Factors associated with smoking cessation using the 2005 National Health Interview Survey

Objectives. The objective of this study is to compare the socio-demographic, behavioral, and access to care characteristics of smokers who have quit smoking for one or more years and current smokers who have made an attempt to quit smoking within the last year. ^ Methods. Data from the 2005 National Health Interview Survey (NHIS) were used to compare current smokers who have tried to quit (n=2747) and former smokers who have quit for one or more years (n=6194). The data was analyzed using STATA 9.0 to perform statistical calculations. ^ Results. Age, education, race and income were associated with smoking status. Respondents aged 65 and older were 36 times more likely to have quit smoking. Education and income had higher odds ratios among quitters (OR=1.27 and OR=1.21) and Non-Hispanic Whites were the most likely to have quit smoking compared to Hispanics and Blacks. Adults with health insurance coverage were 3.44 times more likely to have quit smoking. ^ Discussion. Existing research suggests that individual factors relating to demographics behavior and access to care can impact a smoker's ability to quit smoking. This paper discusses the factors that affect cessation and which populations would benefit from additional research and targeted smoking cessation programs. ^

A STUDY OF INFLUENZA AND OF STATISTICAL METHODS USED TO REPORT INFLUENZA

This paper defines and compares several models for describing excess influenza pneumonia mortality in Houston. First, the methodology used by the Center for Disease Control is examined and several variations of this methodology are studied. All of the models examined emphasize the difficulty of omitting epidemic weeks.^ In an attempt to find a better method of describing expected and epidemic mortality, time series methods are examined. Grouping in four-week periods, truncating the data series to adjust epidemic periods, and seasonally-adjusting the series y(,t), by:^ (DIAGRAM, TABLE OR GRAPHIC OMITTED...PLEASE SEE DAI)^ is the best method examined. This new series w(,t) is stationary and a moving average model MA(1) gives a good fit for forecasting influenza and pneumonia mortality in Houston.^ Influenza morbidity, other causes of death, sex, race, age, climate variables, environmental factors, and school absenteeism are all examined in terms of their relationship to influenza and pneumonia mortality. Both influenza morbidity and ischemic heart disease mortality show a very high relationship that remains when seasonal trends are removed from the data. However, when jointly modeling the three series it is obvious that the simple time series MA(1) model of truncated, seasonally-adjusted four-week data gives a better forecast.^