Epidemiology Report: Evaluating the Wilson et al. Study on Adolescents

This report provides an evaluation of a research study conducted by Wilson et al. (2011) examining the effect of the ACT (Active by Choice Today) intervention on moderate to vigorous physical activity (MVPA) among adolescents. The report critically analyzes the study's methodology, including the structured PICO question, selection bias, study design (randomized controlled trial), potential confounders, blinding procedures, data collection methods, and analytical approaches. It assesses the representativeness of the sample, the appropriateness of the study design, and the control of potential biases. The report highlights the strengths and weaknesses of the study's design and implementation, providing a comprehensive overview of the study's methodology and findings. The report concludes with an overall assessment of the study's rigor and validity.

Epidemiology

Table of Contents
INTRODUCTION...........................................................................................................................1
PART 1............................................................................................................................................1
Structured PICO question............................................................................................................1
A. Selection Bias..........................................................................................................................2
B. Study Design...........................................................................................................................3
C. Confounders............................................................................................................................5
D. Blinding...................................................................................................................................6
E. Data collection methods..........................................................................................................7
F. Withdrawals and drop-outs....................................................................................................10
G. Intervention integrity.............................................................................................................12
H. Analysis.................................................................................................................................13
CONCLUSION..............................................................................................................................15
REFERENCES..............................................................................................................................16

INTRODUCTION
Epidemiology refers to the process of studying health consequences events and applying
necessary control or preventive measures to minimize it. Wilson and et.al., (2011), conducted a
research study to examine the effect of ACT (Active by Choice Today) intervention on the
MVPA (Moderate to vigorous physical activities) among adolescent. The current essay aims at
evaluating its sample biasness, study design, data collection & analytical methods used.
PART 1
Structured PICO question
P: Patient/population/problem: In the given study, 6th grade student of 24 middle
school of South Carolina were selected to participate. There were total 24 schools with 2500-
3000 students which is the population. However, the sample size was 1,563 students, out of
which 141 excluded and total of 1422 students participated in the study in either intervention or
comparison group over 17 week. Wilson and et.al., (2011), studied the key problem of increasing
prevalence of health care problem obesity among people, especially, Youth. The key reason
behind such issue is that they do not take regular exercise and physical activities (PA) which is
declined by 50% between middle and elementary school. The problem is mainly encountered
among low-income and minority people. Therefore, in this study, researcher examine the result
of ACT (Active by Choice Today) trial for MVPA (Moderate to vigorous physical activities)
among adolescent belongs to low-income & minority people.
I: Intervention: ACT intervention
C: Comparison or control: Integration program combining motivation & behavioral
skills at school
O: Outcome: The study found that at mid intervention, students who fall in intervention
conditions have greater MVPA each day in comparison to control students. However, no
significant impact is determined of ACT intervention on MVPA outside the school time for 2 wk
post-intervention.
1

A. Selection Bias
Q1.
Very Likely
It is necessary to select a representative sample which provides highly realistic and
justifiable results representing the entire group (Dimmock, Kouwenberg and Wakker, 2015). For
the given study, 2500-3000 students were the total population in 24 schools, out of which 1563
students selected which belongs to 6th grade. All the participants that were selected for the
sample belong to low-income group families and public schools of South Carolina who delivers
free or reduced lunch meals to the students. Free lunch simply is a sign that they were from the
poor family background. Students were decided to select as per their parental consent, therefore,
during the selection procedure, 141 students has been excluded who refused to participate due to
many reasons whereas remainder 1422 student of all 24 school has been chosen. They were
selected considering several inclusion criteria i.e. overweight that may possibility face the
obesity issues in near future. Besides this, as investigator want to improve physical activities
(PA) among student to reduce obesity, therefore, people who were likely to engage in low level o
f physical exercise, lacked facilities at home and have less opportunity to take participate in the
regular and vigorous PA were chosen (Wilson and et.al., 2011). All people selected were from
poor background henceforth, they have lack access to recreational activities such as gym,
playgrounds and others are limited. Out of this, 729 (12 school) allocated to ACT intervention
and 693 (12 school) were selected for the general healthcare intervention. In order to eliminate
gender biasness in sample selection, researchers had included female participants also because
they belong to minority group in respect to access to physical health exercise. For instance, many
women are engaged mainly in domestic household activities; hence, they do not find time for
outdoor activities i.e. exercise, gym and others. In total, there were 770 female people equal to
54% of total sample size. As investigator selected the required students randomly from the
comprehensive set of students, therefore, the participants selected can be considered as very
likely representative sample of target population and free from sample biasness (Schlenger and
et.al., 2015).
Q2.
80%-100%
2

Out of chosen sample of 1563, 16 do not satisfied inclusion criteria, 55 refused to
participate, 9 moved, 60 alternates and 1 removed due to other reason, hence, 1422 agreed to
participate in the baseline measures. Thus, percentage of students who agreed to take part in the
study was 91% which falls in the band of 80% - 100% agreement. Out of this, in ACT
intervention midpoint and post-intervention, 609 and 673 participated whilst in general health
intervention, 573 and 635 took part in mid-point measure and post-intervention measure.
As the agreement percentage was quite high therefore, it can be ranked 1 means strong.
B. Study Design
3

Indicate design of the study
RCT
The study design was a randomized controlled trial (RCT) which aims at investigating
motivational as well as behavioral skills efficiency on maximizing MVPA among students (Stagl
and et.al., 2015). In this, Wilson and et.al., (2011), randomly allocated eligible candidates to the
control and intervention group At the same time, it was a cohort design to some extent. It is a
design in which more than one sample is followed prospectively to measures outcome separately.
In this, same group is examined before and after the intervention delivered (Lewis, 2015).
Evidencing from the study, there were two groups, 12 intervention & 12 as comparison schools.
There were two group ACT intervention and general intervention, represented in the above
flowchart; both were examined at mid-point (9 week) and 2 week post-intervention (19 week)
measures instead of pre and post.
Was the study described as randomized?
Yes, No-doubt, the study can be described as randomized, because in this, scholars used
word random assignment as they randomly classified 729 students of 12 school people to
intervention and remaining 693 students assigned to the control group.
If yes, was the method of randomization described?
Yes, researchers used randomized selection considering school sizes, proportion of free
& reduced lunch, minorities percentage, rural or urban setting before randomization. For the
baseline measures, blinded schools were randomized through randomly generated numbers at
computer.
If yes, was the method appropriate?
In my opinion, the chosen method for the purpose of randomization was appropriate,
because, scholars were unaware of the individual characteristics of each student; hence, all of
them had an equal chance or opportunity of being included in the sample (McMillan and
Schumacher, 2014). Therefore, it become clear that yes, the method of randomization for the
given case study was appropriate.
4

RATE THIS SECTION STRONG MODERATE WEAK
Rank 1
Overall, for this section, I like to rank 1 implies strong. It is because, the study was based
on RCT, in which, investigators assigned eligible students randomly to both intervention and
controlled group. Moreover, they clearly presented the method of randomization which is
independent from individual own characteristics, therefore, chosen sample is considered as bias
free.
C. Confounders
Confounders are the aspects that are related with the groups and impact the research
outcome to a great extent. It is necessary for the authors of RCT design to clearly report that both
the intervention & controlled groups were adequately balanced at baseline considering important
confounders.
Q1.
No
Demographic characteristic of the students
Participants Characteristics Intervention group
(N = 729 students)
Controlled group
(N = 693 students)
Mean age (standard deviation) 11.32 year (0.56) 11.36 years (0.61)
Percentage of female students 407 (55.83%) 363(52.38%)
African & American % 557 (76.41%) 481 (69.41%)
Waist mean (S.D.) 70.88 cm (12.40) 71.15 cm (12.36)
Body Mass Index mean (S.D.) 22.80 kg/m2 (6.16) 22.86 kg/m2 (5.88)
According to the given result, it can be seen that there were no significant different
between group basic demographic characteristics. It showcase that investigators balanced the
people so as to minimize the differences between both groups and thereby derive reliable and
justifiable results. The result showcase nearby results by presenting mean age to 11.32 & 11.36,
5

female % to 55.83% and 52.38%, African & American % to 76.41% and 69.41%, waist to
70.88cm and 71.15cm ad BMI to 22.80 and 22.86. Thus, it becomes clear that No, there is no
considerable size of differences exists in the groups prior to intervention. It mitigates the
possibility of sampling error in final outcome.
Q2. Percentage of confounders controlled
No significant level of confounders/group differences were detected henceforth, no
measures were undertaken to control it.
RATE THIS SECTION STRONG MODERATE WEAK
1
Confounders were rated to strong, because, there was no significant level of differences
in basic demographical characteristics of the students. The above presented result depicts little
bit differences can be seen, henceforth, on the basis of confounders, it is ranked to strong (1).
D. Blinding
Q1.
No
In the research study conducted by Wilson and et.al., (2011), assessors were in the
position of blinded, it means, they were totally unaware about the participants who had been
randomly assigned to control & intervention group. In the method, it was clearly mentioned that
during the random allocation of students to both the group, assessors were unaware of individual
characteristics and worked blind folded. Thus, units were assigned by assessor unknowingly,
which in turn, helped in minimizing detection biasness.
Q2.
No
Just like detection biasness, it is also necessary to minimize reporting biasness among
chosen participants by not providing them prior information about the research questions. It
6

means that students who took part in the study should be unaware (blinded) of the designed
research questions or hypothesis. Students do not have any kind of knowledge about the ACT
intervention program and they do not know about the formulated questions of the study which
helps to gather accurate responses of the candidates and reduced reporting biasness.
RATE THIS SECTION STRONG MODERATE WEAK
1
Explanation: Evidencing from the study, due to absence of detection and reporting
biasness, it is ranked to 1(strong), in which, research worked as blind during randomization
whereas participants provided authentic responses unknowingly the research questions in
advance.
E. Data collection methods
Data collection is the procedure of extracting or gathering required set of information that
are necessary for examining the research issue (Goodman, Cryder and Cheema, 2013). For the
given research study, investigator held ACT intervention program for a time duration of 17
weeks which was undertaken on three days, Mondays, Tuesdays and Thursdays lasted for 2
hours after the school. On Wednesday, students were asked to practice what they actually learned
in post-school program.
ACT intervention program had three parts, mentioned underneath:
 Snack/Homework: Time duration 30 minutes
 Moderate to vigorous physical activities (MVPA): Time duration 60 minutes
 Behavioral and motivational skill enhancement: Time duration 30 minutes
However, participants who belong to controlled group were provided only with the
general health education that is based on stress management, ways to prevent drug uses, drop-out
intervention, nutrition and other activities without any PA.
Parts of comparison program:
Snack/Homework: 30 minutes
7

Hand on activities about general health: 30 minutes
Q1.
Yes
Validity aspect of data gathering states that research findings truly represent what
researchers claim to identify and measure. In other words, to the extent to which research results
or conclusion accurately present the entire world. In the study, researchers gathered baseline
measure in October, mid-intervention measures and post-intervention measures were taken at 9
weeks (Dec) and 2-weeks later (19 weeks in April).
Methods used for data collection:
 School records: In order to identify students who take free or reduced lunch, data was
extracted from the records maintained by school administration which is a highly valid
information source.
 Systematic observation: In this process, as awell-trained, independent assessor was
assigned to examine intervention program fidelity, given dose and its reach, which
satisfied validity as well as trustworthiness. Process evaluator collected data regarding
given dose out of the planned to the participant, fidelity and reach (% of target people
who attended program every day). Here, fidelity comprises choices, expectations,
belongingness and others. The data were marked to a 4-point rating scale 1 means low
level of fidelity while 4 presents highest. Minimal acceptable completion % was 75%
while fidelity level was 3.
 Intrinsic Motivation Inventory: IMI being used as a multidimensional device of
measurement which is found suitable for evaluating subjective responses of the
participants regarding motivation & physical exercises. IMI examine factors like
enjoyment, competencies, effort, usefulness, values, tension & pressure and others
(Glesne, 2015). However, for the given study, only two subscales were used that are
interest/enjoyment and perceived choices, both were positive predictors of the behavioral
and motivational measure and shown adequate level of validity & reliability.
 Accelerometer: It is an electromechanical device that has been used in the given research
to measure MVPA at baseline, midpoint and 2-week post-intervention. In this,
8

participants wore an accelerometer during 7 days to determine MVPA. Use of Tactical
assist investigator to extract authentic set of data that provided accurate measurement of
energy and monitoring another PA.
All the above stated methods of data acquisition clearly present that all the collected data
set was found valid and trustworthy too. It is because; it specifically addresses research
requirements by providing details regarding individual characteristics, IM and MVPA as well.
Q2
Yes
Reliability aspect states that acquired data set were sufficient, complete and also free
from any error, and considered enough or convincing for the given research aims and objectives
(Goodman, Cryder and Cheema, 2013).
The method of ACT intervention integrates principles of two different theories that are
self-determination (SD) and social cognitive (SC). First taken into account behavioural changes
which is motivated by the intrinsic factors i.e. self-driven, enjoyment and others whereas later is
based on cognitive factors i.e. competency, confidence and others. With reference to the study,
SDT factors choice, autonomy, belongingness and others factors were extracted via systematic
observation via a trained assessor. However, motivational aspects of the participants had been
judged via IMI and the primary outcome (post-intervention MVPA) collected via accelerometer.
However, on the other side, fidelity data were only collected about intervention group and not for
the controlled group (general healthcare education) due to scarce availability of resources.
Besides this, school records were used to select student of free or reduced meal services. Thus, if
any mistakes or errors existed in the record, then it would have definitely affected the research,
still, it has a very little chance. Thus, it becomes clear investigators generated enough amount of
information about the sample including their behavioural, motivational and physical activities
which clearly present that undoubtedly, data collection tools were reliable.
RATE THIS SECTION STRONG MODERATE WEAK
1
On the basis of above discussion, it is made clear that extracted information for the given
research is reliable and valid also, henceforth, ranked strong.
9

F. Withdrawals and drop-outs
Q1.
Yes
The study clearly presented the number of withdrawals and drop out of selected sample
size of 1,563 students from 24 schools of South Carolina along with the reasons of exclusion.
Out of total eligible candidates to 1563, 141 were excluded at baseline point due to below
mentioned reasons:
Reasons for exclusion Number of students
Did not satisfied inclusion criteria 16
Refusal 55
Moved 9
Alternates: 60
Other reasons (Not defined) 1
Total 141
Remainder 1422 randomly allocated to ACT intervention (729) and control group (693).
In both the groups, several participants at midpoint (120) and post-intervention have been lost up
due to following reasons:
Mid-point measures ACT intervention
(12 schools and 609 students)
Control
(12 schools and 573 students)
Moved 43 28
Absent 33 58
Refused to participate 14 23
Don’t know 20 6
Other reasons (Not defined) 10 5
10