Background
Attention-Deficit/Hyperactivity Disorder (AD/HD) is considered to be one of the most frequently diagnosed disruptive behaviour disorders in childhood [
1‐
5], with world wide prevalence rates of 8-12% [
6]. The American prevalence rates range between 3-7% [
1], and 4-12% [
7]. The Australian prevalence rates show 11% of 6-17 year olds are diagnosed with this disorder [
8], where as the English and Welsh AD/HD prevalence rates find 5% of 6-16 year olds have the disorder [
9]. The
Diagnostic and Statistical Manual of Mental Disorders 4
th
Edition Text Revision (DSM-IV-TR) [
1], is the most widely used classification system for mental disorders [
10,
11]. The DSM-IV-TR characterises AD/HD as inappropriate, chronic levels of inattention, hyperactivity and impulsivity [
1]. These children continually experience difficulties in academic achievement, and behavioural control, and as a consequence, they have difficulty in establishing positive relationships with family, authority figures and their peers [
12‐
14]. As a result, much attention has been devoted to the development and evaluation of assessment and treatment for this disorder over the last fifty years [
2,
15‐
17]. The majority of the AD/HD literature is dedicated to the treatment of this disorder [
2,
15‐
18]. Most of this research can be found in the area of pharmacological therapies [
12,
16,
17], with less emphasis in psychotherapy and other psychosocial interventions [
19]. There is even less research in the area of AD/HD and complementary and alternative medicine (CAM) therapies [
20,
21].
Even though psychostimulants are the first line of therapy for paediatric AD/HD [
2,
12,
22,
23], the evidence reveals that up to 30% of these children do not show clinically significant outcomes, and others experience side-effects [
12,
24‐
28], and need to discontinue their medications [
5,
28]. For these children alternative strategies need to be considered and instigated [
5,
27,
29].
In general, parents seek CAM therapies for their children for various reasons, such as they "feel mainstream medicine has let them down" [[
30], p. 573], because a particular treatment was considered ineffective, fear of drug adverse effects and a need for more personal attention [
31,
32]. Furthermore, parents often prefer to try something 'natural' for their children [
20,
29,
30,
33].
It is obvious that parents with children diagnosed with AD/HD seek CAM therapies [
20,
34‐
38]. In fact, CAM therapies are sought more often by parents who have children with developmental and behavioural disorders such as AD/HD, than with any other condition [
20,
33,
34,
39]. Controversy over the safety and appropriateness of stimulant treatment has led to increased parental anxiety and the increased use of CAM therapies [
20,
31,
40]. Major concern regarding the side effect profile of stimulant medications [
29,
31,
34,
41‐
43], has been the main reason parents have turned to alternative therapies [
20,
34‐
36,
38,
42,
43]. Many parents and even teachers are receptive to, and have a preference for non-pharmacological or behavioural therapies for children with AD/HD [
44,
45]. In fact, parents and teachers show preferences for multidisciplinary approaches, which lead to reductions in medications [
44,
46,
47].
In different surveys conducted around the world, CAM use for AD/HD ranged from 12% in Florida USA [
37], 28% in Shaare Zedek, Israel [
36], 54% in Boston USA [
40], 64% in Perth Australia [
38], and 68% in Melbourne Australia [
34]. The American Academy of Paediatrics recognised the increasing use of CAM therapies in children and, as a result, assembled a Task Force on Complementary and Alternative Medicine in 2008 to address issues related to the use of CAM for this population [
31]. This task force found that chiropractic care is one of the most common CAM practices provided at the professional level [
31]. Other studies have also confirmed this finding [
32,
48,
49]. Up to 10% of the US population seek care from chiropractors for non-musculoskeletal conditions [
48,
50,
51]. Studies have confirmed that up to 14% of all chiropractic visits were for paediatric patients [
39,
52], and that chiropractors were the most common CAM providers visited by children and adolescents [
31,
52]. One study indicated that paediatric populations seek chiropractic care predominantly for non-musculoskeletal conditions or when asymptomatic [
53].
A survey conducted in the USA on the presenting complaints of paediatric patients (under 18 years of age) for chiropractic care found that parents consulted chiropractors for their children's musculoskeletal (MSK) and non-musculoskeletal (non-MSK) conditions in addition to wellness care [
53]. Of these paediatric chiropractic visits, 44% were for MSK conditions and 56% were for non-MSK conditions [
53]. In this USA survey, included in the list of the most common non-MSK conditions parents sought chiropractic care for their children was hyperactivity [
53].
A survey conducted in Australia of paediatric chiropractic care for children under 18 years of age found that parents (like their American counterparts) also sought care from chiropractors for their children's MSK and non-MSK complaints [
54]. Within the Australian survey, parents consulted chiropractors for their children's non-MSK conditions, and included in that list were irritability, behavioural problems, AD/HD, and learning difficulties [
54]. These two surveys have found that parents seek chiropractic care for their children's AD/HD, irritability, attentional and behavioural issues, as well as their learning disabilities from chiropractors, both in Australia, [
54] and the USA [
53].
Although figures appear low, parents are presenting to chiropractors with their children [
50,
53,
55], looking for alternative therapies for AD/HD [
42,
43]. Anecdotally it has been suggested that AD/HD may be managed by chiropractic care, however to date no systematic review on the safety and efficacy of chiropractic care for paediatric and adolescent AD/HD has been conducted. A systematic review conducted to determine whether evidence exists for the therapeutic application of manipulation for paediatric health for musculoskeletal and non-musculoskeletal conditions revealed only low levels of scientific evidence [
56]. In view of the large numbers of children and adolescents being diagnosed with AD/HD and the increased use of CAM therapies, of which chiropractic care is one of the most common, this review is relevant and important.
Discussion
An important result of this review is that the authors found that no studies met the inclusion criteria for this topic. The natural conclusion one draws from such a discovery, is that no evidence of studies for or against this treatment (chiropractic care) for this condition (paediatric and adolescent AD/HD) using RCTs (Level II evidence) were found. The reviewers then questioned whether or not their eligibility criteria were too strict or inappropriately defined [
89]. In fact, evidence at lower levels of the hierarchy of evidence, such as non-randomised, quasi-experimental group designs or single-subject experimental designs could exist and could contribute valuable information [
90]. The reviewers discovered that no RCTs existed on the subject matter and after discussion and reviewing the EPOC guidelines the eligibility criteria were extended to include Level III evidence (Table
2). Despite this extension of evidence to include Level III evidence the four intervention studies that were found did not meet the inclusion criteria (Refer to Table
4).
Researchers have used the term 'empty' review when a search to address a research question yields no eligible studies [
89,
90]. At first this may appear as though the review has no intrinsic value. However, knowing that there are no studies of a particular type on a specific topic has the potential to generate meaningful and useful information [
90]. For researchers, empty reviews serve the purpose of highlighting research gaps and directing future original research projects, as was the case for these authors. There was a gap in the knowledge that needed an answer to an important clinical question: "does chiropractic care have a role to play in the treatment and/or management of paediatric and adolescent AD/HD?"
The inclusion of a log of rejected trials is an important aspect of any systematic review [
90]. As part of the Cochrane review process a log of rejected trials is expected, outlining the studies that were excluded as well as listing the reasons for their exclusion [
91]. Table
4 outlines the rejected studies and the reasons they were rejected.
This 'empty review' allows for the opportunity to learn from the excluded studies. For instance: What were the predominant types of research designs used? What types of populations have been studied? Which types of chiropractic interventions have been tested? What types of outcome measures if any, were used?
According to this systematic review, 15 case studies have been published [
67‐
69,
71‐
74,
79,
82‐
88]; three case series [
70,
80,
81]; one single subject design study (n = 7) [
76]; two uncontrolled, non-random experimental trials (n = 41 and n = 13) [
75,
77]; and one controlled, non-random, experimental clinical trial (n = 24) [
78] for AD/HD and chiropractic care. Of these, two studies targeted adult AD/HD populations [
70,
75], three studies targeted paediatric and adolescent populations [
76‐
78]. It is obvious from this review that there is a paucity of studies on paediatric and adolescent AD/HD and that the most predominant type of research design is the case study.
As for the types of chiropractic interventions investigated it was not a homogeneous finding. The chiropractic profession has over one hundred different techniques [
59], and there was no shortage of variety in the studies found for this review. The following were some of the techniques investigated in the chiropractic and AD/HD literature: Diversified, Gonstead, Sacro-Occipital Technique (SOT), Craniosacral Therapy, Pettibon, Toggle Recoil Technique, Thompson Technique, Torque Release Technique, Network Spinal Analysis, Chiropractic Biophysics, and Activator Technique. As part of the interventions described in the published articles, advice on exercise and/or dietary modifications was also given in conjunction with some form of chiropractic treatment in seven of the studies reviewed [
67,
69‐
71,
79,
81,
86] (Refer to Additional files
4 and
5).
In regard to the outcome measures used in these studies very few chiropractors actually used validated psychometric measures, in fact only one paediatric study used a known psychometric measure i.e. Werry-Weiss-Peters Parent Rating Scales [
76]. However, according to Miller and colleagues this psychometric measure is best used when AD/HD is present with mental retardation [
92]. This study also used electrodermal activity of skin conductance, and cervical x-rays [
76]. The only other studies that used a psychometric outcome measures were the two adult AD/HD studies. One study used the Test of Variables of Attention (TOVA) [
70] and the other used the Conners' Continuous Performance Test (CCPT) [
75]. When reviewing the literature it is important to evaluate whether the patients (i.e. children and adolescents) presented to a chiropractor for treatment of traditional musculoskeletal conditions or whether they presented with a primary diagnosis of AD/HD. In every single case study the parents presented their child or adolescent to the chiropractor with a primary complaint of AD/HD, and chose to seek chiropractic care for their child's or adolescent's AD/HD symptoms. An interesting finding was that chiropractors used outcome measures that they would traditionally use for musculoskeletal conditions (i.e. x-rays, thermal scans, and surface electromyography) for AD/HD. These types of outcome measures are not used for AD/HD symptomatology in AD/HD studies published in the medical literature. One study used thermal scans with surface electromyography (sEMG) pre and post intervention as a measure of outcomes [
68]. Two studies used sEMG as outcome measures [
69,
70], and another two studies used paraspinal thermal scans [
67,
79]. Two studies used rating scales designed by the chiropractor rather than using established reliable and validated psychometric rating scales [
69,
78]. Furthermore, all of the studies used subjective statements of a child's improvement taken from parents and/or teachers, and even a bus driver [
67]. In all fairness many case studies presented were retrospective (although many were ambiguous) in nature and as a result it is highly probable that these chiropractors did not have any intentions of publishing and as a result did not seek out and use appropriate outcome measures for AD/HD symptomatology. However, it must be noted that even those few studies that were prospective in nature the chiropractors involved did not seek and use appropriate outcome measures.
When conducting research in the area of AD/HD a good guide to use is the "Practice Parameters for the Assessment and Treatment of Children and Adolescents with Attention-Deficit/Hyperactivity Disorder" [
2]. Choosing psychometric measures that are recommended by the American Academy of Child and Adolescent Psychiatry [
2] (Refer to Table
3), ensures that the outcome measures have normative values and are likely to yield a measure of AD/HD behaviours that are reliable.
For clinicians, an empty review provides valuable information showing that there is no evidence in support of a treatment on the basis of the inclusion criteria used in the review process [
89,
90]. Furthermore, empty reviews inform decision makers in health care when there is lack of robust evidence in favour of (or against) a particular health care intervention [
93]. As was found in this review, there is no robust evidence in favour of chiropractic care for paediatric and adolescent AD/HD. It is important that chiropractors seek out the best evidence available. However, the absence of RCTs in this area does not need to immobilize clinical decision making, nor does it necessarily justify the abandonment of an intervention [
90]. According to Sackett and colleagues [
94,
95], clinical expertise can be defined as "the proficiency and judgment that individual clinicians acquire through clinical experience and clinical practice" [[
94], p.71]. Responsible practitioners need to integrate this evidence with their clinical expertise and should apply a common sense approach to each individual patient. Furthermore, all health care providers have a responsibility to inform their patients when a particular intervention does not have scientific validation, and that all they have is clinical experience and anecdotal evidence to support their treatment strategy, which is in keeping within the scope of evidence based practices [
96].
If the chiropractic profession chooses to conduct research in the area of paediatric and adolescent AD/HD then appropriate study designs need to be followed. The gold standard for claiming a particular intervention caused the desired effect is the randomised controlled clinical trial (RCT). The CONSORT group recommendations are suggested to develop a stringent a set of guidelines designed to improve the reporting of RCTs [
97]. The CONSORT Group also developed an extension of the CONSORT Statement for non-pharmacologic treatments [
98], which can be easily applied to chiropractic intervention studies. If these guidelines are used in the design of a RCT then a robust study can be designed to minimise the risk of bias (internal validity) and to account for the applicability of a trial's outcomes to the target population (i.e. generalisability or external validity) [
99].
With the increase use of CAM therapies the CONSORT group have assessed the quality of randomised trials for paediatric CAM therapies. They found that only 40% of the CONSORT checklist items were included in the published articles [
100]. In order for these types of studies to be a valid source of information about paediatric CAM therapies, they need to be conducted and reported with the highest possible standards [
100]. Unfortunately, the searches for this systematic review did not uncover any RCTs for the use of chiropractic care in paediatric or adolescent AD/HD cohorts. Chiropractic researchers can learn from the CONSORT group in order to design, conduct and report trials that will be valid and applicable in the future.
Lastly, it is important for chiropractors and chiropractic researchers to report any risks, side-effects or adverse events in relation to chiropractic interventions. "Every healthcare intervention comes with risk, great or small, of harmful or adverse effects" [
91]. In all the studies reviewed for this systematic review there was not one mention of side effects or adverse reactions except for one study in which one adolescent girl reported feeling 'high' after her first adjustment [
81]. However, it can not be assumed that the determination of side-effects was a specific goals of any of the studies reviewed, as it was not explicitly stated. It is strongly recommended that future studies for these age groups should include side effect and adverse reaction data. According to the Cochrane review it is important to minimize bias when conducting reviews by including an evaluation of adverse effects [
91]. However, to date only one narrative report [
101], and one systematic review for paediatric spinal manipulation [
102], have been conducted reporting adverse events. Despite these, there are not enough data to evaluate causation or incidence rates of these rare adverse events. The importance of a prospective population-based active surveillance study has been recommended [
102], in order to assess the severity and frequency of adverse events as a result of chiropractic care within the paediatric population. It is recommended that clinicians who administer spinal manipulation to paediatric populations should inform the parents that spinal manipulations may cause rare but serious adverse events [
102].
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
FK, RB and HP conceived the research project. All authors contributed to the writing of the manuscript. All authors read and approved the final manuscript