A Statistical Study of Attention Deficit/ Hyperactivity Disorder, Attention Deficit Disorder, and Hyperactivity and Their Relation to Sleep-Disordered Breathing

Taking medication for ADHD increases the likelihood of SDB by more than 7 times and interrupted sleep by 6 times; mouth breathing nearly quintuples the risk of SDB , and snoring increases the likelihood by more than 3 times.⁶ These risk factors highlight the importance of a comprehensive assessment of children with ADHD to identify potential sleep issues that may contribute to their symptoms. Children with ADHD or sleep issues should be routinely evaluated for nighttime mouth breathing (NTMB). Early detection and correction of NTMB could potentially reduce the need for unnecessary ADHD medication. This proactive approach could help clinicians provide more effective, tailored treatment plans for these children.⁷

Sleep disturbances, indeed, are more common in children with ADHD. These disturbances can manifest in various ways, such as difficulty falling asleep, frequent awakenings, and restless sleep. The discrepancies between subjective reports from parents or children and objective measurements (like polysomnography) further complicate the understanding of sleep issues in this population.⁸

The purpose of this investigation is to aid clinicians in recognizing the interplay between sleep and ADHD. This understanding may lead to more holistic treatment approaches, as better sleep can enhance attention, behavior, and overall functioning and help reduce problems in school. The study highlights the need for clinicians to assess and manage sleep disturbances as part of an early comprehensive treatment plan for children with ADHD, which can improve the future quality of life of these patients.

Materials and Methods

This retrospective study used dental professionals from across the United States who were trained to diagnose and treat SDB. Parents of potential patients were asked to fill out a comprehensive sleep severity questionnaire comprised of several important symptoms of abnormal sleep,⁹ including frequent symptoms often present in children during the day and night while sleeping, along with the severity of each symptom. The degrees of severity consisted of categories from 0 to 5: 0 = not present, 1 = very mild, 2 = mild, 3 = moderate, 4 = pronounced, 5 = severe. The number of symptoms were added to their total severities, resulting in a sleep severity index,¹⁰ which can vary from 0 to 100 or slightly higher. These severity amounts indicate treatment recommendations for each severity grade from 2 through 5. A similar grading system was used for the severity of ADHD ranging from 0 to 10, as well as with “attention deficit” (ADD) and “hyperactivity” (H) when occurring alone, apart from ADHD, and varied from 1 to 5.

A similar back-up analysis to ADHD was used , called DLRS (“doesn’t listen, resists sitting still”), which is identical to ADD and H (P = .001) but is less threatening terminology to a parent who may be afraid that recognition of ADHD might categorize the child in the future. DLRS has the same severity scale as ADHD (2 to 10), and the two components of DLRS (ie, not listening and resisting sitting still) have the same range as ADD and H (1 to 5).

The total sample consisted of 1,129 cases and was subdivided into groups: (A) a longitudinal sample of 220 treatment cases using a 27-symptom questionnaire; patients were treated with a preformed removable appliance (Healthy Start® Habit Corrector [HSHC appliance], manufactured in the United States by Ortho-Tain, Inc., orthotain.com, distributed by Healthy Start, thehealthystart.com) (Figure 1) that is specifically designed to correct sleep problems and various oral habits; (B) a subsample of 445 cases used with style='font-family: "Times New Roman",serif'>the (A) group for statistical analyses; and (C) a separate sample of 464 children whose parents had no knowledge or interest in issues involving sleep problems. Both the (B) and (C) groups had 35-symptom questionnaires.

Suspected obstructive sleep apnea was analyzed in the same way as other symptoms; however, treatment was tested for 3 months, and if there was minimal or no correction during this period the patient was referred to his or her pediatrician for treatment. The justification for this procedure was that suspected apnea had a high treatment success of 100% in 70% of such cases. If the patient had favorable success in this form of treatment for apnea, it would be continued as long as further success occurred. Also, an apnea polysomnograph is very expensive, and because apnea is such a minor contributor to SDB with an occurrence rate between 1% to 5%,¹¹ this approach was deemed to be the most practical and conservative for the patient.

Statistical tests determined the associations of ADHD, ADD, and H with the other symptoms in the questionnaire. Males dominated ADHD with a 60% occurrence (P = .001), while ADD and H exhibited no gender preference. As a result, males and females were pooled for statistical tests. Statistics used in this study were the Pearson product-moment correlation analysis, t-test, and chi square. The significance level was set at P = .05.

Incidence

The incidence of ADHD in a non-biased sample of 464 patients with an age range of 2 to 17 years was 24.7%, while ADD occurring alone was 6.1% and H alone was 17.6% (Table 1, middle column). These 464 patients, who were only visiting the dentist for a dental checkup, were given a sleep severity questionnaire, and it was found that 92% of these children had a sleep severity index of 30.5 and the abnormal sleep symptoms were 7.7 per child.¹⁰ A sleep severity index of 30.5 is a grade 3, which is assessed as a moderate condition with a recommendation for treatment.¹⁰

A separate sample of 665 patients, aged 2 to 15.7 years, included those who had an interest in sleep or did not answer the question regarding their dental visit. Table 2 provides information on various rankings of severity (minor, moderate, and severe), their percentages, means, and gender data for ADHD, ADD, and H.

The DLRS analysis in the bias sample involved 665 cases (Table 1) with an incidence of 40%, and the same statistical tests were analyzed as with those of ADHD. DLRS was found to be 13% more accurate in its severity assessment than ADHD (P = .001), although both are correlated with each other (P = .001). These variations might be due to the resistance of parents answering questions regarding ADHD as opposed to DLRS, which is perceived as a less threatening condition to a parent.

Causes and Interactions Between SDB Symptoms

Because 78% of 27 frequent symptoms of SDB are correlated with ADHD (Table 3), and no one cause corrects 100% of symptoms all of the time, this might indicate that more than one cause is probably present. Two possible factors, NTMB and “wakes up at night” (WUAN), may be at least partial possible causes of ADHD.

NTMB occurs when the external pterygoid and masseter muscles relax while sleeping, which allows the mandible to be opened and posteriorly displaced. Because the tongue is connected to the genial tubercle of the mandibular body anteriorly, the base of the tongue also displaces and impinges on the anterior area of the oropharynx. The muscles of the oropharynx also relax while sleeping, allowing the oropharynx to be partially or even completely closed, which can significantly reduce the amount of oxygen intake. This reduction in blood oxygen can affect the neural (ie, brain function), immune, and endocrine systems, and possibly cause many symptoms of SDB. These three systems can have far-reaching effects on many diverse symptoms.

Another possible cause of SDB is WUAN, which might have an adverse effect on reparative sleep by interrupting REM and slow wave sleep patterns. WUAN is correlated with ADHD (P = .001) and DLRS (P = .001). It is more severe when NTMB is present (P = .001), while ADHD, ADD, and H are 15% more severe when NTMB is present (P = .001) (per unpublished research by the author [EOB]).

NTMB is 31.4% more influential on ADHD than WUAN (P = .001). There is no difference between the severity of ADD or H in relation to WUAN. WUAN on the other hand is correlated with H (P = .001) but not with ADD. It is possible that these symptoms of SDB with their influence on the endocrine system could affect various hormones that might directly influence many emotional, cognitive, growth, and behavioral problems in patients. In summary, both NTMB and WUAN seem to have an effect on the severity of ADHD, while WUAN does not seem to have an effect on the severity of ADD when occurring alone apart from ADHD.

Allergies, which can contribute to difficult nasal breathing and NTMB, may be present in various ways such as asthma, eczema, anaphylaxis, nasal congestion, hay fever, etc., and may be genetic or environmental. Allergies were present in 45% of individuals (Table 1, nonbiased sample) and correlated with ADHD (P = .001), difficult nasal breathing (P = .001), WUAN (P = .001), NTMB (P = .001), and 86% of 29 SDB symptoms. Allergies are more prevalent in males (P = .01) and in industrialized countries. Their significant relation to difficult nasal breathing (P = .001) and NTMB (P = .001) is particularly meaningful in that there is a significant difference (P = .001) between normal and difficult nasal breathers¹⁰ and the severity of NTMB present in each. This indicates that there is a strong relation between difficult nasal breathing and increased severity of NTMB, and because allergies are correlated with both difficult nasal breathing and NTMB the interrelations between these three symptoms becomes obvious.

NTMB is correlated with 90% of 29 SDB symptoms, and these correlations agree with the same ADHD correlations 73% of the time (Table 3). Of interest is that NTMB is not correlated with school problems, although ADHD is (P = .001) (Table 4). When NTMB is present with ADHD, however, school issues are increased by 60.9% compared to when NTMB is not present with ADHD (P = .001). Also, when ADHD is present, school problems increase by 6.4 times as compared to when ADHD is not present (P = .001).

Both WUAN and NTMB have an increased presence of 28% when associated with ADHD than when not (P = .001). Also, WUAN has a 19% greater severity when NTMB is present than when it is not (P = .25).

Diagnosis

A diagnosis of ADHD is often determined by asking questions about behavior and personality, and answers to these questions might indicate the presence of ADHD.¹² A simpler approach would be to ask the patient if he or she experiences ADD or H and what the severity of each is.⁹ DLRS is another comparable analysis that combines two similar symptoms style='font-family: "Times New Roman",serif'>(ie, “doesn’t listen” and “resists sitting still”) , and if the questions are answered honestly, both ADHD and DLRS will usually produce similar results.

A rating scale for ADHD and DLRS has severity ranges from 2 to 10 (Table 2). This resulting index is further divided into three categories for a treatment analysis (Table 2),¹⁰ namely minor (2 to 3 severity), moderate (4 to 6 severity), and severe (7 to 10 severity). A similar rating scale for ADD occurring alone and H occurring alone ranges from 1 to 5 and is also divided into three categories: minor (1 to 2 severity), moderate (3 severity), and severe (4 to 5 severity) (Table 2). All three severity categories are recommended for treatment, whether analyzed for ADHD or DLRS, since even a minor severity obtains a mean treatment success of 55.8% (Table 2).

To analyze ADD alone and H alone together with “doesn't listen and resists sitting still,” categories are similarly defined as minor (1 to 2 severity), moderate (3 severity), and severe (4 to 5 severity). Again, the least category of minor is recommended for treatment with correction of 50% for H and 81.3% for ADD (Table 2).

A further helpful analysis is when various symptoms are rated according to their occurrence frequency and the ranking of their correlation coefficients in relation to their association with ADHD. The ranking of each symptom’s position for their incidence and their correlation coefficient (added together) is used to rank each symptom in order of their importance. Such a ranking indicated that two of the four most influential symptoms are WUAN and NTMB. These two symptoms are very important since they represent reparative sleep and oropharyngeal restriction during sleep.

WUAN and NTMB are each correlated with 90% of 29 SDB symptoms and indicate no difference in their correlation coefficient amounts. An illustration of the most important symptoms according to this analysis are as follows:

1. Allergies ranks #4 for incidence and #9 for its correlation = a total of 13.

2. NTMB total rank = 17 (1 + 16)

3. Restless sleep = 17 (7 + 10)

4. WUAN = 19 (6 + 13)

5. Resists directions = 22 (18 + 4)

6. Nightmares = 23 (11 + 12)

7. Bruxism = 23 (5 + 18)

8. Daytime mouth breathing = 23 (3 + 20)

9. School problems = 24 (19 + 5)

10. Daytime sleepy = 25 (10 + 15) style='color: black'>.

Lists of the most influential symptoms, such as the one above, can be used to assess the seriousness of an ADHD problem. The 10 SDB symptoms listed above are both the most frequently occurring and the most highly correlated with ADHD.

Both WUAN and NTMB have an increased presence of 28% when associated with ADHD than when not (P = .001). Also, WUAN has a 19% greater severity when NTMB is present than when it is not (P = .025).

Treatment

Many procedures/methods are available for treating ADHD and can be generally categorized as oral appliances, medications, adenotonsillectomies, other surgical procedures, myofunctional therapy, continuous positive airway pressure (CPAP) appliances, and commercial and home aids for sleep. Some of these are corrective while others are aids for temporary relief. The HSHC appliance was used in a subsample of 220 children for a mean 6.4 months¹³; it was worn passively while sleeping. The mean correction of ADHD was 55% for the 49% who had changes, and its initial severity was reduced from 5.5 to 3.7 (P = .001). The mean percentage improvement for 27 symptoms of SDB using this appliance was 59%, varying from 91% for headaches in the morning to 37% for “hard time listening and frequently interrupts.” The appliance increased mandibular growth by 54% in 39 months over a control sample’s growth with only nighttime wear, possibly by increased blood oxygen by preventing mandibular posterior positioning during sleep.¹⁴

Compared to male patients, female patients were 68.6% more successful in treating ADHD (P = .025), 56.3% more successful in treating ADD (P = .025), and 11.2% more successful in treating H (non-significant). Traditionally, females are more cooperative than males,¹⁵ being 80% cooperative as compared to 60% for males, which is probably due to males being 2 years less mature than females from at least 6 to 18 years of age.^16,17

Discussion

The public generally associates ADHD with school issues, but not sleep problems. Recent research, however, strongly links ADHD with sleep-disordered breathing. The results of the present investigation found that ADHD was correlated with 78% of 27 of the most important symptoms of SDB, of which 70% are P = .01 and P = .001. Poor school performance is ranked No. 4 with the highest correlation coefficient, while the three symptoms ranked above school problems are “nervous habits and can’t sit still,” “frequently interrupts and doesn’t listen,” and “resists directions.” Obviously, “can’t sit still” and “doesn’t listen” are similar to H and ADD. Just below “school issues” in this ranking, not including suspected “obstructive apnea” and “bluish color in A.M.” due to their minimal incidence, are “difficult nasal breathing,” “allergies,” “restless sleep,” “ear infections,” “nightmares,” “WUAN,” “depression,” “daytime sleepy,” and “NTMB.” The severity of 26 symptoms were compared between NTMB and WUAN and 73% had similar severities while 20% favored NTMB for greater symptom severities (P = .001).

A practical way to predict the severity of ADHD is to combine the various symptom correlation rankings with ADHD (Table 3) and the ranking of their percentage of occurrence (Table 1). The five most influential symptoms are allergies at 45% incidence with a correlation of P = .001, NTMB at 45% and P = .001, restless sleep at 42% (P = .001), WUAN at 42% (P = .001), and resists directions at 23% (P = .001). Such a list can indicate the seriousness of various symptoms with ADHD. The next five symptoms on that list are nightmares, bruxism, daytime mouth breathing, school problems, and daytime sleepiness. All 10 of these symptoms are correlated with ADHD at P = .01 or P = .001, and all but one (school problems) fall above the mean of 25% for incidence.

These rankings attest to a symptom’s importance and the strong relation that these SDP symptoms have with ADHD. NTMB is the most frequently occurring symptom of SDB with 62% incidence, and even though ADHD ranks No. 16 with an incidence of 25%, the correlation of these two symptoms is significant (P = .001).

NTMB is not correlated with school issues, but when present with ADHD, school issues are 61% more severe than when ADHD has no NTMB presence (P = .025). It is possible that since NTMB is correlated with WUAN (P = .01) and WUAN is correlated with ADHD (P = .001), school issues might be affected due to this interaction.

When analyzing the mean treatment success with a removable appliance (Figure 1) designed to correct NTMB,¹³ use of the appliance corrects 66% of the condition (P = .001) and also corrects WUAN with a mean improvement of 55% (P = .001). ADHD is corrected with the same appliance with 42% improvement, school issues at 40%, and all at P = .001. Because NTMB is not correlated with school issues, there possibly are many interrelations between several different symptoms to achieve school changes.

Gender differences involved 52% of the 29 symptoms, and all were males except two (7%) that were females. In that males are 2 years behind females in maturity from about 6 to 18 years of age,^16,17 this probably exerts a severe impact on neural and behavioral development that could involve at least nine of 29 various symptoms. Regarding school issues, the problematic symptoms involved behavior (27% incidence), reading (23%), writing (18%), and spelling (15%), and all were male dominant, while math (14%) and science (3%) were more prevalent for females (P = .001).

Offering an alternative condition to ADHD can be useful when parents are hesitant to fill out data in a sleep questionnaire. DLRS is a substitute diagnostic for ADHD and is disguised with two symptoms: “interrupts and doesn’t listen” combined with “nervous habits and resists sitting still.” This diagnostic can substitute for ADHD since there are no severity differences between the two, there is no difference in their correlations with the other symptoms, and ADHD and DLRS correlations are in agreement 85% of the time with SDB symptoms. These two analyses are correlated (P = .001), which indicates that DLRS can substitute for ADHD with few differences in their conclusions.

Conclusion

Based on this retrospective study, a number of conclusions can be made. ADHD is present in 25% of the population, while ADD occurring alone is present in 6% and hyperactivity occurring alone is present in 18% (49% of the total population). ADHD is correlated with 78% of 27 frequently occurring SDB symptoms, ADD occurring alone is correlated with 37%, and hyperactivity occurring alone is correlated with 56%. The most frequently occurring symptoms of SDB are NTMB, snoring, daytime mouth breathing, allergies, bruxism, WUAN, restless sleep, and talks in sleep. Male dominance occurred in 20% of 30 symptoms, with no gender dominance in 80% of the other symptoms. The mean treatment success for ADHD was 37%, while ADD occurring alone was 65%; for hyperactivity it was 54% using an SDB modifying appliance at night. The 10 most importantly associated symptoms with ADHD, in descending order, based on the sum of their rankings of incidence plus their correlation coefficients are WUAN, NTMB, restless sleep, allergies, resists directions, daytime sleepiness, school problems, talks in sleep, nightmares, and difficult nasal breathing. Finally, ADHD does not appear to self-correct due to insignificant changes with increase in age.

DISCLOSURE

Dr. Bergersen is the innovator of the appliance used for treatment in this study and an advisor on the board of directors of the manufacturing company that distributes the appliance. He volunteers his time and takes no compensation for his role. Neither author has any commercial interest in the company, although they both have a family relationship with the owners and chief executive officer of the company.

ABOUT THE AUTHORS

Earl O. Bergersen, DDS, MSD
Former Assistant Professor for 25 years, Northwestern University Dental School, Graduate Orthodontic Department, Chicago, Illinois; Private Practice in Orthodontics (ret.), Winnetka, Illinois

Brooke Stevens-Green, BS, DDS
Private Practice, Pontiac, Michigan

AUTHOR CONTRIBUTIONS

Dr. Bergersen contributed to the data required, its interpretation, and the statistical analysis of the data, discussion, and conclusions. Dr. Stevens-Green collected and interpreted the data of the treatment sample (N = 220), researched the literature, and wrote the description and review of the literature section.