Background
Cigarette smoking is a global public health problem associated with high morbidity and mortality [
1]. It is a major risk factor for health problems, such as cancer, cardiovascular and respiratory diseases. Besides, smoking is a risk factor for oral diseases, such as oral cancer, periodontitis, gingival recession, tooth loss and implant failure [
2‐
4]. Smoking is also associated with higher costs in periodontal treatment [
5‐
7] and increases the cost of life-time periodontal treatment from 8.8% up to 71.4% [
7].
There is overwhelming evidence about the benefits of smoking cessation to general health [
8,
9]. Quitting smoking also improves oral health conditions. Two interventional studies observed greater probing depth reduction and clinical attachment gain in periodontitis patients that quit smoking when compared to non-quitters [
10,
11]. Moreover, observational studies have shown that former smokers lose fewer teeth than current smokers [
12,
13]. Therefore, smoking cessation therapy (SCT) is recommended as an important component of periodontal treatment [
14].
Medical literature demonstrated that SCT is cost-effective because it reduces health care expenditures associated with the effects of smoking [
15,
16]. There are some economic evaluations of periodontal therapy in the literature [
17‐
20]. However, to the extent of our knowledge, no study has evaluated the cost-effectiveness of the implementation of smoking cessation therapy for periodontitis patients. When allocating resources, different sources of resources may be considered when considering For a given budget (e.g. those specifically assigned to the oral health department or section) and considering an acceptable outcome (tooth loss), economic evaluations may support an intervention that results in any improvement (in this case, in oral health) that may justify an optimal reallocation of health care resources [
21]. Otherwise, lack of data about these consequences may prevent such type of reallocation in real life [
21].
We hypothesized that SCT is a cost-effective intervention because it reduces the risk of tooth loss and consequentlyreduces the costs associated with therapies aimed to replace teeth (prosthesis and implants). Such appraisal could be extremely relevant in decision-making since smoking cessation therapy is available in the Brazilian public health system (PHS), but is underused by dentists. Additionally, it may contribute to direct the need of reallocation of resources in oral health care, reinforcing the resources allocation focused on general health. Therefore, this study aimed to evaluate the efficiency in resource allocation when implementing smoking cessation therapy (SCT) vs. its non-implementation in smokers with periodontitis that received periodontal treatment to prevent tooth loss, in the context of the Brazilian PHS.
Discussion
The results of this study show that implementation of SCT is an efficient way of allocating resources compared to its non-implementation, in the perspective of the Brazilian PHS. Even considering additional costs related to cognitive-behavioral intervention for smoking cessation, they tended to be favorably balanced by the health gain achieved. In some circumstances, the initial extra expenses in implementing the SCT may be compensated by costs saved in the subsequent dental treatment for smokers.
We chose the PHS perspective because all Brazilian citizens are entitled to the services provided by the public system [
41]. Currently, SCT in the PHS is conducted by a multi-professional team that does not include dentists. However, this intervention could be delivered by dentists in the Brazilian PHS [
42], especially if one considers that SCT conducted by oral health professionals increases tobacco abstinence rates [
43].
The cost-effectiveness of smoking on periodontal therapy in private practice was evaluated by Fardal et al. [
7], based on costs of the American Dental Association. They found an increased cost of periodontal treatment for smokers and that the lifetime cost of periodontal therapy is equivalent to about 25% of the cost of smoking for patients who smoke 20 cigarettes per day. However, the authors did not evaluate the cost-effectiveness of SCT. Feldman et al. [
44] compared the cost-effectiveness of a high-intensity therapy with a low-intensity smoking cessation intervention in a Swedish dental setting. Their results favored the high-intensity smoking cessation intervention when willingness to pay was €4000/QALY. The utility weights were derived via general health-related quality of life questionnaire. Their results are important because this high-intensity smoking cessation intervention is very similar to SCT adopted in Brazilian PHS. According to their findings and our results, SCT is a cost-effective intervention for periodontitis patients, which empathizes the necessity to increase application of SCT in the Brazilian PHS dental setting.
From the Brazilian PHS perspective, SCT could be considered a cost-effective option (probability from 79–99%) even considering a low WTP threshold (US$), as we assumed to exemplify. Even when uncertainties are considered, we could observe that a low incremental cost (not exceeding US$350) could be expected. Since we do not have a known WTP for the health effects considered as outcomes in the present evaluations, we analyzed the CEA and the probabilities of simulated cases yield on different quadrants to permit the decision-maker to judge this information and consider if it is acceptable. It is possible to opt for the different preferences for inefficiencies occurring in different quadrants since both size and nature of risks may be presented [
45].
Although in both analyses, CEA and CUA, the implementation of SCT was cost-effective, the health effects in CUA were smaller than in CEA. This result was expected because effects in subjective measures are less evident than in objective measures.
We chose tooth loss as the outcome of CEA because it is considered the true endpoint of periodontal disease [
46] and the most important objective outcome to the patient [
47]. Therefore, tooth loss should be considered the most appropriate outcome in an economic analysis [
31]. Some studies used surrogate outcomes, such as bleeding on probing, plaque index, probing depth reduction and clinical attachment gain [
48‐
50]. However, the precise impact of these surrogate outcomes on the patient is unclear [
31].
We included a CUA in our study because this type of analysis allows including a measure of the patients’ preferences and perception about their health. The importance of patient-related outcomes is impacting more studies with CUA in dentistry [
51]. However, there is still a scarcity of this type of analysis in some regions, such as in South American countries [
51]. Medical literature usually uses EuroQOL-5Dimension (EQ-5D) [
52] or Structured Form 6 Dimension (SF-6D) [
53] to determine QALY. However, these instruments evaluate general health, which is unlikely to be sensitive to important domains of oral health, such as chewing ability or aesthetics [
54]. In the absence of a suitable measure in Dentistry, and considering the low sensitivity of medical questionnaires, we assumed that the utility was proportional to quality of life. This methodology was proposed by previous studies that converted scores from quality of life questionnaires to utility scores between 0 and 1 and reflected changes related to periodontal conditions/treatment [
55,
56]. Even though this is not the ideal approach, this interim tool allowed the incorporation of a patient-centered approach into the analysis. We believe that these limitations do not impact ur findings because the utility scores were used in both analysed strategies.
The inclusion of CUA also allows verifying that there are situations in which SCT could be cost-saving compared to its non-implementation, which is an additional argument to endorse this therapy in the PHS. On the other hand, our CUA should be interpreted carefully, because it is not possible to affirm that quality of life has a linear relationship with the utility. We expect that oral health-related quality of life instruments may have registered oral health conditions that are important to the patient, which are not necessarily comparable to general health status. Oral health-specific utility measures are probably more sensitive in capturing the effectiveness of oral health interventions [
50]. Instruments that use an indexed scale for oral health-related aspects need to be developed to improve the comparison between studies and different therapies.
We applied sensitivity analysis to characterize the uncertainty of our results. We tested a model with implants because this type of rehabilitation presents better utility scores than a partial prosthesis, but it also includes higher costs in the model. However, even considering these higher costs, rehabilitation with implants did not affect the results and the implementation of SCT remained a cost-effective therapy. Further, it was dominant over the non-implementation of SCT. We did not test partial fixed prostheses in our models since they are not available in the Brazilian PHS. Moreover, they may result in additional harm such as pulp exposure, which may lead to endodontic treatment [
18].
We considered different scenarios of the use of resources in SCT [
34] when the SCT costs were varied in the sensitivity analysis. We observed a marginal dominance, since when varying the SCT costs, the therapy remained cost-effective (but not dominant, as in the base case). These findings also provide evidence that SCT should be implemented for PHS. Since cognitive-behavioural therapy is responsible for most SCT costs, we believe the variation in SCT costs considering these different scenarios may also reflect possible variations proposed in different protocols for SCT, even those different from Brazil. Although the costs for professional training were not included in the primary model, even under a conservative approach, these additional costs did not impact the results.
The number of sessions of cognitive-behavioral therapy in SCT seems to exert the same influence on the cost-effectiveness of the therapy. The number and frequency of periodontal maintenance sessions can vary according to clinical conditions, such as extension and severity of the disease. The number of SCT sessions was fixed, as in the Brazilian protocol for SCT, but different scenarios were used in the sensitivity analyses to explore these possible variations. Even when we tested variations in SCT costs, the implementation of SCT remained cost-effective. Therefore, despite exploring possible variables and uncertainties related to our models for CEA and CUA, we reinforce that the SCT may be a cost-effective therapy for periodontitis patients to be implemented in the Brazilian PHS.
Economic evaluations are a standard tool in the assessment of health care technologies to maximize benefits from the available resources [
57]. The need to allocate public finances increased the interest in cost-effectiveness research in dentistry [
58]. A cost-effectiveness criterion can play an important role by guiding the incorporation of new technologies into the population. Policymakers from some countries, such as Australia, Canada and European countries, have adopted economic evaluations to their drug guidelines and reimbursement [
59]. The present findings are especially important for the Brazilian PHS and demonstrate that SCT should be implemented. It is necessary to emphasize that studies with different populations, costs and perspectives should be conducted to confirm the cost-effectiveness of the implementation of SCT concerning tooth loss in different scenarios. This model can be used as a model for future cost-effectiveness analysis with costs and effects from other countries. As we adopted a model in which repetitions of SCT were not included due to smoking relapse, other models may also test the influence of variables related to that in the cost-effectiveness of SCT.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.