The Complete Guide to Orthodontics: the classification map, the decision framework for treatment timing, and the retention phase

A direct answer in under 60 words

Malocclusion is common among children and adolescents ⁷; the timing of treatment varies with the type and the severity of the anomaly ¹⁴⁵. The retention phase is part of the course of care ³², and what is actually done has to be assessed by a dentist ⁴¹.

Scope: This article is general oral-health education based on international literature. It does not address any country's insurance or regulations; consult local rules for care pathways and costs. Every medical statement in this article is anchored to international peer-reviewed literature and cross-society clinical guidelines, and every fact is marked geo: universal; the statutes, insurance coverage and fee regulations of any one country lie outside the scope of this article — please see the corresponding local canonical card ⁴².

Where this article sits: what it answers and what it does not

The dental content system has two layers. One is the canonical card — one question per card — which answers a patient's concrete question, for example how much braces cost, or how long a palatal expander has to be worn. The other is the domain authority article, the layer this article belongs to, which handles the gaps between the cards: what this domain is actually dealing with, how the classification is drawn, what the criteria for timing are, which families of appliances there are, and how far the strength of the evidence reaches in each case ⁴⁰.

So the way this article is written is: it gives you the map and the framework, not a personalised answer. Any concrete question that belongs to a particular canonical card is covered here in a single sentence with a pointer to where it lives; the content of that card is not rewritten and its details are not expanded ⁴⁰. The section "Downstream links" at the end lists every canonical card in this domain.

Three things to state up front ⁴¹:

• Every classification and framework in this article is a structure for communication, not a diagnostic tool, and it is not any individual's treatment plan.
• This article quotes no price; the section on cost writes only about components and variables.
• This article makes no performance claim about any brand of orthodontic appliance and uses the terminology of the literature throughout (for example clear aligners); brand-related concrete questions are handled by the corresponding canonical cards ⁴⁵.

(The basis on which this topic was selected is an internal analysis of search data; that is an editorial matter and its details are not disclosed externally ⁴⁰.)

1. What orthodontics deals with: turning crooked teeth into problems that can be discussed

1-1 First the order of magnitude: how common malocclusion is

A global systematic review and meta-analysis that included 77 studies pooled the prevalence of malocclusion in children and adolescents at 56% (95% CI 11 to 99), with no difference by sex; by region, Africa 81%, Europe 72%, the Americas 53%, Asia 48% ⁷. This figure has to be read together with its confidence interval: the width of the interval, 11 to 99, shows that the studies differ enormously among themselves, so it is suited to grasping the order of magnitude of how common this is, and is not suited to being used as a risk estimate for any particular region or any particular individual ⁷.

The same analysis also records that, from the primary to the permanent dentition, overall prevalence did not change (54% in both); individual traits, however, do move — crossbite and diastema fall during the permanent dentition, while scissor-bite and dental crowding rise ⁸. That explains a common puzzle: things looked all right in childhood and yet feel more crowded later, and at the population level of the data those two observations do not contradict one another ⁸.

1-2 Three directions: sagittal, transverse, vertical

When the literature describes orthodontic anomalies, it separates out at least the different spatial directions. The German S3-level cross-society clinical guideline gives different timing recommendations for skeletal or dental class II anomalies, for skeletal or dental class III anomalies (antero-posterior, that is, sagittal) and for pronounced skeletal or dental transverse anomalies ⁴⁵⁶. And in the literature on maxillary expansion in adults, the sagittal and the vertical are two separate sets of indices, measured and reported apart from one another ²⁸.

Put differently, in professional description "crooked teeth" is not one problem but a set of positional relationships in at least three directions, and the logic of management and the recommendation on timing differ in each direction ⁴⁵⁶²⁸. That is also why two people who look alike may end up with completely different treatment plans ⁴¹.

1-3 Three levels of aetiology: skeletal, dental, functional

In the literature, one and the same appearance is separated into three aetiologies — skeletal, functional and dental — and handled separately ¹⁰. This is not a detail of classification but the fork in the road that decides the treatment approach ¹¹. This article does not undertake to define the mechanism of each aetiology; it only relays how the sources triage on that basis ⁴⁵.

The evidence shows this clearly. A meta-analysis including 21 randomised controlled trials and 854 children aged 6 to 12 years explicitly split anterior crossbite into three categories for handling: 18 studies on skeletal, 2 on functional, and 1 on dental ¹⁰. Its conclusions are likewise triaged by aetiology — intraoral non-skeletally anchored appliances suit dental and functional crossbites and mild skeletal discrepancies; for a true skeletal anterior crossbite, that analysis holds facemask therapy combined with rapid maxillary expansion (FM-RME) to be the more reliable way of achieving significant skeletal correction ¹¹.

The guideline side draws the same distinction: its timing recommendations are consistently written as "skeletal or dental class II" and "skeletal or dental class III", placing bone and tooth side by side within a single criterion ⁴⁵.

So before asking which appliance to use, one has to ask whether this is a problem of the bone or of the teeth — a question that requires clinical examination and imaging, is decided by a dentist, and cannot be answered here ⁴¹.

1-4 How common lay terms line up with the terminology of the literature

Everyday Chinese expressions and the terminology of the literature do not map one to one, so they are aligned first to avoid misreading ⁴⁵:

• Buck teeth (暴牙): the situation that commonly corresponds in the literature is class II malocclusion caused by mandibular retrusion, a situation often managed with functional appliances ¹⁷, while the subjects of the relevant comparative research were growing patients ²⁰. Note, however, that class II malocclusion is not always caused by mandibular retrusion; the two must not be equated ¹⁷.
• Underbite (戽斗, 地包天): corresponds to skeletal class III malocclusion and to anterior crossbite, which sit under different headings in the literature but whose subject matter overlaps ¹⁴⁹.
• Dental crowding (齒列擁擠): listed in the global prevalence analysis as one of the traits that rise during the permanent dentition ⁸.

From here on this article uses the terminology of the literature throughout, with the Chinese equivalent appended where that helps ⁴⁵.

2. Treatment timing: why there is no single answer to when to start

2-1 In the literature this question is itself contested

What the German S3-level guideline states at the outset is this: the ideal timing of orthodontic treatment is discussed controversially depending on the type and the extent of the anomaly present, particularly with regard to efficiency, patient burden and treatment effort ¹. That is the key background to this section — at what age to start is not a question with a standard answer but a judgement that varies with the classification ¹.

The methodological scale of that guideline is worth noting, because it determines how much weight these recommendations carry: 309 eligible studies were identified out of more than 11,000 screened sources and critically appraised one by one, and representatives of 21 German scientific societies and organisations agreed on 19 evidence-based statements and recommendations through a nominal consensus process ².

2-2 How the guideline triages

• General principle: most malocclusions can be effectively treated in the early mixed, the late mixed and the permanent dentition alike ³. That sentence anchors the whole section — it rules out in advance the notion that treatment becomes impossible once a certain age has been missed ³.
• Class II (which includes what is colloquially called buck teeth): therapy of a pronounced skeletal or dental class II anomaly can be started early in order to reduce the risk of anterior tooth trauma, whereas for a moderate class II anomaly it is preferably carried out before or during the pubertal growth peak ⁴.
• Class III (which includes what is colloquially called an underbite): therapy of a skeletal or dental class III anomaly should be started early, as this also reduces the need for later surgery to correct the anomaly ⁵.
• Transverse anomalies (the common indication setting for maxillary expansion): a pronounced skeletal or dental transverse anomaly should be started on early in the upper jaw, in order to make use of the high adaptivity of the maxillary structures in young patients ⁶.

The meta-analysis on anterior crossbite in children is consistent with this: early orthodontic treatment is effective in growing children, and the outcome is to a large extent influenced by the underlying aetiology and by the type of appliance ¹¹. The same paper also cautions that untreated anterior crossbite "may" lead to occlusal dysfunction, impaired craniofacial growth, and long-term periodontal or temporomandibular complications — that sentence comes from the background narrative of the study and is phrased as a possibility; it is not a statistical result of that study ⁹.

2-3 And adults: can bone still move?

On the skeletal response in adults and in late adolescence the evidence speaks more cautiously. An analysis including 11 studies, 7 of which entered the meta-analysis, assessed the skeletal effects of miniscrew-assisted rapid palatal expansion (MARPE) in late adolescents and adults: in the sagittal dimension SNA increased by 0.79 degrees and ANB by 1.18 degrees, reaching statistical significance; the vertical indices remained relatively stable and did not reach statistical significance ²⁸. But the authors are very direct about what these changes mean clinically — the magnitude is relatively small, and the impact on maxillary retrusion and mandibular protrusion is clinically modest ²⁹.

Another related clue comes from the comparative research on root resorption: its inclusion criterion was restricted to patients older than 14 years with fully developed anterior roots ²⁷. Note that this criterion states only root development; that source does not describe skeletal development. And "older than 14 years" still covers many adolescents who are in their growth phase, so it is an inclusion threshold defined by a root condition, and is not equivalent to a definition of an adult ²⁷.

So the framework for adult orthodontics is this: in late adolescents and adults, changes in the skeletal indices may reach statistical significance, but the authors assess their clinical impact as modest ²⁸²⁹; what an individual adult case can achieve has to be assessed by a dentist against the imaging and the clinical conditions ⁴¹.

3. The appliance map: four families and how far the evidence reaches for each

3-1 Fixed appliances

In the two comparative studies set out below, fixed appliances are the family used as the comparator ²⁶²³. In the comparative research on root resorption they are the control group ²⁶; in the systematic review on clear aligners they are again the conventional comparator ²³. Their known costs are set out in the risk section further down — root resorption ²⁶ and white spot lesions ³⁸ were both quantified in the treatment context of this family.

3-2 Functional appliances (class II in growing patients)

Class II malocclusion caused by mandibular retrusion is commonly managed with functional appliances, yet the comparative effectiveness of removable versus fixed modalities remains uncertain ¹⁷. A systematic review including 9 comparative clinical studies with 377 participants in total (188 using the Forsus fatigue-resistant device, 189 using the Twin Block) notes that both families improve overjet, overbite and molar relationship, but that the route by which correction is achieved differs — the skeletal contribution of the Twin Block is stronger, whereas the Forsus achieves correction predominantly through compensation at the dentoalveolar level (proclination of the lower incisors) ¹⁸.

This contrast is itself a decision framework: for the same goal of improving buck teeth, one family leans towards changing the skeletal relationship and the other towards changing tooth position, and the settings each suits and the subsequent costs each carries are different ¹⁸. The same review also notes that the included studies are predominantly of non-randomised design ¹⁸.

3-3 Clear aligners (invisible braces)

The conclusions in this family have to be read together with the level of the evidence ⁴¹.

A systematic review searching from January 2015 to March 2025 included 15 papers out of 196 screened and 55 read in full text; 13 of them were non-randomised studies and 2 were randomised trials. Assessed with ROBINS-I, 3 were at low, 3 at moderate and 7 at high risk of bias, while the 2 randomised trials were rated low risk under RoB-2 ²². Its conclusion is that, relative to conventional fixed appliances, clear aligners perform better on dental aesthetics and periodontal health and carry potential benefits in treatment length and patient comfort ²³. Note the word "potential" in the wording of that conclusion, and note that 6 of the included studies were long-term follow-ups without a control group ²⁴.

On the complication side the quantification is clearer. A meta-analysis measuring with cone-beam computed tomography included 4 high-quality cohort studies with a total of 1,204 anterior teeth: root resorption in the clear aligner group was significantly lower than in the fixed appliance group (weighted mean difference -0.47 mm, 95% CI -0.55 to -0.40; versus -0.54 mm, 95% CI -0.75 to -0.33) ²⁶; but the authors also write plainly that the absolute difference is modest, and that its clinical meaning lies in aligners possibly suiting patients at higher risk of root resorption ²⁷.

How this set of figures is to be read has to be marked clearly: the two weighted mean differences above are the amount of root resorption in each group separately, not a between-group difference; their confidence intervals overlap substantially, and the source does not report their difference as an effect estimate ²⁶. The phrase "significantly lower" and the p value are that source's own narrative, relayed here as written, with the reminder that these two figures by themselves are not sufficient to establish the statistical significance of a between-group difference ²⁶. Beyond that, the subjects included in that analysis were limited to patients older than 14 years with fully developed anterior roots, and the result may not be extrapolated to children whose roots are not yet fully developed ²⁷.

As for the mandibular advancement version for class II in growing patients (CAMA), the literature positions it as an aesthetic alternative for correcting class II malocclusion, and its efficacy relative to traditional functional appliances remains debated ¹⁹. A meta-analysis including 9 studies (1 randomised trial plus 8 non-randomised studies) and 465 participants shows that, compared with the traditional Twin Block, there was no statistically significant difference in skeletal angular change or in the vertical dimension; the difference in the amount of overjet reduction reached statistical significance but was clinically minimal (mean difference -0.46 mm); and lower incisor proclination was markedly less (mean difference -0.90 degrees) ²⁰. The authors' conclusion is written very conservatively: the available evidence is predominantly retrospective, these findings should be interpreted with caution, and they need confirmation by future rigorous randomised controlled trials ²¹.

3-4 Expansion and traction appliances (including the palatal expander)

• Facemask therapy combined with rapid maxillary expansion (FM-RME): in the meta-analysis on anterior crossbite in children it was rated the more reliable way of achieving significant skeletal correction in truly skeletal cases, with pooled estimates showing ANB increased by 3.54 degrees, SNA by 1.37 degrees and SNB reduced by 2.14 degrees, although heterogeneity between studies was substantial ¹¹.
• Skeletally anchored facemask: compared with the conventional tooth-borne facemask, it did not demonstrate a clinically meaningful advantage in overall sagittal correction (ANB mean difference 0.07 degrees), although the same analysis also recorded a small additional improvement in maxillary advancement (SNA mean difference 0.59 degrees) ¹². How to read this entry: within this particular comparison, adding skeletal anchorage did not bring a clinically meaningful advantage in overall sagittal correction, and the extra benefit was seen only in the single index of maxillary advancement; that result is limited to the setting of facemask treatment for anterior crossbite and may not be extrapolated into a claim that no skeletally anchored appliance carries any additional benefit ¹².
• Intraoral non-skeletally anchored class III appliances: extra-oral appliances are bulky, unaesthetic and heavily reliant on compliance, which is why intraoral alternatives have attracted attention ¹⁴. A review including 8 randomised trials with 336 growing patients, of which only 2 were suitable for quantitative pooling, shows that lower clear plate intermaxillary traction (LCP-IMT) and the reverse Twin Block performed more visibly on skeletal improvement in the sagittal dimension ¹⁵; but the overall strength of the evidence lies between low and very low, because of methodological limitations, small sample sizes and heterogeneity between studies ¹⁶.
• Miniscrew-assisted rapid palatal expansion (MARPE): the population it suits is late adolescents and adults; the skeletal changes reach statistical significance but their clinical magnitude is modest ²⁸²⁹.

3-5 An evidence-strength map (usable when reading any orthodontic information)

Placing the authors' own assessments from the sections above side by side yields a very useful reading habit ⁴¹:

• Anterior crossbite in children: overall certainty of the evidence ranged from very low to moderate ¹³.
• Intraoral class III appliances: low to very low ¹⁶.
• Efficacy of clear aligners: 13 of the 15 included papers were non-randomised, and 7 were at high risk of bias ²².
• Clear aligner mandibular advancement: predominantly retrospective evidence, requiring confirmation by randomised trials ²¹.
• Retention approaches: low- to very low-certainty evidence, with no firm conclusion possible about any one approach ³⁴.
• Downgrading on indexing status: 5 of the sources used here describe themselves as systematic reviews or meta-analyses, but on actual inspection their PubMed PublicationType is not indexed as Systematic Review or Meta-Analysis (1 of them is moreover an ahead-of-print online publication whose MEDLINE indexing is not yet complete); this article downgrades its relaying of all 5 accordingly ^1316182229.

So whenever you meet any assertion that one option is better than another, the reasonable next question is: what design was that evidence, how many people were in it, and how did the authors themselves rate it ⁴¹.

4. Retention and relapse: the course of care does not end on the day the appliance comes off

This is a stretch of the orthodontic domain that is very easily missed and yet is stated very clearly in the literature.

The background of the Cochrane systematic review puts it directly: without a phase of retention after successful orthodontic treatment, teeth tend to "relapse", that is, to return to their initial position; retention is achieved by fitting fixed or removable retainers ³². Retention is not an optional extra but part of the structure of the course of care ³².

So which retention approach is better? At present there is no clear answer to that question. The review included 47 studies with 4377 participants in total; 28 of them were judged to be at high risk of bias, 11 at low risk and 8 unclear. The authors' conclusion is that the evidence is of low to very low certainty, so no firm conclusion can be drawn about any one approach to retention over another ³³³⁴. The review also lets you see the shape of the trade-offs between approaches, but change the wear regimen and the direction can reverse: in one study comparing part-time wear of a removable retainer with a fixed retainer, the removable retainer was more likely to cause discomfort but was associated with less retainer failure and better periodontal indices ³⁶; while in another study comparing full-time wear of a removable retainer with a fixed retainer, the removable retainer was likewise better on the periodontal indices, yet carried a higher risk of retainer failure (risk ratio 3.42, 95% CI 1.38 to 8.47) ⁴⁶. So the statement that removable retainers fail less often cannot be read as a general rule; it holds only in the part-time wear setting of the former study ³⁶⁴⁶. Another study showed no evidence of a difference in stability between part-time and full-time wear of the same Hawley removable retainer ³⁷. These are all results from single studies with sample sizes in the order of tens of participants, and cannot be taken as general rules ³³.

There is one further boundary that has to be disclosed honestly: this Cochrane review explicitly excluded studies using aligners ³⁵. That is to say, this body of evidence on retention approaches after fixed appliances are removed does not automatically apply to the situation after a clear aligner course has finished ³⁵.

As for retention after maxillary expansion, a separate systematic review deals with it directly, and its subjects were limited to growing patients with posterior crossbite ³⁰: across the 6 studies it included, the length of the retention period ranged between 4 weeks and 16 months ³⁰; the authors' conclusion is that 6 months of retention with either fixed or removable appliances seems to be enough to avoid relapse, or to keep the changes seen in short-term follow-up minimal ³¹. The point here is not that figure but that gap — this range is the span of the retention protocols adopted by each of the 6 studies, not anyone's recommended length of wear; the review does not argue for the cause of the gap, and this article does not infer one on its behalf ³⁰. The authors' wording is "seem to be", and it is limited to short-term follow-up ³¹. The concrete question of how long a given person's expander has to be worn falls within the scope of a supplementary canonical card and is not answered here ⁴⁰.

5. What costs are made of and what makes them vary

This article quotes no price and does not touch any country's fee regulations or insurance coverage ⁴².

The item-by-item breakdown framework for an orthodontic quotation (which items have to be lined up, which ones are easily charged separately) is already handled by a canonical card and is not expanded again here — see canonical card KM-DENTAL-08 (in production) ⁴³.

This section does only one domain-level thing: it points out which clinical variables make the number of items and the structure of the course of care heavier, because the literature carries clues about these variables, and they are precisely the extension of the classification and timing frameworks in the sections above ⁴³:

• Classification and aetiology: skeletal cases and dental cases take different management routes, the former possibly involving a facemask, an expansion appliance or a subsequent surgical assessment, so the number of items is naturally larger ¹¹⁵.
• Timing: early intervention and intervention at the growth peak are different arrangements of the course of care, and the guideline gives different timing recommendations for different classifications ⁴⁵⁶.
• Type of appliance: clear aligners are described as carrying a potential benefit in treatment length, but the level of the evidence behind that conclusion is limited ²³²².
• Whether adjunctive procedures are used: for example whether skeletal anchorage is added — in the facemask comparison for anterior crossbite, adding skeletal anchorage did not bring a clinically meaningful advantage in overall sagittal correction (ANB mean difference 0.07 degrees), with only a small additional improvement in maxillary advancement (SNA mean difference 0.59 degrees) ¹².
• The length and the manner of retention: this is a stretch that continues to exist after the active course of care has ended ³²³⁰.

Local systems and costs: the scope of insurance coverage, fee regulations and channels for verification in each place fall outside the scope of this global-track article; please see the corresponding local canonical card ⁴².

6. A pre-visit checklist and an overview of information red flags

6-1 Six questions you can ask at the consultation

The questions below were compiled by this site from the forks in the road set out in the sections above; their purpose is to help you follow the explanation you are given. They are not a scoring sheet and not a diagnostic tool ⁴⁴:

1. In terms of classification, is my situation more skeletal or more dental, and what is that judgement based on ¹¹⁴?
2. Given my age and stage of growth, what is the difference between starting now and starting later ¹³?
3. What appliance options are there, and does each of them work by changing the skeletal relationship or by changing tooth position ¹⁸²⁰?
4. Roughly how strong is the evidence behind this recommendation ¹³¹⁶²²?
5. What is the plan for retention, how long does it have to be kept up, and what happens if a retainer fails or is lost ³²³⁴?
6. Which items does the cost include, and which are charged separately ⁴³?

The explanation of a medical procedure and consent to it fall within the scope of local statutes; for their concrete requirements please see the corresponding local canonical card ⁴².

6-2 Situations during the course of care that call for a return visit

The following are safety reminders, not diagnostic criteria; if they occur, contact your dentist for assessment and do not deal with the appliance yourself ⁴¹:

• The appliance comes loose, is deformed or is damaged, or a sharp part is scraping the mucosa ⁴¹.
• Persistent pain, swelling or bleeding of the teeth or gums — periodontal indices are in any case one of the outcomes that orthodontic research tracks ³⁶.
• White patches appear on the tooth surface — white spot lesions are a common problem that has been quantified in fixed appliance treatment ³⁸.
• A retainer no longer fits, cannot be seated, or has been lost for some time — because without a retention phase teeth tend to relapse ³².

6-3 Information red flags: statements that should prompt one more question

The following are this site's reading reminders. They are editorial judgements, and they refer to no individual and no institution ⁴⁴:

• Giving an appliance conclusion straight off, without first distinguishing skeletal from dental ¹¹.
• Presenting one family of appliances as generally superior, without mentioning the design of that evidence and its risk of bias ²²²¹.
• Not mentioning retention at all, or presenting retention as optional ³².
• Presenting the result of a single study as a general rule — for example presenting the finding that removable retainers fail less often as a general conclusion, without mentioning that the direction reverses when the regimen changes to full-time wear ³⁶⁴⁶³⁷.
• Using time pressure or a price incentive to push you into deciding on the spot — this has nothing to do with the classification of the treatment and belongs to solicitation technique ⁴⁴.

Risk factors (indications, side effects, contraindications)

Indications and triage

• Treatment timing varies with the classification: a pronounced class II anomaly can be started on early in order to reduce the risk of anterior tooth trauma, while a moderate one is preferably handled before or during the pubertal growth peak ⁴; a class III anomaly should be started early, and this also reduces the chance of needing later surgery to correct the anomaly ⁵; a pronounced transverse anomaly should be started on early in the upper jaw ⁶.
• Appliances are triaged by aetiology: intraoral non-skeletally anchored appliances suit dental and functional discrepancies and mild skeletal ones; a truly skeletal anterior crossbite is more reliably handled with facemask therapy combined with rapid maxillary expansion ¹¹.
• Most malocclusions can be effectively treated in the early mixed, the late mixed and the permanent dentition alike ³.
• Surgery is a subsequent option in some cases: what the guideline writes in its class III recommendation is that early treatment reduces the chance of "needing later surgery to correct the anomaly", and that sentence also shows that surgery does exist in the pathway of some cases ⁵. Whether it is needed has to be assessed by a dentist together with the relevant specialties, and is not answered here ⁴¹.

Possible side effects and the costs that have been quantified

• Root resorption: orthodontically induced inflammatory root resorption is a common complication of orthodontic treatment ²⁵. A meta-analysis measuring with cone-beam computed tomography shows that the clear aligner group was significantly lower than the fixed appliance group (weighted mean difference -0.47 mm, versus -0.54 mm), but that the absolute difference is modest ²⁶²⁷; these two figures are the amount of resorption in each group separately rather than a between-group difference, the source does not report their difference as an effect estimate, and the subjects included in that analysis were limited to patients older than 14 years with fully developed anterior roots ²⁶²⁷.
• White spot lesions (decalcification of the tooth surface): in studies of fixed appliance treatment, the prevalence of white spot lesions at debonding varied between 12% and 55% ³⁸. Regular fluoride varnish has been shown to have a preventive effect; that review included 7 studies with 666 patients in total, of which only 5 provided data for the meta-analysis, giving a pooled risk ratio of 0.64 (95% CI 0.42 to 0.98); the authors graded the certainty of the evidence as very low ³⁹. The upper bound of the confidence interval, 0.98, is already close to 1, and readers would do well to take that in together with the rest ³⁹. Whether it is carried out and at what frequency is a clinical decision ⁴¹.
• Periodontal and soft tissue: periodontal indices (gingival index, bleeding index) are one of the outcomes tracked in retention research, and the different retention approaches have advantages and disadvantages against one another on these indices ³⁶.
• Discomfort and compliance: in one study comparing part-time wear of a removable retainer with a fixed retainer, the removable retainer was more likely to cause discomfort ³⁶; extra-oral appliances, meanwhile, are described as bulky, unaesthetic and heavily reliant on compliance ¹⁴.
• Relapse: without a retention phase, teeth tend to return to their original position ³².

Limits and the boundaries of the evidence (readable as the part not to expect too much from)

• The magnitude of skeletal change in adults and late adolescents is limited: MARPE reached statistical significance, but the authors assess it as clinically modest ²⁸²⁹.
• Adding skeletal anchorage does not mean a better result: compared with the conventional tooth-borne version, the skeletally anchored facemask did not demonstrate a clinically meaningful overall sagittal advantage ¹².
• Retention approaches are unsettled: low to very low certainty, with no firm conclusion possible about any one approach ³⁴.
• A gap in the evidence on retention after clear aligners: the Cochrane retention review explicitly excluded studies using aligners ³⁵.
• The effect of early treatment is influenced by aetiology and by appliance, and the overall certainty of the evidence ranged from very low to moderate ¹¹¹³.

Actual treatment and its results vary from person to person and must be assessed by a dentist against the individual situation ⁴¹. This article lists no set of contraindications — whether a person is suited to orthodontic treatment, and whether other oral conditions have to be dealt with first, fall within the scope of clinical assessment, can be determined only with imaging and examination, and are not replaced here by text ⁴¹.

Every clinical statement in this article is mapped line by line to its cited source (see the sources and evidence chain below). It has not been clinically reviewed by a licensed practitioner. This is health information, not individual advice; assessment by a clinician is required.