The focus of this paper is to evaluate the methodological quality of the research study by Miller, et al., (2014) with the aim of determining if this study addressed, in its design the probability and degree of bias within it. To successfully conduct the appraisal, several questions were used as provided by the critical appraisal tool for randomized controlled trial, thereby creating a standardized platform to assess the article. Such questions include whether the participant groups had similar prognostic details, identification of confounding factors, validity of results, randomization treatment, stratification, completion of follow up, and applicability of results in clinical practice (Godin, Dhillon, & Bhandari, 2011).
Twenty Five Year Follow-Up for Breast Cancer Incidence and Mortality of the Canadian National Breast Screening Study: Randomised Screening Trial by Miller, et al., (2014)
The aim of the study was to compare the incidence and mortality of breast cancer in two groups of women, those exposed to mammography and who were not, with participants aged between 40 and 59 years with a 25year follow-up post mammogram. The research took a randomized controlled trial design where the participants were stratified by age, those between 40 and 49 and 50 to 59 years and further by five year age group. Randomization of participants between control group and treatment elevates the likelihood of obtaining participants with similar characteristics (Candice, David, &Philippe, 2009). All participants were physically examined and randomization was then done by study coordinators in each of the 15 screening centers where eligible participants were independently, randomly, and blindly assigned to receive or not receive mammography, irrespective of the physical examination results. The selected participants to undergo mammography were screened for the first time and an additional four annual screens plus physical examination and a follow up scheduled. Participants assigned not to undergo mammography were attended physical examination procedures at their centers of screening.
On mammography or physical examination, women whose results revealed abnormality were referred a review clinic and diagnostic mammography performed, if indicated. if it necessitated further investigations in diagnosis such as biopsy, they had to see a specialist. Diagnosed individuals with breast cancer underwent treatment under care recommended by their family doctors, and the researchers made a follow-up via communication with surgeon annually. Women with normal results resumed the trial.
Strengths and Weaknesses
This research study was ethical as signing of an informed consent form approved by relevant authorities was done by participants before randomization. For biasness, the study ensured equivalence of treatment groups by the random allocation of participants into the group to undergo or not undergo mammography. As Young & Solomon, (2009) suggests, random controlled trials (RCT) enable similarity of participating groups in unknown and known confounding factors through the random allocation, thus reducing sytematic bias. Additionally, Miller, et al, 2014 further minimized bias by blinding participants during allocation of groups. As McCormack, (n.d) outlines, studies carried out in a blinded manner reduces the prejudice influence. Allocation conceilment was also addressed in Miller, et al, 2014 by allowing central randomization by a third party, study coordinators, as suggested by McCormack, (n.d). However, allocation conceilment using internet-based randomization system or computer program would have been more efficient. Since there was no proposed hypothesis or prediction of the results, it was difficult to trace impact bias in the study. Intention to treat analysis was not utilized in the study, and the study failed to mention if there was any discontinuation of participation due to any other reason apart from death. The study is applicable to clinical practice where breast cancer in women aged between 40 and 59 years is in consideration. Therefore, the study design used in this study is valid with randomization, similar prognostic factors of participants ensured, bias reduced and ethical issues addressed.
Confounding factors may have influenced the results of this study. As Andale (2016) states, confounding factors can introduce bias and give false research results. For instance, the study failed to consider the breast cancer family history of participants, which have been shown to determine the likelihood of developing the disease. Also, participants’ behavioral factors such as smoking, and drinking, environmental factors such as the type of environment they are exposed to in their work places, utilization of contraceptives and existence of any other chronic conditions are also potential confounding factors that may influence the study results, by increasing the probability of developing breast cancer or accelerating death. Most of these factors were not awarded any consideration, since there is no mentioning of any exclusion criteria basing on such factors.
With regard to real world factors that may affect the quality of this research, diagnosis with breast cancer is a scary experience and different women show different attitudes towards mammography screening. There was a possibility of increased droup out from participation of the study due to fear or attitudes, thus affecting the quality of the research. Additionally, the twenty five year follow up seems to be a limited time to enable complete exploration of the effects of mammography screening and other external factors that may influence development of breast cancer. Also, the lifestyle of individuals has been proven to play significant role in determining the likelihood of developing the disease. Since participants in this study have different lifestyles, the results may not be accurate, lowering the quality of the research. In this regard, the quality if the research may have been influenced by different factors.
Although follow up was complete, it was not awarded adequate time since the awarded 25 years leaves no room for investigation for the influence of external factors. For the incomplete follow up, death was the reason. The trial’s results indicated no decrease in mortality associated with breast cancer upon mammography screening. However, significant results were obtained in difference in survival upon cancer diagnosis between physically diagnosed and mammography diagnosed cancers.
Therefore, the research study by Miller, et al. (2014) complied with most of the recommended guidelines of undertaking a randomized control trial, rendering their study valid and of high quality. Particularly, the study design was appropriate, the outcome, death and tumor, to measure the impacts of mammography on mortality and incidence rates in women with breast cancer was suitable, and the study is clinically applicable. However, a number of aspects were not appropriately addressed as per the provisions of randomized control trial critical analysis tool. The follow up length, for instance, was not adequately consistent with the evolution of the disease. In this regard, Miller, et al., (2014)’s article is, in general, of good quality.
Andel E,. (2016). Confounding Variable: Simple Definition and Example. Statistics how to. Retrieved April 28 from
Candice E., David J.T., Philippe R,.(2009) How to perform a critical analysis of a randomised controlled trial. Best Practice and Research: Clinical Rheumatology, Elsevier, 23 (2):291-303.
Godin, K., Dhillon, M., & Bhandari, M. (2011). The Three Minute Appraisal of a Randomized Trial . Indian Journal of Orthopaedics , 45(3): 194-196.
McCormack, J. (n.d). How to Critically Appraise an RCT in 10 Minutes.
Miller, A., Wall, C., Baines, C., Sun, P., To, T., &Narod, S. (2014). Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. BMJ, 348(feb11 9), g366-g366
Young, J. M., & Solomon, M. J. (2009). How to Critically Appraise an Article . Nat Clin Pract Gastroenterol Hepatol , 6(2): 82-91.