In this paper I will discuss an article from the Seattle Press written by Associated Press Medical Writer Lindsay Tanner. I will discuss the statistical procedures used in the study and provide an analysis of the conclusions both the researchers and the reporter drew from the study results to consider the statistical significance of the data presented. The article explains the research of Dr. Sayeed Ikramuddin of the department of surgery, University of Minnesota, Minneapolis and his colleagues relating to bariatric surgery its potential role in glycemic control in type two diabetics.
Irkamuddin states that at 12 months, “the primary goal was considered successful if patients achieved the composite of the triple endpoint: HbA1c of less than 7. 0%, an LDL cholesterol level of less than 100 mg/dL, and systolic blood pressure of less than 130 mmHg at the 12-month visit” (2013, p. 2242). The study was conducted at four sites, including New York, Minnesota, and two hospitals in Taiwan and lasted 12 months.
Patients with mild to moderate obesity and adult-onset diabetes mellitus were offered a free intensive intervention and recruited using mass media advertisements, practice-based databases and contact with professional groups, leading to an unblinded randomized group after screening 2648 candidates. Body mass index in participants ranged from 30. 0 to 39. 9, with a mean of 34. 6. The study participants had been diagnosed with diabetes for a mean of 9. 0 years.
The intensive intervention included lifestyle modification with daily caloric intake counts, increased physical activity, daily weights and counseling sessions, and intensive medical management including close monitoring and treatment from doctors with medications for cholesterol, blood pressure, and glycemic control. Sixty of these participants were randomly selected to receive Roux-en-Y gastric bypass surgery. Two of the participants selected declined to have the surgery, and two others were randomly selected to undergo the procedure.
These patients continued to receive the same intensive intervention during the entire 12 months. The frequency of sessions and monitoring decreased for both groups during the study period as planned, going from weekly to monthly over the course of the study. The author of the newspaper article summarizes the study’s results by stating, “The surgery group lost on average nearly 60 pounds and 75% lowered blood glucose levels to normal or near-normal levels. The non-surgery group lost on average 17 pounds and just 30% reached the blood-sugar goal.
The surgery group also needed less medication after the operation” (Tanner, 2013, p. 2). The last portion of this statement is misleading as the study evaluates and considers only medications to manage blood glucose levels, dyslipidemia, and hypertension (Ikramuddin, 2013, p. 2245). The study reports, “Patients who underwent gastric bypass surgery experienced 50% more serious and 55% more nonserious adverse events than did those in the lifestyle-medical management group” (Ikramuddin et al. , 2013, p. 2248).
It is important to consider the frequency and rate of adverse events, in addition to classification of such data. Researchers also state that the surgical group took 66% fewer medications than the nonsurgical group. The most common adverse event in the surgical group that was not considered peri- or late post-operative was nutritional deficiencies, especially iron and vitamin B levels, requiring treatment with medications and supplements. Nutritional deficiencies are often long-term and require frequent monitoring and dose adjustment.
Because of the time limitations of the study and the relatively small sample size, long-term cardiovascular and nutritional effects of bariatric surgery in the mild to moderately obese population for glycemic control remains to be determined. Both the study and the newspaper article assert that there is no conclusive evidence on long-term effects when bariatric surgery is used in conjunction with lifestyle and medical management for glycemic control on mild to moderately obese patients.
Other serious adverse events documented and considered by the authors of the study included motor vehicle accidents and pancreatic cancer, both of which occurred in the medical management group, decreasing the difference between the medical management group and the Roux-en-Y gastric bypass group. Pregnancy was considered as well. The total number of serious adverse events was 15 for the medical management group and 22 for the surgical group. The study does not discuss or measure in any way the adherence of participants to the intensive intervention in either group.
Patients’ adherence to recommended regimen including decreased caloric intake, recommended physical activity, and medication compliance is a variable that was not assessed by Ikramuddin and his colleagues during the study, limiting its scope. In general, the newspaper article correctly interprets the statistics and draws conclusions from this study appropriately. The surgical group tended to have less hypertension and elevated LDL cholesterol levels, but this data lacks statistical significance. Tanner stated in the newspaper article “those between-group differences could have been due to chance” (2013, p. ). The study states a confidence interval of 95%, which correlates with a 0. 05 significance level. Many of the values representing these measures were marginally different. Many other studies have reported effects on these values as individual variables. There are many other factors involved in cholesterol and blood pressure management that must be identified and considered before drawing conclusions on such data. This study was very detailed and presented a large amount of data. Despite this, the overall conclusions of the newspaper article were accurate in relation to the data.
The newspaper article summarized the conclusions of the researchers appropriately. The statistics representing HbA1c as part of the composite endpoint are reliant on weight loss and can be primarily attributed to this factor (Ikramuddin et al. , 2013, p. 2240). The newspaper article states, “The researchers say the diabetes changes were likely due to the weight loss but that hormonal changes affecting blood sugar may have contributed” (Tanner, 2013, p 2). This was determined by using regressive analyses of the available data.
The study suggests that altered gastrointestinal hormone secretion as well as differences in individual medication use during the study may have attributed to the results, but acknowledges that the study lacks the data to determine this. By going to the source of news-reported statistical studies, the reader can evaluate the statistical significance of the data and determine if the data were presented accurately and if the conclusions drawn are reliable. Research and comprehension of data and statistical significance is essential to best evidence-based practice in the field of nursing and beyond.