear disease from Eustachian tube malformation. Alaska Med 1972: 14:90-4.

(14.) Fysh PN. Upper respiratory infections in children a chiropractic approach to management. Int Rev Chiro 1990; Mar/Apr: 29-31.

(15.) Teele DW, Klein JO, Rosner B. Epidemiology of otitis media during the first seven years of life in children in greater Boston: a prospective, cohort study. J Infect Dis 1989; 160:83-94.

(16.) Stool S. Otitis media: update on a common, frustrating problem. Postgrad Med 1989; 85:40-53.

(17.) Bluestone CD. The ear. In: Behrman RE, Vaugh VC, eds. Nelson textbook of pediatrics. 12th ed. Philadelphia: WB Saunders, 1983:1022-31.

(18.) Hayden GF. Acute suppurative otitis media in children: diversity of clinical diagnostic criteria. Clin Pediatr 1981; 20:99-104.

(19.) Klein JO, Bluestone CD. Acute otitis media. Pediatr Infect Dis 1982: 1:66-73.

APPENDIX I
Retrospective Ear Study Clinical Data Report

Name:
Age:
Hx antibiotic use: yes no unc. Tubes: yes no uncertain
Hx no. previous episodes +0 +1 +2
Initial treatment
Treatment dates until improvement or amelioration
Antibiotics used this time: yes no
Result: worse unchanged improved uncertain

Recurrence
Antibiotics used this time: yes no
Treatment dates until improvement or amelioration
Result: worse unchanged improved uncertain

Recurrence
Antibiotics used this time: yes no
Treatment dates until improvement or amelioration
Result: worse unchanged improved uncertain

Recurrence
Antibiotics used this time: yes no
Treatment dates until improvement or amelioration
Result: worse unchanged improved uncertain
~~~~~~~~

By Rosann M. Froehle, D.C.

Private practice of chiropractic, Blair, Wisconsin. Submit reprint requests to: Rosann M. Froehle, D.C., Route 2, Box 152, Blair, WI 54616

Share this with your friends