The Use of Decision Analysis in Treatment Planning for Patients with Periodontal Disease

Rakhi Sinha, BS* ; Ira B. Lamster, DDS, MMSc**

ABSTRACT
The past several years have seen dramatic changes in the diagnosis and treatment of periodontal disease. Research has focused on the development of new methods to aid the clinician in the evaluation and treatment of patients. This research has caused a reassessment of how treatment decisions are made for the patient with periodontal disease. The purpose of this article is to discuss the use of decision trees for the examination and treatment of periodontal patients. Decision trees can allow clinicians to consider all options that are available when formulating a treatment plan, and modify approaches to therapy as treatment progresses. (Col Dent Rev 2:35, 1997)

INTRODUCTION

Studies during the past ten years have suggested that many long held beliefs concerning the diagnosis of periodontal disease need to be re-examined. First, our understanding of the natural history of these common disorders has been modified based on longitudinal clinical trials. The progression of periodontal disease had been considered to be slow and continuous throughout the adult life of affected individuals. We now understand that the periodontal diseases are similar to other chronic disorders in that they are characterized by periods of exacerbation (brief) and remission (longer).6

At the same time, our approach to the diagnosis of these diseases is changing. For many years, diagnosis has relied on clinical parameters such as probing depth and radiographic evidence of existing alveolar bone loss. These techniques are recognized as quite useful in evaluating past disease, but offer little insight in evaluating patient's risk of experiencing active disease in the near term. For this reason, there is considerable interest in developing new approaches for diagnosis of

patients. These new diagnostic tests seek to identify the "biochemical" lesion prior to the progression to the clinical lesion. Therefore, the dentist/periodontist of the future will rely on both traditional and diagnostic methods to evaluate patients. The use of these diagnostic methods involves the identification of microbial challenge and mediators of the host response ( Figure 1).

Research aimed towards new diagnostic tests for periodontal disease has focused on identification of bacterial DNA1, cell surface antigens7 as well as the presence of specific organisms as identified by culture.5 Measurement of the host's immune and inflammatory response is usually performed by analysis of gingival crevicular fluid (GCF), serum, or saliva.4 These tests, perhaps used in combination, appear to offer a sensitive method to determine the risk for active periodontal disease.3 As a result, there are important therapeutic implications of this new understanding about the natural history of periodontal disease. Specifically, it is critical to identify patients at risk for periods of active disease.

DECISION TREES

With this new appreciation of the natural history of periodontal disease, and the options available to the clinician regarding periodontal diagnosis, it is reasonable to suggest that care for the periodontal patient will improve with the use of clinical decision analysis. The practical application of decision analysis is the decision tree. This is a schematic diagram of potential pathways taken in the course of diagnosis and treatment. It is a visual representation of the action being considered by the clinician. Decision trees generally read from left to right. The point from which the tree originates is called the decision node.2 The initial choices that are available to the clinician are the first branches that emanate from the decision node. These branches lead to other choices along a particular pathway.

This type of analysis has been used in medicine for many years, and now should be applied to dental patients. Figure 2 is a decision tree that provides options in the treatment of patients with mild, moderate, or advanced periodontitis. In addition, probabilities can be included in a decision tree. These probabilities allow clinicians to assess the chance of success (or failure) and the risk associated with a particular procedure. These values allow informed decisions to be made.

DISCUSSION

Periodontal disease is a multifactorial disorder affecting the supporting structures of teeth. It is the complex nature of this chronic disorder that requires a clinician to thoroughly investigate each possible therapeutic option to fit the patient's particular need. Patients with mild disease can be evaluated after initial therapy in order to assess the response to therapy. Assuming an appropriate response, the clinician would then decide upon the recall interval. Patients with moderate disease are evaluated after initial therapy and the clinician decides whether to continue with nonsurgical procedures, or if the patient would benefit from anti-infective therapy or periodontal surgery. With the increase in severity of the disease, more options are now available to the clinician. New diagnostic methods would be most beneficial for the patient with advanced disease. The development of decision trees may be helpful for considering options available in the diagnostic and therapeutic phases of periodontal therapy. The "therapy" decision tree would be modified based on findings from the "diagnostic" decision tree. While no specific therapy would be dictated by any diagnostic finding, the identification of low, moderate, or high risk patients would trigger different treatments. These treatments would be aimed at halting progressive disease, and the intensity of treatment regimen would depend on the degree of risk.

CONCLUSION

It is evident that the use of decision analysis can help quantitate therapeutic options. Decision trees offer the clinician a logical approach to tailoring the diagnostic and treatment needs for individual patients.

REFERENCES

1. Dewhirst FE, Paster BJ (1991). DNA probe analysis for detection of periodontopathic bacteria in clinical samples. In: Hamada S, Holt SC, McGhee JR, editors. Periodontal diseases: pathogens and host immune responses. Tokyo: Quintessence 367-377.
2. Kramer MS (1988). Clinical Epidemiology and Biostatistics. Springer-Verlag, Berlin.
3. Lamster IB, Celenti RS, Jans HH, Fine JB, Grbic JT (1993). Current status of tests for periodontal disease. Advances in Dental Research 7: 182-190.
4. Lamster IB, Grbic JT (1995).Diagnosis of periodontal disease based on analysis of the host response. Periodontology 2000 7: 83-99.
5. Listgarten MA (1992). Microbiological testing in the diagnosis of periodontal disease. Journal of Periodontology 63: 332-337.
6. Socransky SS, Haffajee AD, Goodson JM, Lindhe J (1984). New concepts of destructive periodontal disease. Journal of Clinical Periodontology 11: 21-32.
7. Snyder B, Ryerson CC, Corona H, Grogan EA, Reynolds HS, Contestable PB, Boyer BP, Mayer J, Mangan T, Norkus N, Zambon JJ, Genco RJ (1996). Analytical performance of an immunologic-based periodontal bacterial test for simultaneous detection and differentiation of Actinobacillus actinomycetemcomitans, Porphyromonas gingivalis, and Provotella intermedia. Journal of Periodontology 67: 497-505.