Maximizing a Student's Understanding of Biostatistics Through the Review of Medical Research Studies

Introduction

In order to maximize understanding of biostatistics and its applications, a great practice for students is to review medical research studies. When reviewing medical research studies, it is important for students to recognize the type of study (randomized clinical trial, case control study, cohort study, or longitudinal study) employed and whether the study is retrospective or prospective. Understanding the type of research used and whether it is retrospective or prospective will aid a student in determining a study's validity in an objective manner. This paper will briefly describe the preceding terms and then go on to applying these terms to five medical research study abstracts that I will describe in detail while pointing out each study's potential strengths and weaknesses.

Types of Studies

There are four types of studies that can be used in the design of a medical research study, those being a randomized clinical trial, case control study, cohort study, and longitudinal study. A randomized clinical trial study is "a clinical study where volunteer participants with comparable characteristics are randomly assigned to different test groups to compare the efficacy of therapies" (Randomized Clinical Trial, 2011, para.1). A case control study is epidemiological study used to identify factors that may contribute to a medical illness by comparing those who have the disease or condition of interest (the cases) with those who do not have the disease or condition of interest (the controls). A cohort study is "a study in which subjects who presently have a certain condition and/or receive a particular treatment are followed over time and compared with another group who are not affected by the condition under investigation" (What is Cohort Research?, 2011, para.1). A longitudinal study is a study that follows patients over a prolonged period of time. Some sources also describe a longitudinal study as one in which the same patients are examined on two or more occasions.

A study can either be retrospective or prospective. Retrospective studies look backwards in time and examines exposures to suspected risk or protection factors in relation to an outcome that is established at the start of the study. In contrast, a prospective study looks forward in time and watches for outcomes, such as the development of a disease, during the study period and relates this to other factors such as suspected risk or protection factor(s).

Research Study Abstract #1

The first medical research study abstract that will be described in detail is "A Large Study of Long-Term Daily Use of Adult-Strength Aspirin and Cancer Incidence" by Eric J. Jacobs, Michael J. Thun, Elizabeth B. Bain, Carmen Rodriguez, S. Jane Henley, and Eugenia E. Calle.

This study shows evidence that aspirin use correlates with lowered risks of colon cancer and possibly several other cancers, including prostate and breast cancer. The study method examined associations between long-term daily use of adult strength aspirin (325 mg/day) measuring overall cancer incidence and incidence of 10 types of cancer among 69810 men and 76303 women. These men and women, who were relatively elderly, took part in the Cancer Prevention Study.

Since aspirin use was reported at enrollment in 1992-1993 and updated in 1997, 1999, and 2001, this is a cohort study. Individuals were already classified as groups at the beginning of the study into those who were taking 325 mg/day of aspirin and those who were not. The study is also retrospective in nature because it looked at survey data that was already provided by the American Cancer Society.

Results of the experiment showed that during a follow up in 2003, that Long-term (5 years) daily use of adult-strength aspirin, in comparison with no aspirin use, was associated with lower overall cancer rate in men (multivariable-adjusted RR = 0.84, 95% confidence interval [CI] = 0.76 to 0.93). Since the confidence interval does not include 1 and RR (rate ratios) for men was < 1, results are deemed statistically significant. A non-statistically significantly lower overall cancer rate was reported in women (multivariable-adjusted RR = 0.86, 95% CI = 0.73 to 1.03). Although the RR (rate ratios) for women was < 1, the confidence interval includes 1; results a deemed to be non-significant because variation can account for the observed correlation. Results also show that long-term daily aspirin use correlated with lower incidence that was statistically significant of colorectal cancer (RR = 0.68, 95% CI =0.52 to 0.90 among men and women combined) and prostate cancer prostate cancer (RR =0.81, 95% CI = 0.70 to 0.94). Non-statistically significantly lower overall female breast cancer (RR = 0.83, 95% CI = 0.63 to 1.10) was shown by the results.

Since this study is a retrospective cohort study, it does have several specific strengths and weaknesses. Strengths are that it is good for displaying multiple outcomes and is relatively inexpensive. Major weaknesses include that the investigator has little or no control over exposure and outcome, the temporal relationship is often difficult to determine, it requires large samples for rare outcomes, and the comparability between exposed and unexposed is difficult to achieve.

Research Study Abstract #2

The second medical research study abstract that will be described in detail is "Effects of Immediate Versus Delayed Pushing during Second-Stage Labor on Fetal Well-Being: A Randomized clinical trial" by Kathleen Rice Simpson and Dotti C. James.

This study shows evidence that it is more favorable for the well-being (in terms of fetal oxygen desaturation/saturation and variable decelerations of the fetal heart rate) of an infant for mothers at 10-cm cervical dilation to delay pushing until the woman feels the urge to push versus being coached to immediately start pushing at 10-cm cervical dilation. The study also shows that women who pushed immediately had more perineal lacerations. No differences were shown in the results between pushing and delayed pushing in length of labor, method of birth, Apgar scores, or umbilical cord blood gases.

Since women were randomly allocated to receive one or other of the alternative treatments under study (to push or delay pushing) this is a randomized clinical trial. It is also prospective in nature because it looked forward in time and watched for outcomes.

Results of the experiment showed significant difference between groups in fetal oxygen desaturation during the 10-cm cervical dilation (immediate: M = 12.5; delayed: M = 4.6) F (1, 43) = 12.24, p =.001, and in the number of >=2-min epochs of fetal oxygen saturation <30% (immediate: M = 7.9; delayed: M = 2.7), F (1, 43) =6.23, p =.02. There were more variable decelerations of the fetal heart rate in the immediate pushing group (immediate: M = 22.4; delayed: M = 15.6) F (1, 43) = 5.92, p =.02. Women who immediately at pushed 10-cm cervical dilation had more perineal lacerations (immediate: n = 13; delayed: n = 5) [chi] 2 (1, N = 45) = 6.54, p =.01. Assuming that the alpha is.05; the p-values fall well below the threshold and are deemed to be statistically significant; in other words, the null hypothesis is rejected in variable decelerations of the fetal heart rate, fetal oxygen saturation, and perinatal lacerations.

Since this study is a prospective randomized clinical trial, it does have several specific strengths and weaknesses. Strengths include elimination of bias in treatment assignment, facilitation of blinding (masking) of the identity of treatments from investigators, participants, and assessors, and it permits the use of probability theory to express the likelihood that any difference in outcome between treatment groups merely indicates chance. Weaknesses include potential limitations of external validity on where the study was performed, on characteristics of the patients, on study procedures, on outcome measures, and in incomplete reporting of adverse effects of interventions. In addition, randomized clinical trials can be quite expensive to perform.

Research Study Abstract #3

The third medical research study abstract that will be described in detail is "An Intervention to Increase Safety Behaviors of Abused Women" by Judith McFarlane, Ann Malecha, Julia Gist, Kathy Watson, Elizabeth Batten, Iva Hall, and Sheila Smith.

This study shows evidence that telephone intervention is an effective way increase safety behaviors of abused women who are victims of intimate partner violence. Intervention is administered immediately after an abusive behavior and remains effective for 6 months after the treatment. A two-group trial randomized 75 abused women to receive six telephone intervention periods on safety behaviors. A control group of 75 women received standard care. Women in both groups were re-interviewed at 3 months and 6 months for post-initial measurement.

Since women were randomly selected to receive the alternative treatment or be part of the control group under the study (to receive intervention or not receive intervention) this is a randomized clinical trial. It is also prospective and longitudinal in nature because it followed subjects over a prolonged period of time through re-interviewing the women at both 3 months and 6 months.

Results of the experiment showed significant difference between groups who received intervention to those who did not. More adopted safety behaviors were reported by women in the intervention group [F (2,146) 5.11, p =.007] than women in the control group at both the 3-month [F (91, 74) = 19.70, p <.001] and 6-month [F (1, 74) = 15.90, p <.001] interviews. Assuming that the alpha is.05; the p-values fall well below the threshold and are deemed to be statistically significant; the null hypothesis is rejected in women who received intervention at both the 3-month and 6-month interviews.

As previously discussed, since this study is a prospective randomized clinical trial, it does have several specific strengths and weaknesses. Strengths include elimination of bias in treatment assignment, facilitation of blinding (masking) of the identity of treatments from investigators, participants, and assessors, and it permits the use of probability theory to express the likelihood that any difference in outcome between treatment groups merely indicates chance. Weaknesses include potential limitations of external validity on where the study was performed, on characteristics of the patients, on study procedures, on outcome measures, and in incomplete reporting of adverse effects of interventions. In addition, randomized clinical trials can be quite expensive to perform.

Research Study Abstract #4

The fourth medical research study abstract that will be described in detail is "Treatments of Coronary Artery Disease Improve Quality of Life in the Long Term" by Hannele Lukkarinen and Maija Hentinen.

This study shows evidence that long-term health-related quality of life (HRQoL) after 8 years is more favorable for patients receiving a bypass operation or angioplasty than the baseline of 1 year after surgery. Also, in patients taking medication, the study shows that there were no significant changes were reported in the HRQoL of the patients on medication when comparing the baseline to 8 years after treatment. It is important to note that a significant impairment on the response variable of sleep had taken place during the 8 years follow-up after angioplasty. The initial study was made up of 280 patients where 100 of those patients underwent a bypass operation, 100 had an angioplasty, and 80 were prescribed medication. After 1 year, 81 bypass operation patients, 74 angioplasty patients, and 64 patients in the medication group responded. In the final analysis after 8 years, 63 bypass operation patients, 57 angioplasty patients, and 34 patients in the medication group responded.

Since patients with coronary artery disease are compared by their method of treatment this is a case control study. It is also prospective and longitudinal in nature because it followed subjects over a prolonged period of time through interviewing patients after 1 and 8 years.

Results of the experiment showed patients had a statistically significantly better HRQoL 8 years after the operation than at baseline on the response variables of mobility (p <.001), energy (p =.003), and pain (p =.031). Angioplasty patients had a statistically significantly better HRQoL 8 years after the intervention on the response variables of emotional reactions (p =.002), pain (p =.003), mobility (p =.004), and energy (p =.005). A significant deficiency on the response variable of sleep had taken place after 8 years follow-up after angioplasty (p =.018), Assuming that the alpha is.05; the p-values fall well below the threshold and are deemed to be statistically significant; the null hypotheses after 8 years regarding HRQoL and sleep are rejected.

Since this study is a prospective case control study, it does have several specific strengths and weaknesses. Strengths include that it is good for rare diseases, requires little time to conduct, has the possibility of exploring multiple exposures and is relatively inexpensive. Weaknesses include the reliance on recall or historical data, that temporality can be difficult to establish, and the comparability of cases and controls.

Research Study Abstract #5

The fifth medical research study abstract that will be described in detail is "Caregiving Experiences After Stroke" by Cynthia S Teel, Pamela Duncan, and Sue Min Lai.

The purpose of this study was to examine correlation between patient characteristics, caregiver characteristics, and caregiver managing resources with caregiver physical and mental health results at 3 and 6 months after a dependent adult's stroke. Another objective was to compare family members' assessments of patient disability with assessment by doctors. The caregiver study was conducted in partnership with a study of patient effects after stroke. 83 Caregivers completed and mailed back questionnaires at 1, 3, and 6 months after the patient's stroke. The surveys asked questions on fatigue and energy, assessment of mood disturbance, stress, spirituality, and reactions to the caregiving situation. This data recorded by respondents provided a detailed assessment of caregiver characteristics, coping resources, and physical and mental health status.

Since Individuals were already classified as a group at the beginning of the study into caregivers and patients, this is a cohort study. It is also prospective and longitudinal in nature because it followed subjects over a prolonged period of time through interviewing caregivers at 1 month, 3 months, and 6 months.

Results of the experiment showed correlation between physical health and depressive symptoms are parallel at 3 and 6 months. Perceived stress was correlated to mental health at 3 and 6 months. Caregiver ratings of disability at 1 month matched doctor's assessments using the Orpington Prognostic scale. Results also show evidence of a caregiver's stable perceptions of fatigue, vigor, recurrent sorrow, perceived stress, finances, family support, physical health, and depression symptoms at 1, 3, and 6 months after a dependent adult's stroke. It was determined that a comprehensive approach to stroke rehabilitation should include comprehensive assessment of caregiver functioning soon after a dependent adult's stroke. Early assessment might identify persons at greater risk for physical and mental health problems in a continuing caregiving role.

Since this study is a prospective cohort study, it does have several specific strengths and weaknesses. Strengths are that there is the possibility to study multiple exposures and multiple outcomes in one cohort and rare exposures can be studied. Major weaknesses are that it is not possible to establish causal effects and it is easily susceptible to selection bias. Also, prospective cohort studies can be quite expensive to perform.

Research Study Abstract #6

The sixth medical research study abstract that will be described in detail is "Daily Stress and Gastrointestinal Symptoms in Women With Irritable Bowel Syndrome" by Vicky L. Hertig, Kevin C. Cain, Monica E Jarrett, Robert L. Burr, Margaret M. Heitkemper.

The purpose of this study was to examine the correlation of daily self-reported stress to gastrointestinal and psychological distress symptoms both across women and within woman in a comparison group of women without Irritable bowel syndrome (IBS) and among subgroups of women with IBS.

Since respondents are compared by daily self-reported stress to gastrointestinal and psychological distress symptoms both across women and within woman in a comparison group of women without IBS and among subgroups of women with IBS, this is a case control study. It is also prospective and longitudinal in nature because it followed subjects over a prolonged period of time through the testing of respondents daily for a month.

This study shows evidence that gastrointestinal symptom discomfort is associated with self-reported stress in women with IBS; stress has been associated to be a contributing factor to launching bowel and discomfort symptoms and making the problem worse in patients with IBS. Results of the experiment showed significant across-women correlations among mean daily stress, psychological distress, and GI symptoms in the total IBS group and the IBS bowel pattern subgroups. Across-women and within-woman analyses were used. Women with IBS (n = 181; age = 18-49 years) were divided into subgroups based on bowel pattern (constipation, n = 52; diarrhea, n = 67; alternating, n = 62) and were compared to a group of women without IBS (n = 48). Self-report stress measures were abdominal (abdominal pain, bloating, and intestinal gas), bowel pattern (constipation, diarrhea), and intestinal gas; and psychological (anxiety and depression) distress symptoms were obtained daily over 1 month. The across-women relationships between daily stress and gastrointestinal symptoms were less when anxiety and depression were controlled in the test. Although within-woman analyses showed little evidence of correlation between day-to-day variations in stress and day-to-day variations in GI symptoms, stress was strongly related to anxiety and depression.

As previously discussed, since this study is a prospective case control study, it does have several specific strengths and weaknesses. Strengths include that it is good for rare diseases, requires little time to conduct, has the possibility of exploring multiple exposures, and is relatively inexpensive. Weaknesses include the reliance on recall or historical data, that temporality can be difficult to establish, and the comparability of cases and controls.

Summary

In summary, reviewing medical research studies can help maximize a student's understanding of biostatistics and its applications. When reviewing these studies, it is important for students to comprehend the type of study used and the potential strengths and weaknesses associated with each study. After this understanding is achieved, a student will be able to question the validity of medical research that he or she is reading in an objective manner.

References

What is Cohort Research? (2011). Retrieved April 7, 2011, from Cha Cha: http://www.chacha.com/question/what-is-cohort-research

Randomized Clinical Trial . (2011). Retrieved April 7, 2011, from The Free Dictionary: http://medical-dictionary.thefreedictionary.com/randomized+clinical+trial