This screen also contains the data input fields for smoking history and diabetes. The green bar represents deletion of smoking and diabetes from the algorithm. …
Development and Design of A Risk Calculator
for Future Periodontal Disease
John A Martin, DDS
Roy C Page, DDS, PhD
[SLIDE - Purpose]
Our knowledge and understanding about periodontal diseases have greatly
increased during the last several decades We now know that risk and
susceptibility varies between individuals and for an individual over time
coincident with a change in an individuals risk factors The application
of this knowledge would be expected to drive an expansion of treatment that
currently is based on repairing lesions to treatment that additionally
incorporates managing risk factors The effect of applying risk concepts
in the formulation of treatment decisions could be expected to result in a
decreased incidence of periodontitis, a significant reduction in the
periodontal disease treatment burden in the population and reduction in the
cost of care Success of this enhanced treatment approach is dependent in
large measure on the ability of practitioners to accurately assess risk and
institute risk reduction as an integral part of prevention, treatment and
maintenance A
problem is that there is no standardized validated method
to quantify risk and risk factors to guide treatment decisions The
purpose of this presentation is to describe the development and design of a
tool to calculate the risk for future periodontal disease The next two
sessions will report on the validity and accuracy of this new tool
[SLIDE - Design Parameters]
We used thirteen design parameters to develop the risk model
1 The model calculates the risk for future periodontal disease for those
patients who do not yet have it and risk for future progression of
periodontal disease for those who already have it The risk assumes the
patient will have no treatment
2 A risk factor was defined as a factor that is part of the causal chain
of disease, or exposes the patient to the causal chain, which if present
directly increases the probability of disease occurring and if absent
reduces the probability
3 A risk factor must have a scientific basis that is supported by
publication in refereed scientific journals
4 The application of risk assessment information through the development
of treatment recommendations to reduce risk must occur
5 All
requisite information must be obtained during a traditional
periodontal examination
6 The time required for data collection and input must fit within the
usual time allocated for a traditional periodontal examination
[SLIDE - Design Parameters, continued]
7 Risk is to be expressed using a 5-point scale We chose a 5-point scale
to balance the sensitivity of risk assessment with the time and cost
required to obtain the necessary information We felt that the 3-point
scale used in caries risk assessment trials could be expanded for
periodontal disease to improve treatment specificity Further expansion
to a 10-point scale we thought would require too many data points
8 A risk score must have the same meaning for all patients at all times
9 The use of a risk factor in our model is based on its weight and its
impact on the risk score
10 Risk factors do not necessarily have a linear relationship to risk
11 Risk factors do not have the same magnitude of influence on risk
12 A risk factor can potentiate or negate the effect of another risk
factor
13 A risk factor to be operative might require the existence of another
factor
The risk assessment
tool is computer-based A report that includes the
risk score, risk factors, risk-based treatment suggestions, and diagnosis
is generated
[SLIDE - Factors]
Twelve factors are used They are grouped into 3 categories - history,
clinical and radiographic The factors that we used are:
Patient age
Smoking history
Diagnosis of diabetes
History of periodontal surgery
Dental care frequency
Pocket depth
Furcation involvements
Restorations or calculus below the gingival margin
Bleeding on probing
Radiographic bone height
Furcation involvements
Vertical bone lesions
Radiographic evidence of root surface calculus
Two of the factors, periodontal pocket depth and radiographic bone height,
each require one value per sextant for 12 total data points The remaining
ten factors each account for 1 data point The total number of data points
is 22
[SLIDE - Screen View Clinical Data 1]
This is a view of the first of two clinical data entry screens
Surrounding the tooth diagram are the data entry fields for pocket depth
This screen also contains the data input fields for smoking history and
diabetes
The value for pocket depth is the deepest pocket in the sextant recorded
as
5 mm, 5 - 7 mm, and 7 mm Measurements are from the gingival margin to
the base of the pocket taken at the classical 6 sites for each tooth
[SLIDE - Screen View Clinical Data 2]
This is a view of the second of two clinical data entry screens
Surrounding the tooth diagram are the data entry fields for bone height
The value for bone height is the greatest distance from the cemento-enamel
junction to the bone crest in the sextant recorded as up to 2 mm, 2-4 mm,
and 4 mm or greater
[SLIDE - The Risk Algorithm]
The risk algorithm is a 7-step process The base risk score is calculated
using a method that quantifies the severity of periodontal disease adjusted
by age The risk score is increased if there is a positive history of
periodontal surgery, if the patient smokes more than 10 cigarettes per day,
or the patient has diabetes that is poorly controlled The existence of
furcation involvements, vertical bone lesions, or subgingival restorations
or calculus increase risk when risk is 4
[SLIDE - Deletion Trials]
To determine the effect of each factor on outcome, we compared the results
of using all of the factors to the deletion of one or more We used data
on 523 subjects that
were enrolled in the VA Dental Longitudinal Study
This data set was also used to test for validity and accuracy Dr Roy
Page and Dr Lloyd Mancl will present these findings A risk score was
calculated for each subject for each deletion trial We evaluated the
effect of deleting:
1 Smoking and diabetes
2 Pocket depth
3 Bone height
4 Furcation involvements, vertical bone lesions, and subgingival calculus
5 Smoking, diabetes, furcation involvements, vertical bone lesions, and
subgingival calculus
[SLIDE - Deletion Trials bar graph]
The vertical axis is the number of subjects The horizontal axis is
divided into the five risk score groups Each trial is represented by a
different color in the bar graph The yellow bar represents the control
Regardless of which factor is eliminated risk group 5 is not affected by
the deletion of any factor except for pocket depth The green bar
represents deletion of smoking and diabetes from the algorithm This
decreases the risk score as shown by the slightly increased number of
subjects in risk groups 2 and 3 The dark blue bar is when bone height is
not used This reduces the risk score and notably shifts a large number of
subjects to risk
groups 1 and 2 The beige bar is when we deleted all
except pocket depth and bone height This eliminated all subjects from
risk group 4, halved the number from group 3, and tripled the number in
group 2
[SLIDE - Scorecard Summary Report]
Our model creates informative and actionable information, which is
presented in a scorecard summary The information includes:
A numerical and descriptive diagnosis
The risk score
The risk factors
Changes that occurred if the calculations were done for 2 time periods
Risk-based treatment suggestions
Rules for treatment can be based on this information For example high
risk could justify treatment urgency, aggressiveness, and more frequent
periodontal maintenance Treatment could be targeted to the factors It
is logical that treatment that eliminates risk factors would reduce risk,
which would result in less disease and less reparative treatment
[SLIDE - Conclusions]
We designed a computer-based tool that calculates the risk for future
periodontal disease Its use can be easily incorporated into the clinical
setting Only 22 data points are required and they are obtained during a
traditional periodontal examination A report is
generated that includes a
risk score, risk factors, risk-based treatment suggestions and numerical
and descriptive periodontal diagnosis Risk scores can be used to guide
treatment urgency and aggressiveness Changes in risk status can be used
to measure treatment effectiveness Use of this risk assessment tool over
time may be expected to result in more uniform clinical periodontal
decision making, a reduction in disease incidence, improved oral health, a
significant reduction in the need for complex periodontal treatment and
reduction in the cost of care
Source:saut.ac.tz
