Data mining approach for exploring socioeconomic patterns in cancer
- Leila Hosseini ,
- Hossein Vatanpour ,
- Mehdi Mohammadzadeh ,
- Mohamadreza Abdolahi ,
- Rita Motidostkomleh ,
Background: Cancer has become an increasingly important issue for health expenditure and in the public sector. Therefore, identification of socioeconomic patterns is essential for developing novel methods of cancer prevention. A cancer risk prediction system is proposed here, which is easy and cost-effective, also saves time.
Methods: Initial data were collected from surveys with a digital audio recorder on 2014 cancer dependent and independent smokers. Patterns were found using a decision tree model. As socioeconomic patterns are neglected altogether, this study will pave the way for public health policies.
Results: This study shows that a death rate in cancers by smoking around 24.3%. According to them, the type of cancers attributed to smoking was oral cavity (8.2%) and bladder (8.2%) cancer. Some risk factors related to smoking and cancers were included male patients (66%), age <35 years old group (56.3%), employee (86.3%), married (86.2%), as well as have very good public insurance (96.5%). In addition, 98% cancer patients were known have public insurance.
Education levels have been shown most of the patients with cigarette smoking had less than a college education and in the absence of police, they use fewer safety belts.
Those with risk factors for failing to wear a seat belt under a doctor to quit smoking. Although
The health situation in cancers attributed to smoking vs. non-smoking attributed to worse evaluated.
The age factor is effective in counseling for the doctor to quit smoking, and people aged 35-65 are the most advised to quit smoking.
Conclusion: Smoking-related costs in Iran are high. The cost methodology presented is useful for policy making. Cost estimates can be used to evaluate the level of cigarette taxes and other policies related to smoking.