A Bayesian approach can be used to estimate the probability that a solitary lung nodule is malignant based on the clinical features. The data provided was limited to males.
Parameters studied:
(1) The age of patient.
(2) The diameter of the lung nodule.
(3) The patient's smoking history.
(4) The prevalence of lung cancer in the population being studied.
|
Parameter |
Finding |
Likelihood Ratio for Malignancy |
|
diameter of nodule |
< 1.5 cm |
0.1: 1 |
|
|
1.5 - 2.2 cm |
0.5: 1 |
|
|
2.3 - 3.2 cm |
1.7: 1 |
|
|
3.3 - 4.2 cm |
4.3: 1 |
|
|
4.3 - 5.2 cm |
6.6: 1 |
|
|
5.3 - 6.0 cm |
29.4: 1 |
|
patient's age in years |
<= 35 years of age |
0.1: 1 |
|
|
36 - 44 years of age |
0.3: 1 |
|
|
45 - 49 years of age |
0.7: 1 |
|
|
50 - 59 years of age |
1.5: 1 |
|
|
60 - 69 years of age |
2.1: 1 |
|
|
70 - 83 years of age |
5.7: 1 |
|
smoking history |
never smoked |
0.15: 1 |
|
|
pipe or cigar only |
0.3: 1 |
|
|
ever smoked cigarettes |
1.5: 1 |
|
|
current smoker, averaging 1-9 cigarettes per day |
0.3: 1 |
|
|
current smoker, averaging 10-20 cigarettes per day |
1.0: 1 |
|
|
current smoker, averaging 21-40 cigarettes per day |
2.0: 1 |
|
|
current smoker, averaging >= 41 cigarettes per day |
3.9: 1 |
|
|
quit smoking <= 3 years ago |
1.4: 1 |
|
|
quit smoking 4-6 years ago |
1.0: 1 |
|
|
quit smoking 7-12 years ago |
0.5: 1 |
|
|
quit smoking >= 13 years ago |
0.1: 1 |
|
overall prevalence |
clinical settings |
0.7: 1 |
|
|
community survey |
0.1: 1 |
odds of malignancy =
= (likelihood ratio for prevalence of malignancy) * (likelihood ratio for size of lesion) * (likelihood ratio for age of patient) * (likelihood ratio for smoking history)
probability of cancer in percent =
= (odds of malignancy) / (1 + (odds of malignancy)) * 100
Specialty: Pulmonology, Hematology Oncology
ICD-10: ,
