Need to Pay More Attention to Attendance at Follow-Up Consultation after Cancer Screening in Smokers and Drinkers

BACKGROUND
Follow-up clinical consultations could improve overall health status as well provide knowledge and education for cancer prevention.


MATERIALS AND METHODS
This is the cross-sectional study using the Korean Community Health Survey (KCHS) 6th edition for 2012, with 115,083 respondents who underwent cancer checkups selected as subjects. Associations between the presence of consultation and the socioeconomic status were determined using statistical methods with the SAS 9.3 statistical package (Cary, NC, USA).


FINDINGS
Among the recipients, 32,179 (28.0%) received clinical consultations after cancer screenings. Those in rural areas (odds ratio, OR=0.71, 95% confidence interval (CI), 0.69-0.73) visited follow-up clinics less frequently than did those in urban areas. Starting at the elementary school level, as the education level increased to middle school (OR=1.26, 95% CI: 1.19-1.34), high school (OR=1.29, 95% CI: 1.23-1.36) or college (OR=1.76, 95% CI: 1.65-1.89), the participation rates also increased. When compared with the lowest quartile group, the quartile income level showed a statistical trend and difference as follows: second lowest quartile (OR=1.11, 95% CI: 1.07-1.16), third lowest (OR=1.12, 95% CI: 1.07-1.17) and highest quartile income (OR=1.29, 95% CI: 1.23-1.35). In addition, the people with economic activities (OR=0.87, 95% CI: 0.84-0.90) visited follow-up clinics less frequently than did the others. Current smokers (OR=0.93, 95% CI: 0.89-0.98) and inveterate drinkers (OR=0.88, 95% CI: 0.85-0.94) had a tendency to visit less often than did non-smokers and other drinkers with all cancers combined.


INTERPRETATION
We suggest primary prevention through lifestyle modifications including smoking and drinking, and environmental interventions may offer the most cost-effective approach to reduce the cancer burden.


Introduction
Cancer has been the leading cause of death in Korea since 1983 and is associated with the largest disease burden (Jung et al., 2010;Statistical Korea, 2014).The cancer burden in Korea was estimated as 1,525 person-years per Statistical Korea, 2014).
Although the main purpose for the checkups is early cancer detection, they can also provide many additional health benefits and suggestions for lifestyle modifications.For example, the cancer screening program includes regular blood tests and physical examinations.Accordingly, the follow-up consultations could improve overall health status and provide knowledge and education for cancer prevention and treatment (Cho et al., 2013).Thus, we analyzed the factors associated with clinical consultations following cancer screenings and introduced solutions to increase the follow-up rate.

Materials and Methods
Data from the Korean Community Health Survey (KCHS: Korea Centers for Disease Control and Prevention, KCDC) in 2012 was obtained for this study.In 2008, the KCDC initiated the KCHS, the first nationwide survey to gather data that could be used to plan, implement, monitor and evaluate community health promotion and disease prevention programs.The standardized KCHS questionnaire encompasses a wide variety of health topics which can be used to assess the prevalence of personal health behaviors related to disease causes.
Because we focused on cancer screening in the present study, we only included data from 2012, which had additional information pertaining to cancer.A total of 228,921 people were enrolled in the 2012 survey.Among them, we selected those respondents who had received cancer checkups (115,083 subjects).
Demographic characteristics included gender, age group, basic living status, geographic area, unmet needs, number of chronic diseases, quality of life and other socioeconomic factors.We included stomach, colorectal and liver cancers for both genders and breast and cervical cancers for females.In the latter group, gender was not included in the statistical analysis.
The associations between consultations after cancer screenings and the demographic and socioeconomic status were determined using a chi-square test.To examine the multiple associations between consultations and many factors, we conducted a logistical analysis using the SAS 9.3 statistical package (Cary, NC, USA).

Results
The general characteristics of the recipients enrolled in this study (115,083) are shown in Table 1.Among them, 32,179 recipients (28.0%) underwent a clinical      1).First, we performed the chi-square test to evaluate the results of the bivariate analyses (Tables 1, 2, 3).Among all cancer screening groups, demographic and socioeconomic variables including gender (p=0.006),age group (p<0.001),presence of spouse (p<0.001),living area (p<0.001),educational level (p<0.001),basic living status (p<0.001),income level (p<0.001),economic activities (p<0.001) and health-related variables, such as perceived health status (p<0.001),major depressive disorders (MDDs, p<0.001), number of chronic diseases including hypertension, diabetes, arthritis and dyslipidemia (p<0.001) and quality of life (EQ-VAS), were statistically different, while unmet needs was not.Stomach (p=0.058), and liver (p=0.100)cancer screening groups were not associated with basic living status (Tables 2, 3).Gender  was not statistically associated with cancer type (p=0.306).Interestingly, despite the absence of a statistically significant difference in unmet needs the cancers screened, all organ subgroups evaluated except for liver showed statistical differences (Tables 2, 3).In the cervical cancer screening group, the follow-up consultation was not associated with quality of life (p=0.155).
We also performed the multivariate analysis using logistical models for all cancer screening groups and then for each specific cancer group.Lifestyle and behaviors such as smoking, alcohol consumption and amount of sleep were adjusted (Tables 4, 5).
We also analyzed the effects of health-related factors at the follow-up visits.Perceived health status was correlated negatively with the follow-up visits, in that the recipients who answered their health status as 'excellent' (OR=0.53,95% CI: 0.48-0.59)tended to follow-up the least.The group with MDDs (OR=1.22,95% CI: 1.13-1.31)diagnosed by doctors, chose to visit more frequently.The number of chronic diseases was related positively to the number of visits.For example, the recipients with one (OR=1.19,95% CI: 1.15-1.23)or two or more (OR=1.34,95% CI: 1.29-1.40)chronic diseases visited more than those without chronic disease.The recipients with unmet needs (OR=0.96,95% CI: 0.92-1.00)did not visit as often as those without, and this was statistically different.Lastly, the recipients with the best quality of life (OR=1.06,95% CI: 1.02-1.11)visited the follow-up clinics more often than did those with a poorer quality of life.
The results were nearly the same among cancers affecting both genders, such as stomach, hepatocellular carcinoma (HCC) and colorectal cancers, in terms of statistically meaningful independent covariates.However, regarding colorectal cancer, the age group from 65-74 years (OR=1.07,95% CI: 0.99-1.16)did not show a statistical difference regardless of statistical trend.In the female-specific cancers such as cervical and breast, the presence of unmet needs was not associated with the follow-up visits, but in all other aspects, the results were similar to those of the all cancers combined group.
To determine their association with follow-up visits, the logistical outcomes were analyzed in greater detail (Figure 2).Current smokers (OR=0.93,95% CI: 0.89 -0.98) had a tendency to visit less often than did non-smokers in the all cancers combined and male groups.Regarding alcohol consumption, there was a dose-dependent relationship among the all cancers combined and female groups; the male group was not associated with social drinking (defined as the amount equivalent to two or three times per month or less), and the female group was not associated with smoking.results of this study, the follow-up visits in males were related to smoking status.In contrast, smoking did not have any influence in females, and among males, ex-smokers visited more often than did non-smokers.However, the number of follow-up visits among females was related to alcohol consumption in a dose-dependent manner.Overall, smoking and alcohol use were negatively associated with the frequency of follow-up consultations

Discussion
Based on the results from this study, the recipients with low socioeconomic status, such as low education and income levels, those with unmet needs, and those living in a rural area tended to visit follow-up clinics less frequently.Moreover, regarding health-related factors, a greater number of chronic diseases, presence of MDDs and a high quality of life were associated with more frequent follow-up visits to the clinic.
Several studies have been conducted on the factors associated with cancer screening.Kang et al. used data from the Korea National Health and Nutrition Examination Survey (KNHANES), a cross-sectional nationwide study (Shin and Lee, 2012;Kang et al., 2014).In terms of education level, disparities in attendance were observed for the opportunistic screening program.In another study, Lee et al. conducted a study regarding factors associated with the use of breast cancer screening services by females in Korea (Lee et al., 2010).They concluded more attention should be given to under-represented groups, particularly the elderly, those with a low education level, smokers and those with a negative attitude towards screening tests.In our study, similarly, the independent variables associated with low socioeconomic status, such as low income level and low education level, were negatively correlated with the dependent variables.Because no previous study has evaluated the factors associated with follow-up consultation after cancer screening in Korea, confirming our hypothesis and comparing our results with other studies are difficult.
However, we believe that education regarding lifestyle modifications, including smoking and alcohol use, is more important for cancer prevention.In fact, population-based efforts to lower tobacco use, mainly cigarettes, have reduced the lung cancer rates (Bamoya and Glantz, 2004;Thun and Jemal, 2006;Jemal et al., 2008;Glantz and Gonzalez, 2012).A 2008 U.S. annual report to the nation showed a decline in the incidence and death rates of all cancers and reasoned that reductions in tobacco use was the largest single factor preventing nearly one-third of cancer deaths by utilizing the existing knowledge.Another U.S. study concluded that reductions in tobacco use in the last half of the 20 th century accounted for approximately 40% of the decrease in overall male cancer death rates and prevented at least 146,000 lung cancer deaths in males between 1991and 2003(Jemal et al., 2008)).The gold standard for cessation treatment is the five A's (Ask, Advise, Assess, Assist, And Arrange) (Schroeder, 2005).Knowledge that most smokers require multiple attempts before they succeed, that rigorous studies show long-term quit rates of 14-20%, with one report as high as 35%, that cessation rates for users of telephone quit lines and integrated health care systems are comparable with those using individual clinicians, and that no other clinical intervention can offer such a large potential health benefit may help counter clinicians' pessimism regarding cessation.In our study, the current smokers visited followup clinics less often than did non-smokers, although ex-smokers visited more often than did non-smokers.However, to accomplish the objective of cancer screening programs, current smokers should be encouraged to attend the follow-up visits, obtain information on smoking cessation, and enroll in a quit smoking program.
Regarding alcohol use, the association between drinking alcohol and cancer has been demonstrated.A total of 21,201 Japanese males completed a self-administered questionnaire on various health habits, including alcohol consumption.The risk for any cancer was significantly higher in ex-drinkers than never-drinkers.A dose-response relationship between the amount of alcohol consumed and the risk of cancer among current drinkers showed 17.9% (95% CI 3.1-30.5) of cancer risk was attributable to drinking habits.
Additionally, a well-established between cancer incidence and socioeconomic deprivation has been demonstrated; poor socioeconomic groups have high of cancer (Limb, 2014).Ramsay et al. studied socioeconomic inequalities among cancer mortality cases in Britain between 1978 and 2013 (Ramsay et al., 2014).The hazard ratio for cancer mortality between manual and non-manual social classes remained unchanged: 1.62 (95% CI 1.17-2.24)from 1980-1990 and 1.65 (95% CI 1.14-2.40)from 1990-2000 among males aged 50-59 years.The absolute difference (non-manual minus manual) in probability of surviving cancer at 70 years remained at 3% over the follow-up period.The consistency of risks over time was similar between smoking-related and nonsmoking-related cancer mortalities.Another study was conducted in France on educational inequalities between males and females and cancer mortality in the 2000s (Menvielle et al., 2013).In this study, significant relative inequalities were found among males for all cancers and for smoking and/or alcohol-related cancer mortality.Furthermore, this disparity could induce different treatment outcomes.Mahdi et al. investigated racial disparities among the 30-day morbidity and mortality rates after surgery for endometrial cancer (Mahdi et al., 2014).According to this study, African American patients with endometrial cancer had more preoperative morbidities and postoperative complications and were less likely to undergo minimally invasive surgery.Although European American females are more likely to be diagnosed with breast cancer than are African American females (Office of Public Health Statistics and Information Services DoHaEC, 2006), the latter are more likely to die from breast cancer than the former (Ries et al., 2005).In addition, the average cancer stage at diagnosis was significantly higher in African American than European American females.
Therefore, controlling cancer-related lifestyles, such as smoking and drinking, especially in patients with low socioeconomic status are important.However, this group of recipients was not willing to attend the follow-up consultations, which contribute to lifestyle modifications and public health education related to cancer prevention.
Health education could significantly improve healthrelated behaviors in cancer patients, resulting in more favorable outcomes (Goss et al., 2014).Marek et al. (2012) evaluated the effect of an educational intervention on Hungarian adolescents' awareness, beliefs and attitudes on the prevention of cervical cancer.A self-administered anonymous questionnaire was completed by 394 male and female adolescents in September 2010 in Hungary.Half of the students (48.5%) received a one-on-one educational intervention regarding cervical cancer, consisting of a 45min lesson, while the remaining participants (the control group) did not receive the educational intervention.Three months following the intervention, both groups were retested using the same questionnaire.A significant increase in cervical cancer awareness was detected.Health-related beliefs were enhanced, such as 'HPV may cause cervical cancer' (64.9%"81.0%,p<0.05) or 'cervical cancer may be prevented by vaccination' (66.5%"85.3%,p<0.05).Health education may also increase the cancer screening rates of participants.Huang et al. performed a community-based 2-year health educational intervention, after which 24.5% of the 1,041 respondents underwent a fecal occult blood test and 12% a colonoscopy examination during the study period, both substantially higher than the background screening rate in Shanghai (Huang and Shi, 2011).
Furthermore, easier accessibility to health clinics is needed in rural areas.In the last 20 years, public health has improved significantly in Korea.As a result, the disparity in accessibility to medical care between urban and rural areas has decreased dramatically.However, some inequity remained in this study due to geographic inconvenience.Thus, we suggest another venue is necessary to provide more information on current health status and education for cancer prevention.Fortunately, the emergence of hightech communication devices, such as smartphones and smart TVs may, be the most cost-effective distribution method in rural areas (Wentzensen and Schiffman, 2014) In Korea, if the patients are unable to follow-up, physicians traditionally report the results of cancerscreening programs by telephone or mail with only a "yes" or "no" regarding the presence of cancer.Therefore, the reporting system for results, including health education, needs to be upgraded from that of minimal information reporting.
There were several limitations in this study.First, due to its cross-sectional design, causal relationships could not be determined.To address this issue, we intend to analyze this topic further using a panel survey.Second, the cancer screening results could not be determined, and if a patient's results are negative for cancer, they potentially may not consider the clinic follow-up visit necessary.Consequently, we could not measure the unmet needs, which may require further cancer evaluation for suspicious positive results, but do not involve additional follow-ups.Investigating the actual unmet needs among different socioeconomic classes and lifestyles may prove more meaningful.
In conclusion, To the best of our knowledge, this is the first study to determine the associated factors to followup consultations after establishing a cancer screening program in Korea.Patients with low socioeconomic status and living in rural areas visited follow-up clinics less often.Furthermore, we believe that primary prevention through lifestyle modifications and environmental interventions addressing various health determinants may offer the most cost-effective approaches to reducing the cancer burden.Because the several risk factors discussed here are common to other diseases, controlling such risk factors may have a positive impact on a population's health.

Figure 1 .
Figure 1.The Participation Rates of Follow-Up Consultation after Cancer Screening Examinations by Total and Each Cancer Group.Participation rate for each

Figure 2 .
Figure 2. Adjusted Odds Ratio (OR) for the all Cancers Combined and Specific Gender Groups.According to the

Table 1 . Demographic, Socioeconomic and Health- Related Characteristics between Groups in the all Cancers Combined Screening Group
*life style behaviors such as smoking, drinking, sleep hours, and stress were adjusted Need to Pay more Attention to Attendance at Follow-Up Consultation after Cancer Screening in Smokers and Drinkers Participation rate for each cancer was almost same to the total one, regardless of different numbers of participants DOI:http://dx.doi.org/10.7314/APJCP.2015.16.1.109

Table 2 . Demographic, Socioeconomic and Health-Related Characteristics in Cancer Screenings Both Gender Groups (Continue)
Need to Pay more Attention to Attendance at Follow-Up Consultation Cancer Screening in Smokers and Drinkers DOI:http://dx.doi.org/10.7314/APJCP.2015.16.1.109