br Table br Multivariable adjusted hazard ratios for inciden
Multivariable adjusted hazard ratios for incidents of cancer (n = 177).
Median 71 Reference
Quartile 22 Reference
P for trend
Clinical 20 Reference
P for trend
*Adjusted by age category, gender, alcohol drinking status, BMI category, and physical activity levels as appropriate.
Among all variables associated with CRP levels, BMI is the most significant and consistent one. Our results are in line with the previous literature reports in several different populations, including Mexican American [25–29]. We also found a significant trend of decreasing median CRP levels with the increase of physical activity levels even after adjustment of BMI and other co-variants, indicating that being physical active may reduce Doxorubicin directly. In fact, our results are supported by the findings from other studies [30–34]. In the mul-tivariable analysis, we also found a protective association of alcohol drinking. Compared to never drinkers, current drinkers had lower median CRP levels in the univariate analysis (P < 0.001). The differ-ence was only evident among women. In the multivariable analysis, current drinkers had 21% decreased likelihood of having high CRP le-vels than neve drinkers did (OR = 0.79, 95%CI: 0.60, 0.97). Moderate alcohol drinkers is known to have lower CRP compared to never drin-kers and heavy drinkers [35,36]. Overall, Hispanics are less likely to drink alcohol at all than are non-Hispanic Whites. In fact, Mexican Americans have the lowest rate of binge drinking among all Hispanics. Although the current study doesn’t have the data on the amount and type of alcohol consumed, it is still reasonable to assume that most of the current drinkers in our study only drink lightly or moderately.
Among those born in Mexico, median levels of CRP were gradually increased with years of living in U.S, particularly among men. Such association was further confirmed in the multivariable analysis. Coupled with the observation that median CRP levels were higher among those born in U.S. than their counterpart in both men (P = 0.029) and women (P = 0.076), we can assume that the adoption of American culture and behaviors, such as being overweight or obese, leading a sedentary lifestyle, being lack of sleep, and eating fast or processed foods, during the immigration and acculturation may influ-ence CRP levels. With more accustomed to American lifestyle, the prevalence of such unhealthy behaviors is likely to increase. As shown in this study, both being over-weight or obese and physically inactive are correlated with increased CRP levels. Although we already included BMI, physical activity, cigarette smoking, and alcohol drinking in the analysis, there are still a few potential confounding factors left out, such as sleep and diet. The impact of sleep loss on CRP levels has been re-ported previously . And dietary fatty acid intake has been shown to affect serum CRP levels .
In addition, this study examined the relationship between elevated CRP and incidents of cancer among Mexican Americans who experi-enced cancer at follow-up. We found a significant dose-response trend between increased CRP levels and increased risk of all cancer. Our re-sults are consistent with literature reports from several prospective cohort studies that have shown that elevated serum CRP levels are as-sociated with increased risk of cancer [9–14]. Since our study is a prospective cohort study, we have the ability to determine temporal exposure in relation to disease and evaluate the possibility of reverse causation. As the associations we observe are only evident among cases occurring more than 5 years of follow-up, our results suggest increased levels of serum CRP may be a cancer risk factor, not a biomarker of subclinical cancer in Mexican Americans. Our observations are also consistent with the notion that CRP is a marker for systemic in-flammation.
The major strengths of this study include its large sample size, de-tailed epidemiologic questionnaire data, and unique Mexican American study population. One weakness of our study is that we were unable to obtain repeated measures of CRP over time because one time quanti-fication may not echo CRP over a long period. Nevertheless, our results provide the first evidence in Mexican Americans that higher CRP levels were significantly associated with increased overall cancer risk. Future large-scale and well design prospective studies are required to further validate the observed association of our study. Cancer Epidemiology 60 (2019) 1–7