Clinical Research
Cost-effectiveness of Different Post-retained
Restorations
Falk Schwendicke, PhD,* and Michael Stolpe, Dr†
Abstract
Objectives: Dentists can choose between metal and
fiber post systems to provide post-retained restorations.
The risk of tooth loss and other complications differs be-
tween different post systems, as do the initial treatment
costs. We aimed to assess the cost-effectiveness of (1)
cast metal (MC), (2) preformed metal (MP), (3) glass fiber
(GF), and (4) carbon fiber (CF) post-retained restorations.
Methods: A mixed public-private payer’s perspective
within German healthcare was taken. Risks of complica-
tions were extracted from systematic reviews. Costs were
estimated by using fee items and 2016 material costs.
A Markov model was constructed to follow up an
endodontically treated molar receiving a post-retained
crown in an initially 50-year-old patient during his life-
time. Monte Carlo microsimulations were performed to
assess lifetime costs and tooth retention time. Results:
MPs were least costly (692V), retaining teeth for
26.7 years. GFs were more costly (745V), retaining teeth
for 27.6 years. MCs were minimally more effective but
also more costly than GFs (774V). CFs were less effective
and most expensive (825V, 26.7 years). For payers
willing to invest more than 60V per tooth retention
year, GF was cost-effective. Payers willing to invest an
additional 670V found MC to be cost-effective. These
findings were found robust in sensitivity analyses. Con-
clusions: For payers not willing to invest additional
money for longer tooth retention, MP seemed most suit-
able to retain restorations. For payers with additional will-
ingness to pay, GF seemed suitable, retaining teeth for
longer. MC was only cost-effective under very high will-
ingness to pay. CF is not recommendable on the basis
of their cost-effectiveness. (J Endod 2017;43:709–714)
Key Words
Computer modelling, dental care, fiber posts, health
economics, Markov model, prosthetics
From the *Department of Operative and Preventive Dentistry, Charit
e – Universit€atsmedizin Berlin, Berlin, Germany; and †Kiel Institute for the World Economy, Kiel,
Germany.
Address requests for reprints to Dr Falk Schwendicke, Department of Operative and Preventive Dentistry, Charit
e – Universit€atsmedizin Berlin, Campus Benjamin
Franklin, Aßmannshauser Str. 4-6, 14197 Berlin, Germany. E-mail address: falk.schwendicke@gmail.com
0099-2399/$ - see front matter
Copyright ª 2017 American Association of Endodontists.
http://dx.doi.org/10.1016/j.joen.2017.01.002
JOE — Volume 43, Number 5, May 2017
T o restore endodonti-
cally treated teeth with
limited coronal tooth hard
Significance
We found preformed metal post-retained restora-
tions to be least costly, whereas glass fiber post-
tissue, post-retained resto-
retained restorations were moderately more
rations (crowns) are often
expensive but also more effective. Clinical
required. For such post-
decision-making should consider both the initial
retained restorations, cli-
treatment costs and also costs associated with
nicians can use metal posts
possible long-term complications.
or non-metal fiber posts.
Metal posts can either be
cast, often as post-core casts, or preformed, with the core being directly placed after
cementation of the post. Fiber posts are usually preformed and contain carbon, quartz,
or glass fibers, which are embedded in an epoxy or methacrylate matrix. These are
usually adhesively luted, with the core being directly placed. In contrast to metal posts,
the elastic modulus of fiber posts is similar to dentin (1, 2), which should assist to
distribute the stress under load, reducing the risk for vertical root fracture (3–6).
In contrast to the results from in vitro studies, clinical studies found the risk of such
fatal complications to be similar in teeth with metal versus fiber post-retained restorations;
however, risks seem to differ in different types of metal and fiber posts (7). Moreover, the
risk of non-fatal complications might be different between different post types too (7).
At present, there is ambiguity with regard to the suitability of different post systems, with
no clear guidance being available for clinicians as to which system is most appropriate (8).
One aspect that has so far not been assessed when comparing different post-retained
restorations is cost-effectiveness: The placement of different posts involves different
efforts; for laboratorial manufacturing or adhesive luting, these differences will lead to
different initial treatment costs. Moreover, different risks will generate different long-
term costs for mending complications such as recementation, re-restoration, or replace-
ment. Assessing such long-term consequences of treatments and retreatments is complex.
The present study aimed to assess the cost-effectiveness of different post-retained
crowns. To reflect the discussed long-term aspects, a model-based approach was
chosen. The findings of this study are relevant for payers, clinicians, patients, and
healthcare researchers alike because they might assist to guide clinical and non-
clinical (health services) decision-making as well as the conduct of future studies.
Methods
Setting, Perspective, Population, Horizon
This study adopted a mixed public-private payer perspective in the context of
German healthcare. We modeled a population of initially 50-year-old men with a molar
tooth that had completed endodontic treatment of a vital, painless pulp with 3 root
canals. Molars were assumed to require a post-retained crown and were followed
Cost-effectiveness of Different Endodontic Posts 709
Clinical Research
during the patient’s lifetime (TreeAge Pro 2013; TreeAge Software,
Williamstown, MA), which was determined by age and gender.
All cost-effectiveness evaluations were performed per 1 molar to
avoid clustering and the associated issues of correlation and to increase
the ease of interpretation of our findings. Note that for this study, we did
not specify any further factors that influence the survival of post-retained
teeth such as root canal and post shape, post preparation type, or
specific cementation or adhesive luting materials, etc.
Comparators
We compared 4 strategies:
1. A cast metal post-retained crown (MC)
2. A preformed (passive) metal post-retained crown (MP)
3. A glass fiber post-retained crown (GF)
4. A carbon fiber post-retained crown (CF)
Metal posts were assumed to be cemented conventionally, whereas
fiber posts were assumed to be luted adhesively. The placed crown was
assumed to be a full non-precious metal crown, as is standard under the
assumptions of the statutory insurance in Germany for molars.
Model and Assumptions
Molars were assumed to experience fatal and non-fatal complica-
tions, the risks of which were extracted from a recent systematic review
for the different comparators (7). Fatal complications were those leading
to the tooth being extracted (mainly root fractures). Non-fatal complica-
tions were those needing retreatments, including endodontic complica-
tions, decementation or crown dislodgments, fractures, secondary
caries, etc. The risk of complications and the probabilities of allocation
to different treatments mending these complications are given in Table 1.
Endodontic complications were assumed to be mended by
nonsurgical (orthograde) or surgical retreatment (apisectomy). The
risks of endodontic complications were derived from existing studies
in the field and systematic reviews (Table 1). Teeth that had experienced
orthograde retreatment were treated surgically in case of further end-
odontic complications; those that had received surgical retreatment
were extracted in case of further endodontic complications. We varied
the proportion of endodontic complications being mended surgically or
non-surgically between 0% and 100%. In an additional analysis, we
assumed teeth with adhesively placed posts to not receive orthograde
but only surgical retreatment, whereas those with metal posts were
all first non-surgically retreated in case of complications.
Restorative complications (Table 1) were mended by recementation
of the crown or renewal of the crown (involving the renewal of the post and
core too). We assumed 60% of such complications to be decementations,
ie, requiring recementation, and 40% to require renewal of the crown.
In case of teeth that needed removal, their replacement by using
implant-supported single crowns was modeled. We assumed 50% of
teeth to be replaced in the base-case scenario; this was varied to account
for heterogeneity (of patients, etc.). Implant-supported crowns were
assumed to be prone for complications from the implant (peri-implan-
titis, implant fracture), which we assumed to lead to implant removal
and in 50% of the cases renewal of the implant, and complications of
the crown (decementation, fracture, etc), which we assumed to lead
to crown replacement or recementation. Risks of complications of
implant-supported crowns were derived from a systematic review (13).
The constructed model is shown in Figure 1. Model validation was
performed internally by varying key parameters to check their impact
on the results, by evaluating different model structures, and by perform-
ing sensitivity analyses.
Health Outcomes and Measurement of Effectiveness
The health outcome was tooth retention years, ie, the mean time a
tooth was retained in a patient’s mouth. Transition probabilities to allow

TABLE 1. Parameters Used for Effectiveness Estimation

Transition
probability Triangular Allocation
Health state Reference per year distribution* Allocation to probability
Fatal failures
MC Figueiredo et al, 2015 (7) 0.0034 0.1, 1.0, 2.2 Removal 0.50
MP 0.0055 0.8, 1.0, 1.2 Removal and replacement 0.50†
GF 0.0036 0.2, 1.0, 1.8
CF 0.0057 0.6, 1.0, 1.4
Non-fatal restorative
failures‡
MC Figueiredo et al, 2015 (7) 0.022 0.3, 1.0, 1.7 Renewal post-crown 0.40
MP 0.013 0.7, 1.0, 1.3 Recementation post-crown 0.60
GF 0.014 0.1, 1.0, 2.3
CF 0.023 0.7, 1.0, 1.3
Non-fatal endodontic
complications
RCT Ricucci et al, 2011 (9); 0.0232a
0.823 — Non-surgical re-RCT 0.50
Schwendicke et al, 2014 (10) Surgical re-RCT 0.50†
Non-surgical re-RCT Ng et al, 2008 (11) 0.059 0.3, 1.0, 2.0 Surgical re-RCT Extraction 0.80†
0.20
Surgical re-RCT Torabinejad et al, 2009 (12) 0.080 0.5, 1.0, 2.0 Extraction 1.00
Implant complications Jung et al, 2012 (13) 0.032 0.5, 1.0, 1.7 Renew Remove 0.5
0.5
Implant crown Jung et al, 2012 (13) 0.047 0.6, 1.0, 1.8 Renewal Recementation 0.4
complications 0.6
RCT, root canal therapy.
*Distributions were used to express uncertainty, with triangular distributions being used for random sampling during probabilistic sensitivity analyses.
†
Were varied in sensitivity analyses.
‡
Note that we modeled endodontic complications separately from the remaining complications outlined as ‘‘non-fatal’’ in the systematic review informing this study (7) to allow modeling of follow-up endodontic
complications. Note that this will have increased the overall risk of non-fatal complications, without any difference between groups.

710 Schwendicke and Stolpe JOE — Volume 43, Number 5, May 2017
 Clinical Research

Figure 1. Transition diagram. Fatal or non-fatal complications could occur (see Table 1 for details). Fatal complications led to removal, whereas non-fatal compli-
cations could be endodontic or restorative ones, which could be mended via a range of options but did not lead to tooth loss. However, it should be noted that in the
follow-up states to these non-fatal complications, tooth loss was possible (ie, after non-surgical re-root canal treatment). For implants, implant and crown complications
were modeled. For the purpose of illustrating the application of the transition probabilities shown in Table 1, we also depict these probabilities for MC (p).

teeth to move from one health state to another (ie, risks of complica-
tions and associated retreatments) were estimated as described above
and are summarized in Table 1.
Resources and Costs
This study modeled a publicly insured patient by using a mixed
public-private payer perspective, as is most common in Germany.
Cost calculations were based on the German public and private dental
fee catalogues, BEMA (Bewertungsmaßstab) and GOZ (Geb€uhrenord-
nung f€ur Zahn€arzte) (14, 15). German dentists estimate their costs
by using fee items; this estimation is the basis for claims for
reimbursement from the patient himself or his insurer. Most patients
are publicly insured (16), ie, fee items are drawn from the public cata-
logue BEMA for the majority of treatments (like most conservative or
surgical treatments). Prosthetic treatments are only partially covered;
non-surgical re-root canal treatment and implants are not usually
covered at all. Costs for treatments partially or not at all covered by
the public insurance are estimated by using fee items from the private
catalogue GOZ. Publicly insured patients pay the additional costs out of
their own pocket (in case they have private additional insurance, these
costs might be reimbursed). For GOZ, factoring of item points is com-
mon to determine prices, which allows individualizing of fees. The pre-
sent analysis used the standard multiplication factor (2.3).
Calculating costs by using average national multiplication factors
(17) has not been found to greatly impact on cost-effectiveness (18).
A small proportion of German patients are not members of the
public insurance; for them, all costs are estimated by using fees from
the private item catalogue. Our calculations do not fully apply to these
patients; however, analyses using cost estimations for these patients
have been found to yield only limitedly different results compared
with those yielded for publicly insured patients (18, 19). Because of
the lack of primary data, opportunity costs of patients’ time in
treatment were not accounted for.

JOE — Volume 43, Number 5, May 2017 Cost-effectiveness of Different Endodontic Posts 711
Clinical Research
TABLE 2. Cost Estimation per Treatment Complex
Course of treatment
MC
MP
Core crown with GF
Core crown with CF
Recementation of a crown
Orthograde re-RCT
Apical surgery
Tooth/implant removal
Implant insertion
Implant crown
Costs were calculated by quantification of item fees from public or private item catalogues. For
further details see Appendix, Schwendicke and Stolpe (10), and Schwendicke et al (15).
Additional fee items, laboratory, and material costs were included
in our cost estimations. Laboratory costs were estimated by using fee
items as well, as outlined elsewhere (20). For materials, 2016 market
prices for exemplary materials were used (Appendix is available online
at www.jendodon.com).
Discounting, Currency, and Price Data
All costs were calculated in 2016 euros, and future costs and effec-
tiveness were discounted at 3% per annum (21), discounting accounts for
opportunities forgone if spending money now instead of later or gaining
health benefits later instead of now. Discounting rates were varied
between 0% and 5% to explore the impact of higher or lower discounting.
Analytical Methods
Monte Carlo microsimulations were performed for analysis, with
1000 independent individual molars being followed during the average
expected lifetime of patients (which was 29 years) (22) in annual
cycles. Different lifetimes were modeled (ie, different initial start ages
chosen) within a sensitivity analysis. Convergence was checked by
inspecting sampling distributions.
Costs and effectiveness for different strategies were assessed, and
cost-effectiveness planes were used to depict cost-effectiveness. Strategies
were ranked according to their costs, and incremental cost-effectiveness
ratios (ICERs) were exemplarily used to express cost differences per
Costs (V)
gained or lost effectiveness. Positive ICERs indicate additional costs
488.25 per additional effectiveness, whereas negative ICERs indicate additional
434.49
491.21
costs per effectiveness loss. Strategies that were more costly and less
489.22 effective were dominated; those that were more costly but also more
54.29 effective were undominated.
526.78 To introduce parameter uncertainty, we randomly sampled transi-
154.63 tion probabilities from a triangular distribution of parameters (23). By us-
67.41
958.40 ing estimates for costs (c, in euros) and effectiveness (e, in years), the net
345.23 benefit of each strategy combination was calculated by using the formula:
net benefit ¼ l  De
Dc;
with l denoting the ceiling threshold of willingness to pay, ie, the addi-
tional costs a decision maker is willing to bear for gaining an additional
unit of effectiveness (24). If l > Dc/De, an alternative intervention is
considered more cost-effective than the comparator despite possibly
being more costly (23). We used the net-benefit approach to calculate
the probability of a strategy being cost-effective for payers with different
willingness-to-pay ceiling thresholds. In addition, a number of univar-
iate sensitivity analyses were performed to explore the impact of uncer-
tainty and heterogeneity.
Results
Study Parameters
Effectiveness parameters can be found in Table 1. Cost estimates
for different procedures can be found in Table 2, with details on unit
costs and quantities laid out in Appendix (available online at www.
jendodon.com).
Base-case Scenario
Figure 2A shows the cost-effectiveness plane. MPs were least costly
(692V), retaining teeth for 26.7 years. GFs were ranked next according
to their costs (745V), retaining teeth for 27.6 years. MCs were mini-
mally more effective (27.6 years after rounding) but also more costly
than GFs (774V). CFs were less effective and more expensive than

Figure 2. Cost-effectiveness plane and cost-effectiveness acceptability. (A) Costs and effectiveness of different treatments were plotted. MPs were least costly.
GFs were more costly but also more effective, with ICER of 70V/year per tooth retention year. MCs were more effective but also more costly, with ICER of
670V. CFs were most costly and less effective than alternatives. (B) Cost-effectiveness acceptability curve indicates how with increasing willingness to pay, the
probability of cost-effectiveness for different treatments changes. GF is most likely to be cost-effective for payers willing to invest between 60V and 740V per
year; for those with higher willingness to pay, MC had the highest probability.

712 Schwendicke and Stolpe JOE — Volume 43, Number 5, May 2017

TABLE 3. Strategy Ranking and ICERs, Compared with the Next Higher-
Ranked Strategy
Strategy Rank Costs (V) Effectiveness (y)
MP 1 692 26.7
GF 2 745 27.6
MC 3 774 27.6
CF 4 825 26.7
Rankings were performed according to costs. A positive ICER indicates additional costs per gained
effectiveness; negative ICER indicates lost effectiveness per additional costs.
alternatives (825V, 26.7 years); these were dominated. Rankings and
ICERs are shown in Table 3.
For payers unwilling to pay additional money for gained effective-
ness, MP had >90% probability of being cost-effective (Fig. 2B). This
probability decreased markedly with increasing willingness to pay.
For payers willing to invest more than 60V per tooth retention year,
GF had a probability of >40% of being cost-effective, exceeding 50%
for payers willing to invest 180V to 730V per tooth retention year.
Above this threshold, MC (which was most effective but also more costly
than GF) had a probability of >50% of being cost-effective.
Sensitivity Analyses
Sensitivity analyses were performed to explore parameter uncer-
tainty and heterogeneity (Table 4). In younger patients, the ranking
remained stable, whereas overall, costs increased because of longer
time periods available for retreatments. In older patients, the ranking
was again stable, with reduced total costs. Assuming that non-surgical
(orthograde) endodontic retreatments would never be performed
decreased the costs but also the effectiveness of all strategies, with
the ranking being unchanged compared with the base-case. Assuming
the opposite (all endodontic complications being first mended non-
surgically) increased the costs and minimally also the effectiveness;
the ranking was unchanged. Assuming that adhesively placed fiber
post-retained restorations would not receive non-surgical retreatment
whereas conventionally cemented metal posts would altered the
ranking, with GF now being the least costly option and MC being only
minimally more effective. Assuming removed molars to not be replaced
by using implants decreased the costs significantly (effectiveness was
not affected), with the ranking remaining unchanged. Setting the
discounting rate to 0% or 5% decreased or increased the costs and
effectiveness, respectively, but did not alter the ranking.
Discussion
Assessing long-term cost-effectiveness is complex, especially when
dealing with the consequences of an initial treatment decision, because
Clinical Research
one has to weigh the initial treatment costs against the risk of fatal and
non-fatal complications, which both affect the effectiveness and also the
ICER costs emanating from any treatments for mending complications. The
chosen model-based approach was suitable for evaluating the cost-
— effectiveness of different post-retained crown restorations.
60
670 We found that long term, MC and GF show very similar effectiveness,
–57 which is in line with clinical trial data (7). However, GF was less costly.
That was despite GF generating similar costs for the initial placement, with
costs for the post material and the adhesive luting and core buildup
compensating any cost savings from avoiding laboratorial manufacturing.
Instead, the observed cost savings of GF compared with MC were mainly
due to fewer non-fatal complications occurring, which reduced the costs
of retreatments (such as recementation or renewal of the post-retained
crown). Moreover, we found that one driver of costs is endodontic
retreatments (especially non-surgical retreatment) because the benefit
in effectiveness is limited, whereas costs are substantial. That could be
seen when assuming fiber posts to not receive such treatment because
of difficulties in providing orthograde retreatment. In this case, GF was
nearly as effective but substantially less costly than MC. However, one
should note that the cost difference between non-surgical and surgical
retreatment might not apply to all settings and might well depend on
how exactly both treatments are performed. On the basis of our findings,
decision makers willing to invest additional money can choose both GF
and MC. Because of the high additional long-term costs of MC and the
minimal gain in effectiveness as well as the poor esthetics of MC (which
might be relevant in anterior teeth, not–as modeled–molars), decisions
toward GF might well be justifiable.
MP was the least costly option in all scenarios and might be consid-
ered in case of strict cost containment, ie, in payers with no willingness
to pay for gained effectiveness. It should be noted that these cost advan-
tages were only minimal in case all teeth were replaced, because the
relatively higher risk of fatal complications and the subsequent costs
for replacement nearly compensated initial savings. In such cases, GF
might be a highly acceptable option (because the additional costs are
minimal, whereas the gained effectiveness was substantial).
We found CF to not have acceptable cost-effectiveness because it
was less effective than MC or GF and generally more expensive. More
specifically, CF had the highest risk of non-fatal complications and
also high risks of fatal complications, coupled with higher costs than
GF and MC except for one scenario shown in Table 4, namely ‘‘Revision
only in metal post-retained crowns’’ where CF incurred lower costs than
MC but still higher costs than GF. On the basis of cost-effectiveness
considerations, CF cannot be recommended.
This study has a number of limitations. First, the applied model
greatly simplifies the clinical long-term flow of events, yet it allowed a
range of relevant scenario analyses that confirmed the robustness of

TABLE 4. Cost-effectiveness of Different Strategies in the Base-case and Sensitivity Analyses

MC MP GF CF
Scenario c (V) e (y) c (V) e (y) c (V) e (y) c (V) e (y)
Base-case 774 27.6 692 26.7 745 27.6 825 26.7
Patient aged 40 y 856 36.0 769 34.7 809 36.0 909 34.7
Patient aged 60 y 689 18.8 609 18.4 661 18.8 711 18.4
No revision 722 27.3 645 26.6 693 27.3 768 26.6
All revision 817 27.6 730 26.8 778 27.6 868 26.8
Revision only in MPs 817 27.6 730 26.9 693 27.3 768 26.6
0% replaced 688 27.6 569 26.8 651 27.6 692 26.8
100% replaced 856 27.6 815 26.8 820 27.6 940 26.8
0% discounting 788 28.4 702 27.7 754 28.4 846 27.5
5% discounting 763 27.0 683 26.3 729 27.0 799 26.2
Costs (c) and effectiveness (e) are displayed. The least costly strategy is indicated in bold, and the most effective strategy is underlined. Sensitivity analyses evaluated how different patient ages, treatment decisions
for or against revision in case of endodontic failure, different replacement ratios for missing teeth, and different discounting rates affected the cost-effectiveness.

JOE — Volume 43, Number 5, May 2017 Cost-effectiveness of Different Endodontic Posts 713
Clinical Research
our findings. Second, the input data, mainly regarding the risk of
complications, stemmed from only a few studies (however, these
were systematically collated). Because of the ambiguity of data from
controlled and non-controlled trials regarding the risks of fatal and
non-fatal complications in metal versus fiber post-retained restorations
(7), caution is certainly needed when interpreting our data. On the basis
of the performed scenario analyses, our conclusions that GF and MC
were similarly effective but more expensive than MP might hold when
new trials are published as long as they do not find completely conflict-
ing evidence. Third, the cost estimates used are only valid for a mixed
payers’ perspective in Germany. Because cost estimation differs in other
settings, with labor and material cost structures being different, our
findings cannot easily be transferred. Moreover, we did not account
for costs occurring because of time lost at work or costs for any drug
consumption in case of pain, etc. However, the fact that retreatment
costs were one driver for lifetime costs is likely to apply to other settings
and should be borne in mind when making initial treatment decisions.
Fourth, a number of factors have been found to affect the survival and
success of post-retained restorations, such as the type of tooth (higher
risk of complications in anterior versus posterior teeth because of
disadvantageous load and stress distribution), the remaining tooth sub-
stance (number of walls, ferrule), and the tooth function in a possible
prosthesis (post-retained restorations acting as abutment for removable
prostheses fail most often, whereas those abutting fixed prostheses
showed longest survival) (3). Further factors that come to mind include
the specific type of post, the clinician skills, the laboratory performance,
the possible use of dental microscope, the proper isolation, patients’
caries risk, and many more. These have not been standardized within
the studies informing our analysis, but they play a role in a clinical
setting. Unfortunately, there were insufficient data to consider all these
aspects in our analyses. Having such data and generating cost-
effectiveness estimates for the resulting subgroups would allow more
informed, individualized decision-making. Last, the effectiveness differ-
ences yielded by our evaluation were around 1 year of tooth retention
time. Decision makers may or may not find such difference relevant, and
whether they do should be considered when interpreting our findings.
In conclusion, within the limitations of this study, MP restorations
are the least costly among the evaluated therapy strategies. GF and MC
restorations are more expensive but also more effective, retaining teeth
for longer. Because of the estimated willingness-to-pay thresholds,
payers with additional willingness to pay might find GF restorations
the most acceptable option; MC restorations will only be acceptable
under very high willingness to pay. CF restorations are not recommend-
able because their costs were highest, and their effectiveness was lower
than that of all the alternatives.
Acknowledgments
The authors deny any conflicts of interest related to this study.
Supplementary Material
Supplementary material associated with this article can be
found in the online version at www.jendodon.com (http://dx.doi.
org/10.1016/j.joen.2017.01.002).
714 Schwendicke and Stolpe
References
1. Barbizam JV, White SN. Fatigue susceptibility of an endodontic fibre post material.
Int Endod J 2014;47:202–9.
2. Lassila LV, Tanner J, Le Bell AM, et al. Flexural properties of fiber reinforced root
canal posts. Dent Mater 2004;20:29–36.
3. Goracci C, Ferrari M. Current perspectives on post systems: a literature review. Aust
Dent J 2011;56(Suppl 1):77–83.
4. D’Arcangelo C, Cinelli M, De Angelis F, D’Amario M. The effect of resin cement film
thickness on the pullout strength of a fiber-reinforced post system. J Prosth Dent
2007;98:193–8.
5. Gonzalez-Lluch C, Rodriguez-Cervantes PJ, Sancho-Bru JL, et al. Influence of mate-
rial and diameter of pre-fabricated posts on maxillary central incisors restored with
crown. J Oral Rehabil 2009;36:737–47.
6. Barjau-Escribano A, Sancho-Bru JL, Forner-Navarro L, et al. Influence of prefabri-
cated post material on restored teeth: fracture strength and stress distribution. Oper
Dent 2006;31:47–54.
7. Figueiredo FE, Martins-Filho PR, Faria ESAL. Do metal post-retained restorations
result in more root fractures than fiber post-retained restorations? a systematic
review and meta-analysis. J Endod 2015;41:309–16.
8. AAE. Endodontics: Colleagues for Excellence. Restoration of Endodontically
Treated Teeth. Available at: http://www.aae.org. Accessed July 16, 2016.
9. Ricucci D, Russo J, Rutberg M, et al. A prospective cohort study of endodontic treat-
ments of 1369 root canals: results after 5 years. Oral Surg Med Path Radio Endod
2011;112:825–42.
10. Schwendicke F, Stolpe M. Direct pulp capping following a carious exposure
versus root canal treatment: a cost-effectiveness analysis. J Endod 2014;40:
1764–70.
11. Ng YL, Mann V, Gulabivala K. Outcome of secondary root canal treatment: a system-
atic review of the literature. Int Endod J 2008;41:1026–46.
12. Torabinejad M, Corr R, Handysides R, Shabahang S. Outcomes of nonsur-
gical retreatment and endodontic surgery: a systematic review. J Endod
2009;35:930–7.
13. Jung RE, Zembic A, Pjetursson BE, et al. Systematic review of the survival rate and the
incidence of biological, technical, and aesthetic complications of single crowns on
implants reported in longitudinal studies with a mean follow-up of 5 years. Clin Oral
Impl Res 2012;23(Suppl 6):2–21.
14. KZBV. Catalogue of Fees [Geb€uhrenverzeichnisse]. Available at: http://www.kzbv.
de/gebuehrenverzeichnisse.334.de.html. Accessed July 16, 2016.
15. Schwendicke F, Stolpe M, Meyer-Lueckel H, et al. Cost-effectiveness of one-
and two-step incomplete and complete excavations. J Dent Res 2013;90:
880–7.
16. Gesundheitsberichterstattung des Bundes. Bev€olkerung mit Angaben zur
Krankenversicherung nach Art des Krankenversicherungsschutzes in 1000.
Available at: http://www.gbe-bund.de/. Accessed July 16, 2016.
17. Niehaus F, Keßler T, Finkenst€adt V. Ein Vergleich der zahn€arztlichen Verg€utung
nach GOZ und BEMA. WIP-Diskussionspapier 2/2011; 2011.
18. Schwendicke F, Graetz C, Stolpe M, Dorfer CE. Retaining or replacing molars with
furcation involvement: a cost-effectiveness comparison of different strategies. J Clin
Perio 2014;41:1090–7.
19. Schwendicke F, Stolpe M, Meyer-Lueckel H, Paris S. Detecting and treating occlusal
caries lesions: a cost-effectiveness analysis. J Dent Res 2015;94:272–80.
20. Schwendicke F, Stolpe M. Cost-effectiveness of caries excavations in different risk
groups: a micro-simulation study. BMC Oral Health 2014;14.
21. IQWiG. Appraisal of recommendations by the scientific board of IQWiG regarding
‘‘Methods to assess cost-effectiveness in German Public Health Insurance’’. [W€urdigung
der Empfehlung des Wissenschaftlichen Beirats des IQWiG zur ‘‘Methodik f€ur die Be-
wertung von Verh€altnissen zwischen Nutzen und Kosten im System der deutschen ge-
setzlichen Krankenversicherung’’]. Available at: https://www.iqwig.de/download/09-
03-18_Wuerdigung_der_Empfehlung_des_Wissenschaftlichen_Beirats.pdf. Accessed
July 16, 2016.
22. Statistisches Bundesamt. Mortality table 2009/2011 [Sterbetafel 2009/11]. Avail-
able at: http://www.destatis.de/. Accessed July 16, 2016.
23. Briggs AH, O’Brien BJ, Blackhouse G. Thinking outside the box: recent advances in
the analysis and presentation of uncertainty in cost-effectiveness studies. Ann Rev
Pub Health 2002;23:377–401.
24. Drummond M, Sculpher M, Torrance G, et al. Methods for Economic Evaluation
of Health Care Programmes. Oxford, UK: Oxford University Press; 2005.
JOE — Volume 43, Number 5, May 2017