Introduction
The article “Clinical trial of tailored activity and eating newsletters with older rural women” is a 2009 publication in the Nursing Research Journal volume 58, issue 2 from page 74 to 85. The article is authored by Susan Noble Walker, Carol H. Pullen, Linda Boeckner, Patricia A. Hageman, Melody Hertzog, Maureen K. Oberdorfer and Matthew J. Rutledge. Walker et al (2009) seek to examine the possibility of enhancing the health of rural women by comparing two interventions: individualized Health Promotion Model and an untailored intervention. This is based on the fact that incidences of chronic diseased among rural women (midlife and older women) are high due to poor diet and physical inactivity.
From their community-based controlled clinical trial, it was evident that both interventions resulted into enhanced physical activity and eating behaviors. The most effective approach between the two was the use of tailored newsletters compared to the generic newsletter.
Article Critique
Purpose
By comparing the individualized and untailored health promotion plans, the authors of this article intended to find out which of the two would be more effective in improving healthy eating and physical activities in rural women who are their 50s and 60s. Changes in the two variables were to be determined after a six and twelve month-period into the study. To determine changes in the eating behavior, Walker et al (2009) examined variables such as servings of whole grains, vegetables and fruits as well as calories obtained from fats. Activity behavior was determined using time spent doing moderate exercise, calories used as well as time the subjects were involved in stretching and strengthening forms of exercise.
Activity was expected to affect the strength of the lower body and flexibility and therefore a measure of these factors was necessary. The researchers also used BMI, percentage of body fat and blood pressure measures to determine the influence of eating behaviors and activity due to the two treatment interventions. The measures were based on standard measures set in the Healthy People 2010.
Literature Review
Walker et al (2009) first of all take time to define the term tailoring since it is a significant term that is central to this study. As such, the reader is able to understand the difference between the tailored and the untailored interventions used in this study. An outstanding aspect of the literature review conducted for this study is that the authors utilized meta-analysis data which gives a comprehensive understanding on the topic by reviewing a volume of studies on the subject.
Furthermore, the meta-analyses gives the most current through the early findings on the subject as exemplified by some reviews involving information gathered on the topic for 39 years. Considering that this article was published in 2009, it is evident that Walker et al (2009) reviewed current literature since almost all the reviewed literature had been published in the 21st century except one literature which was dated 1999.
Conceptual framework
The authors did not only select the Health Promotion Model but they also explain that the model was suitable for the study subjects since such subjects seek to be more independent than to prevent health problems by working on their diet and activity level. The framework is effective in determining outcomes of interventions sine it has a basis on behavioral-cognitive measurements. The authors pinned their confidence on this model from the fact that intervention would not only affect health behaviors but also other factors that have an influence on health behaviors. This conceptual framework is therefore commendable since it has the capacity to facilitate the study in implementing interventions and collecting data from outcomes of the intervention.
Research design
Methods
This was a community-based clinical trial that was conducted over a period of 24 months, twelve months of intervention and another twelve months of follow-up. The authors used an experimental design to compare the groups where subjects were randomly assigned as either generic newsletters group or computer-tailored newspapers group. Randomization helped in increasing the external validity of this study and hence the results thereof can be generalized in the general population.
The outcomes were measured using repeated measures, which is effective considering that the outcomes were conducted among the same subjects at different times during the study. This study can be considered to be ethically acceptable as regards to research involving human subjects. This is exemplified by the fact that Walker et al (2009) had the approval of the Institutional Review Board (IRB) in addition to obtaining informed consent from participants.
Sample
The repeated measures analysis of variance (ANOVA) is an importance measure of the power of outcome variables in a study. The authors of this article were therefore on point by utilizing this measure where a power of at least.80 was obtained, indicating that it was possible to get even minimal changes in outcomes of intervention between the two treatment groups.
In selecting the sample, the subjects were sampled randomly and all the 225 women who were sampled were from rural Midwestern state and had to be aged between 50 and 69 years. This inclusion criteria was an important way of selecting only the required sample thus reducing time for sorting out qualified individuals when no exclusion criteria is used. In addition, the authors of this study ensured that internal validity by ensuring that all subjects had almost similar characteristics at the beginning of the study. Such characteristics included computer literacy level, English proficiency and participants who were not under cardiac rehabilitation among other characteristics.
Use of exclusion criteria was also important in getting an almost standard sample which would maintain internal validity by removing the possibility of compounding. Moreover, the characteristics of the comparison group subjects were highly similar to those of the tailored intervention group thus reducing the likelihood of obtaining biased results. The sample size of 225 participants was also sufficient to produce meaningful findings considering that this was an experimental study which maintains its internal and external validity even with small sample sizes (Huitt & Kaeck, 1999).
Measurement of both primary and secondary outcomes of the interventions as well as the HPM determinants was conducted at three time periods. It is therefore likely that there was an effect of maturation and therefore the performance of the subjects may have been influenced. Overall, the internal validity of the study may have been altered by the effect of maturation. Furthermore, the likelihood of some participants dropping out of the study was exacerbated by having the study run for a period of 24 months, inclusive of the follow-up where losing participants in the follow up is likely.
The researchers ensured that the data remained confidential and secure by transmitting it to persons who were part of the investigations only, the research personnel, and via the university’s intranet which was secure. This was an important ethical consideration regarding handling and managing data in research. Furthermore, the researchers used standard protocols in transporting samples (blood samples) to the respective analysis points.
To measure physical activity, Walker et al (2009) used credible measures such as the 7-Day Activity Recall and a 1-mile walk test for testing cardiorespiratory fitness. To test flexibility, sit-and-reach rest was used while repeated timed chair stands helped determine the strength of the lower body. An important aspect in measuring physical activity is that all the subjects were subjected to the same treatment hence the authors prevented bias on the results.
Healthy eating was determined using the Block Health Habits and History Questionnaire, 1998 version. The fact that data on this variable depended on self reports meant that if some participants were aware of being observed, they may have manipulated their answers thus making the results less a reflection of the real situation. Biomarkers of activity and eating such as blood pressure measures were determined using standardized clinical tools thus making the results more credible.
To enable the researchers to tailor interventions, HPM behavioral determinants were used and scientifically acknowledged measurement scales were used. Such scales helped determined individual’s exercise habits, healthy eating habits, and friend and family support in both eating and exercise. The use of scientifically acknowledged measures further strengthens the credibility of obtained results.
The researchers sent tailored and untailored newsletters to the respective groups via mail and the tailored group got an accompanying plan of action. This helped further make the groups more distinct. The fact that all the newsletters (tailored and generic) had a similar appearance in terms of layout and face appearance helped in maintaining anonymity of the treatment and generic-targeted newsletters. Tailored content was generated with the aid of the computer while untailored information on healthy eating and activity was non-specific for individuals but the information was up to standard for both groups, hence the generic treatment group was not exposed to any health hazard whatsoever.
It is important to note that the tailored group was distinctive from the untailored group and the plan of action emphasized on the subjects committing themselves to the plan of action. This included setting goals on healthy eating goals and activity behavior, with assessments being done after every three months.
All the subjects were provided with equipment to facilitate their adherence to eating and physical activity behaviors and these included videotapes with details on how to conduct physical exercise. The participants were then supposed to give a feedback on how they faired with the interventions with the tailored group being guided previously. Such differences helped in creating clear comparison basis between the two groups.
Data analysis
Exploratory data was used to explore the data and descriptive statistics helped reveal the characteristics of the sample. Skewed data affects the test to be used (parametric test for normally distributed data and non-parametric for non-normal distribution) hence the researchers removed any outliers to make the data more normally distributed. Any missing data was handled using the Missing Value Analysis available in the SPSS program. Dealing with missing data as appropriate helps the researcher get the real reflection of the study results following analysis. When analyzing changes in outcome for the same sample over a period of time, an analysis of variance of mean performance/outcome is a valuable statistical tool (Field, 2009).
In this study, Walker et al (2009) utilized the repeated measures ANOVA for the two groups over the different assessment periods and compared the results. A main effect test helps in identifying the exact independent variable that cause a change in the outcome variable. In this study, a main effects test was conducted to determine the differences in outcomes over time. The chi-square test (2X2 contingency table) helps in comparing differences in distributions for each individual. The test was therefore appropriate for comparing the subjects on healthy eating and activity for the time periods. Block Habits and History questionnaire data was analyzed using the Block Dietary Data Systems. Significance was determined at.05 value and.017 for Bonferroni-adjusted value.
Results
The results effectively addressed the research questions as clear differences/effects of the interventions were observed between the groups. There were significant difference noted in the level of education, race and consumption of lipid-lowering drugs (p =.025,.002 and.008 respectively). The fact that there was a high participant dropout (7.8%) in the tailored intervention group compared to participant in the generic group (1.0%) may have compromised the comparison since internal validity was compromised.
The study was able to reveal that physical activity and healthy eating among rural women in their 50s and 60s improves when interventions are made and improvement is more significant if the interventions are tailored (p =.002 after twelve months). The main effects test as a component of ANOVA was indispensable in facilitating this finding. Intake of saturated fats was lower in tailored group compared to the generic group and therefore there was a more significant reduction of calories in this group.
The results demonstrated a main effect on blood pressure measures among other eating and activity biomarkers over time for both treatment groups with the tailored group displaying a significant betterment in blood pressure and “bad” cholesterol measures. While there was a reduction of body fat in the tailored group, the untailored group experienced an increase in body fat. It was further demonstrated that significant differences only emerged at twelve months but not after six months for secondary outcomes. Overall, the outcomes may have been influenced by compounding factors over the period of study, thus the presence of significant differences after more time but the authors of the study do not seem to appreciate this.
Discussion
Walker et al (2009) discuss the result contextually by comparing the outcomes of this study with other studies focusing on increasing healthy eating and exercise among women and they acknowledge their study as bringing new knowledge. The authors acknowledge the uniqueness of their study as having utilized multiple communications over a long period of time, thus achieving significant outcomes.
The authors of the study acknowledge their results to low participant dropout rate coupled with making the study as convenient to the subjects as possible. While they acknowledge a significant difference in dropout for tailored versus the untailored group, the authors point out that the effect of the dropout was not of major concern, thus giving the study more validity. The authors of this study also give commendable explanations to the various outcomes, including explanations for observed differences/no differences in outcome variables between the groups. This helps in making the interpretations more consistent.
Walker et al (2009) have effectively highlighted some limitations related to their study especially limitations in study design where bias may have resulted due to self reporting. The authors also mention the limitation of generalization due to failure to randomize subjects and instead randomizing sites. Bias in sampling may have occurred due to willingness and lack of willingness of some participants to be involved in the study.
Implications
From this study, it is evident that rural women can benefit in achieving healthy eating habits and improved physical exercise if they are given individualized interventions. It is also evident that such interventions are more effective if they are communicated severally and over a substantial period of time since the women would get enough time to apply the interventions. Overall, individualized interventions can be effective in lowering chronic diseases and mortality associated with unhealthy eating and sedentary lifestyles practiced by rural women in their 50s and 60s. It therefore calls for the extension of such services to this group of people if the targets of Healthy People 2010 are to be achieved.
Researcher creditability
This research is well presented and the fact that the authors are credible individuals in the field of nursing makes the study more acceptable globally. Commendable background information, solid research questions and relevant literature review has been provided making the research more acceptable. Overall, the study has all the sections in a credible study (abstract, background information, research questions, methodology, data analysis and interpretation, discussion and conclusion) thus it is well presented. The ethical guidelines regarding dealing with human subjects were observed and there were no conflicting interests.
References
Field, A. P. (2009). Discovering statistics using SPSS. 6th edition. Thousand Oaks, CA: Sage Publications Ltd.
Huitt, W. and Kaeck, H. D. (1999). Internal and external validity: general issues. Web.
Walker, S. N., Pullen, C. H., Boeckner, L. Hageman, P. A. and Hertzog, M. et al. (2009). Clinical trial of tailored activity and eating newsletters with older rural women. Nursing Research. 58(2): 74-85.