EPIDEMIOLOGY : Cancer as the Public Health Issue | Sample Case Study

Table 1: Cross-sectional study

  1. In the study of Seyed et al., (2016), the main inclusion criteria for selecting or collecting data is patients having any type of cancer. Apart from this, the patients are also needed to live in Sothern Iran. In the study of Badar et al., (2016), a cross-sectional study has also been held for estimating cancer at the population level for the Lahore district. Hence, it can be said that the inclusion criteria are evident.
  2. Both the subject and design of the entire study have been described in detail. The study subject is based on the evaluation of the changing epidemiology of different common cancers in Southern Iran. The details of the study designs and subjects have also been discussed in another research project which is also evidence for this study (Sugisaki et al., 2018).
  3. All the collected data have been calculated based on the age-standardized rates per 100,000 person-years(ASRs) following this particular time period. Therefore, it can be stated that the way of calculation is significant.
  4. The main objective of the study has been the evaluation of the changing epidemiology of the most common types of cancers in the Fars province of Iran. As per the need of the measurement, the age-standardized rates per 100000 per year have been calculated. The whole research has been done in a cross-sectional manner and the time period that has been considered for the study is 2007-2010 (Seyed et al. 2016). Age standardization is effective in epidemiological research for comparison in a study where the age profiles of the participants are different (Chikandiwa et al. 2019).
  5. Age and gender are the cofounders which have been considered in this research. Other cofounders are the common cancer tumour sites which are breast, stomach, lung cervix and other common sites. The incidence rate of the top ten cancers has been compared with the previous cancer registry and the possible causes of the cancers have been tried to find out by a cross-sectional study (Seyed et al., 2016).
  6. Age, gender and the top ten cancer sites are the confounding factors upon which the whole study has been conducted. More precisely the result has presented the incidence rate of sex, crude rate, site and ASR of the 100000 people per year basis. How the onset of the common cancers has changed in that particular area of Iran over the years has been evaluated and compared with the data of previous year ranges. It has been observed that the epidemiology of common cancers has significantly changed due to certain reasons throughout Iran over the years (Farhood et al, 2018).
  7. For calculating the ASR of the cancer data, the collected information within 2007-2010 from the hospitals, clinics, central death registry and pathology clinics have been collected. The data have been summarized in MS Excel and the duplicate data have been removed from the list. The data have been computerized using SPSS software which is efficient statistical software and applying the direct standardisation method of the world population, the incidence rate has been calculated. A standardised rate ratio has been used for comparison(Seyed et al., 2016). Sung et al. (2018), in their study, also used a standardised rate ratio for comparing the incidence of lung cancer among different metropolitan cities of Korea.
  8. In the cross-sectional study, the researchers handled a huge amount of data. The main aim of the study has been to estimate the instance of different confounding factors with the changed epidemiology of the top ten common cancers in the Fars locality of Iran (Seyed et al., 2016). A standardised rate ratio is used for the application of a set of rates for estimating the prevalence of certain events. This is also helpful in estimating the morbidity or mortality ratio (Giun-Yi-Hung 2019). A standardised rate ratio is a proper statistical approach for comparing the data to time spans. Therefore, it can be said that proper statistical analysis has been done in the mentioned research.

Table 2: Case-Control Study

  1. The selected case-control study has focused on a particular health issue which is colorectal cancer in females. This particular study has focused on the identification of the link between shift work with the risk of developing colorectal cancer in females (Walasa et al., 2018). The study of Sormunen et al., (2016) has also aimed to identify the association between rectal cancer and physical activity. Based on this evidence, it can be stated that the concern of this study about this particular issue is significant.
  2. In this selected study, the author has followed a case-control method for collecting and analysing data to obtain the results of the specific research questions. In this context, 350 cases of colorectal cancers and 410 controls have been selected as participants. The collected responses or data have then been analysed statistically to meet the objectives of this study. A questionnaire has been developed to collect data from the respondents. It has also been seen that in the study of Sormunen et al., (2016), the research has also used case-control methods to know whether there is a link between physical activity and rectal cancer or not. So, the selection of this method for this particular research is evident.
  3. In this study, the data has been collected from the Western Australia Bowel Health Study(WABOHS). In this WABOHS, the cases had been selected based on the criteria of having colorectal cancer and ages between 40 to 79 years. These cases have been confirmed diagnosed with colorectal cancer by regarding Western Australian Cancer Registry from June 2005 to August 2007. On the other hand, the study of Turati et al., (2017) also recruited the cases by collecting data from 665 hospital controls and 690 bladder cancer cases from an Italian case-control study. Therefore, it can be said that the way of recruiting the cases in this study is acceptable.
  4. In this research project, the controls have also been selected randomly from the Western Australian electoral roll. These controls do not have colorectal cancer. These selected controls have matched with the sex and age of the cases frequently. In the study of Sritharan et al., (2017), data have been collected from the National Enhanced Cancer Surveillance System and the controls have also been selected randomly. On the basis of this evidence, it can be stated that the way of selecting controls in this selected study is acceptable.
  5. It has been sated by the researcher of this study that the use of a job exposure matrix (JEM)for categorising exposure to shift work also helps to make standardisation of the exposure definition. This matrix is also very effective to eliminate reporting bias as these biases can be associated with self-reporting job histories sometimes. Apart from this, sensitivity analysis has also been done using the data to minimise biases. In this study, Graveyard shift work has been considered as the primary exposure. In another study, the data have been collected in more than 1 year to remove bias (Boursi et al., 2016).
  6. As in this study, the data have been gathered from WABOHS, there is no issues related to treatment. No partiality has been taken place at the time of selecting controls and cases as well as collecting their data. It has been seen in another study has data have collected in this similar way for conducting case control study (Sritharan et al., 2017).
  7. Certain demographic factors have been taken into consideration as confounding factors by the authors of this study. These factors mainly include socioeconomic status, education, lifetime alcohol consumption and cigarette smoking. In the study of Shin et al., (2020), these same confounding factors have been considered.
  8. From this research result, it has been found that working in a particular occupation involving long-term exposure (more than 7.5 years) to graveyard shift work is not linked with risk of colorectal cancer. The treatment has lasted for 2 years as WABOHS has been held from 2005 to 2007.
  9. A statistical analysis including regression has been performed to estimate the effect of treatment. This regression analysis has been done for calculating odds ratios (OR) as well as corresponding 95% confidence intervals (CI) to the association between colorectal cancer risks and different shiftwork variables. In the study of Di Maso et al., (2016), statistical analysis has been done to evaluate relation between bladder cancer risk and dietary water intake.
  10. I believe this result because authentic International Standard Classification of Occupation(ISCO-68) coding system, JEM process has been used here.
  11. These results can be used in case cases of local population because it has become evident that the risk of colorectal cancer is not associated with shift work (Sormunen et al., 2016).
  12. There are many evidences which are fitted with the results of this selected study. The study of Boursi et al., (2016) and Turati et al., (2017) are the significant evidences for this research results.

Table 3: A cohort study

  1. The main aim of the cohort study has been to investigate the association of breast cancer with components of metabolic syndrome. Also, the risk of breast cancer among premenopausal and postmenopausal women. The case-control study has been conducted on 22494 women who were recruited in four Italian centers for a time-span of 1993-1998 of EPIC. The follow-up has been done for 15 years. From the vast population a sub cohort of 565 women has been randomly obtained and them the cases of breast cancer are 593. The study has been conducted for measuring the harmful effect of metabolism syndrome on cancer onset. The outcomes reflect that only high blood glucose level has significant interaction with breast cancer among the post-menopausal women (Agnoli et al., 2015). The study of Lohmann et al., (2017), also indicates that metabolic factors, especially blood glucose has a significant impact on breast cancer initiation.
  2. No selection bias has been observed in the cohort. The test cohort, the sub cohort of 565 women has been chosen randomly and the 593 cases of reported breast cancer have also been chosen. However, in the random and cancer diagnosed cohort, the pre- and post-menopausal separation has not been maintained (Agnoli et al., 2015). The separation could have been included in the study. Otherwise, the cohort selection is free of bias.
  3. The cohort study has sed both the subjective and objective measurements. One hand, it has tried to find out the interaction of breast cancer with metabolic syndrome. On the other hand, the study has involved the comparison of breast cancer onset among the pre-menopausal and post-menopausal women. The biomarkers have been adjusted perfectly so the bias can be reduced. A confidence interval of 95% has been applied. However, the estimated outcomes have not been achieved fully, only significant relationship of blood glucose and breast cancer has been observed (Agnoli et al., 2015). 95% confidence interval is the reliability indicator of the outcomes (Amsi, 2020).
  4. As already mentioned in the above part, both objective and subjective measurements have been appliedin the research outcomes. All the analyses have been done based on the comparison of the plasma samples of the participants and it is known that metabolic factors can be detected from the plasma sample very effectively. For the cases a reliable system and same conversion factors have been applied. Cox hazard model weighted by Prentice in association with age variables underlining time variables have been used for assessing the interaction of metabolic syndrome components and breast cancer risk(Agnoli et al., 2015). Cox proportional hazard model is a proper measurement framework for estimating the risk of cancer onset and it has been properly applied here (Hossain et al., 2015).
    1. The confounding factors that have been considered for the analysis part are age, BMI, center specification, menopausal status of the women, menarche age, parity, smoking pattern of the participants, education, alcohol consumption and total physical activities. The cohort study based on the prospective design has been served as one of the basic strengths of the study (Agnoli et al., 2015).
    2. The prospective design has been helpful for conducting the statistical analysis in the cohort study. As the study sample is large and the long-term impact of the factors have to be assessed, therefore, cohort study has been justifiably done here over any other study designs. The Cox hazard model has been run for the total cohort and the cohorts of premenopausal and post-menopausal women. The diabetic women have been excluded from the models as an aspect of sensitivity analysis (Agnoli et al., 2015). Referring to Kreike et al., (2010), it can be said that sensitivity analysis makes the implementation of Cox model more effective.
    1. The women of the cohort study have been followed up from their entry until cancer diagnosis, emigration, death or end of follow-up period. The follow up has been ranged for 15 years. The end of follow-up sessions has been varied for different centers;therefore, an adjustment of different follow up timings has been done. Among 565 cohort sample, 16 women have been diagnosed with cancer during the time of follow-up (Agnoli et al., 2015).
    2. The follow up time for the mentioned cohort study is 15 years which is long time. As mentioned by Wright & Moorin, (2020), the longer the follow-up term of a cohort study, the efficient it is to track the changes.
  5. The Cox hazard model has been applied on total cohort, and the cohorts of premenopausal and post-menopausal women. Comparing the fully adjusted models it can be said that in the 95% confidence interval only the presence of high glucose level has been closely interrelated with breast cancer on post-menopausal women and all women but no significant impact has been observed among the premenopausal women (Agnoli et al., 2015).
  6. The result of the study is precise in nature as the 95% confidence level has been applied here and the comparison of the three well demarcated models have been done.
  7. 9.
  8. The article of Simon et al. (2018) reflect that hyperglycemia is related with cancer driven deaths. It is known that high glucose content induces cell proliferation. Therefore, I agree with the test result which shows significant relationship among blood glucose and breast cancer onset (Agnoli et al., 2015).
  9. 10.
  10. Yes, the result can be applied to the local population as the study has included the women of both pre-menopausal and post-menopausal conditions. Therefore, the outcomes can be applied on local population perfectly.
  11. 11.
  12. The study indicates that no other metabolism syndrome components other than high glucose level has interaction with breast cancer (Agnoli et al., 2015). Post-menopause the insulin activity of the body gets hampered and it can give rise to the glucose content of the blood. Hyperglycemia has been identified as an important factor of cancer related morbidity (Simon et al., 2018). Therefore, it can be said that the outcome of this research is supported by other evidences.
  13. 12.
  14. The evidence of a single study is not enough to reach a final outcome. However, the impact of hyperglycemia on cancer morbidity has been identified in another evidence (Simon et al., 2018). Considering the key essence of the outcomes of Agnoli et al., (2015), it can be said that reducing the blood glucose level among post-menopausal women can reduce the risk of breast cancer.


Agnoli, C., Grioni, S., Sieri, S., Sacerdote, C., Ricceri, F., Tumino, R., . . . Krogh, V. (2015). Metabolic syndrome and breast cancer risk: A case-cohort study nested in a multicentre italian cohort. PLoS One, 10(6) doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1371/journal.pone.0128891

Amsi (2020). Calculating confidence interval. Retrieved from: http://amsi.org.au/ESA_Senior_Years/SeniorTopic4/4g/4g_2content_8.html

Badar, F., Mahmood, S., Yusuf, M. A., & Sultan, F. (2016). Epidemiology of cancers in lahore, pakistan, 2010-2012: A cross-sectional study. BMJ Open, 6(6) doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1136/bmjopen-2016-011828

Boursi, B., Mamtani, R., Haynes, K., & Yang, Y. (2016). Parkinson’s disease and colorectal cancer risk–A nested case control study. Cancer Epidemiology, 43, 9-14. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2016.05.007

Chikandiwa, A., Pisa, P. T., Sengayi, M., Singh, E., & Delany-Moretlwe, S. (2019). Patterns and trends of HPV-related cancers other than cervix in south africa from 1994–2013. Cancer Epidemiology, 58, 121-129. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2018.12.004

Di Maso, M., Bosetti, C., Taborelli, M., Montella, M., Libra, M., Zucchetto, A., . . . Polesel, J. (2016). Dietary water intake and bladder cancer risk: An italian case-control study. Cancer Epidemiology, 45, 151-156. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2016.09.015

FARHOOD, B., GERAILY, G., & ALIZADEH, A. (2018). Incidence and mortality of various cancers in iran and compare to other countries: A review article. Iranian Journal of Public Health, 47(3), 309-316. Retrieved from http://ezproxy.laureate.net.au/login?url=https://search-proquest-com.ezproxy.laureate.net.au/docview/2016980500?accountid=176901

Giun-Yi Hung, Jiun-Lin Horng, Paul Chih-Hsueh Chen, Li-Yih, L., Jinn-Yang, C., Pei-Hung Chuang, . . . Chueh-Chuan Yen. (2019). Incidence of soft tissue sarcoma in taiwan: A nationwide population-based study (2007–2013). Cancer Epidemiology, 60, 185-192. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2019.04.007

Hossain, M. J., Xie, L., & Caywood, E. H. (2015). Prognostic factors of childhood and adolescent acute myeloid leukemia (AML) survival: Evidence from four decades of US population data. Cancer Epidemiology, 39(5), 720-726. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2015.06.009

Kreike, B., Hart, G., Bartelink, H., & van de Vijver, M.,J. (2010). Analysis of breast cancer related gene expression using natural splines and the cox proportional hazard model to identify prognostic associations. Breast Cancer Research and Treatment, 122(3), 711-20. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1007/s10549-009-0588-6

Lohmann, A. E., Ennis, M., Taylor, S. K., & Goodwin, P. J. (2017). Metabolic factors, anthropometric measures, diet, and physical activity in long-term breast cancer survivors: Change from diagnosis and comparison to non-breast cancer controls. Breast Cancer Research and Treatment, 164(2), 451-460. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1007/s10549-017-4263-z

Seyed, M. M., Lankarani, K. B., Honarvar, B., Tabatabaee, S. H., Moghadami, M., & Khosravizadegan, Z. (2016). Changing epidemiology of common cancers in southern iran, 2007-2010: A cross sectional study. PLoS One, 11(5) doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1371/journal.pone.0155669

Shin, C. M., Kim, N., Han, K., Kim, B., Jin, H. J., Tae, J. O., & Lee, D. H. (2020). Anti-diabetic medications and the risk for colorectal cancer: A population-based nested case-control study. Cancer Epidemiology, 64 doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2019.101658

Simon, J., Thomas, F., Czernichow, S., Hanon, O., Lemogne, C., Simon, T., . . . Danchin, N. (2018). Hyperglycaemia is associated with cancer-related but not non-cancer-related deaths: Evidence from the IPC cohort. Diabetologia, 61(5), 1089-1097. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1007/s00125-017-4540-8

Sormunen, J., Talibov, M., Martinsen, J. I., Kjaerheim, K., Sparen, P., Tryggvadottir, L., . . . Pukkala, E. (2016). Perceived physical strain at work and incidence of colorectal cancer: A nested case-control study. Cancer Epidemiology, 43, 100-104. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2016.07.001

Sritharan, J., Demers, P. A., Harris, S. A., Cole, D. C., Peters, C. E., & Villeneuve, P. J. (2017). Occupation and risk of prostate cancer in canadian men: A case-control study across eight canadian provinces. Cancer Epidemiology, 48, 96-103. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2017.04.006

Sugisaki, K., Ueda, S., Ueji, M., Monobe, H., Yako-Suketomo, H., Eto, T., . . . Mori, R. (2018). A cross-sectional investigation of cancer-screening intentions, sources of information, and understanding of cancer in japanese adolescents. Journal of Cancer Education, 33(1), 102-108. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1007/s13187-016-1056-8

Sung, J. H., Chang, S. S., Ock, M., Oh, I., Jeong, K. S., & Yoo, C. (2018). Comparison of a 10-year cumulative age-standardized incidence rate of lung cancer among metropolitan cities in korea (during the 2000–2009 period): Review of occupational and environmental hazards associated with lung cancer. International Journal of Environmental Research and Public Health, 15(6), 1259. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.3390/ijerph15061259

Turati, F., Bosetti, C., Polesel, J., Serraino, D., Montella, M., Libra, M., . . . Negri, E. (2017). Family history of cancer and the risk of bladder cancer: A case-control study from italy. Cancer Epidemiology, 48, 29-35. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2017.03.003

Walasa, W. M., Carey, R. N., Si, S., Fritschi, L., Heyworth, J. S., Fernandez, R. C., & Boyle, T. (2018). Association between shiftwork and the risk of colorectal cancer in females: A population-based case-control study. Occupational and Environmental Medicine, 75(5), 344. doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1136/oemed-2017-104657

Wright, C. M., & Moorin, R. E. (2020). Does minimum follow-up time post-diagnosis matter? an assessment of changing loss of life expectancy for people with cancer in western australia from 1982 to 2016. Cancer Epidemiology, 66 doi:http://dx.doi.org.ezproxy.laureate.net.au/10.1016/j.canep.2020.101705