Abstract
The “Newest Vital Sign” (NVS) is a validated health literacy assessment tool typically administered by clinicians. The objective of this study was to assess if the NVS could be self-administered in adolescents to measure health literacy. Sixth graders in a Colorado middle school were provided a self-administered survey containing the NVS, a section for parent permission, and a section for the student’s age, gender, grade, and previous elementary school. In all, 167 sixth graders returned usable surveys (45% return rate), and the average health literacy score was 3.75 ± 1.70. Almost two thirds (62.9%) of the students scored in the adequate health literacy range, while only 12.6% scored in the limited health literacy range. Health literacy scores were similar when evaluated based on gender. However, when students were grouped based on prior elementary school attendance, students who matriculated from one elementary school had an average NVS score significantly lower than two other elementary schools (p < .001 and p < .05). Self-administration of the NVS was successful and showed similar health literacy scores compared to other studies in adolescents. Using the NVS as a self-administered tool could greatly increase its function as a quick health literacy assessment for adolescents, both in clinical practice and in school-based health education.
Introduction
Health literacy has been defined by the Institute of Medicine (2004) as the “degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions” (p. 4). Making health decisions requires many skills, such as the ability to read and interpret information, ability to perform mathematical calculations, and social and communication skills. Health literacy could be considered a potential surrogate marker for health outcomes in adults, as multiple clinical studies have linked low or inadequate health literacy with poor knowledge of disease state management and/or worse health outcomes (Baker et al., 2002; Fang, Machtinger, Wang, & Schillinger, 2006; Kalichman, Ramachandran, & Catz, 1999; Powell, Hill, & Clancy, 2007; Schillinger et al., 2002; Sentell, Baker, Onaka, & Braun, 2011; Williams, Baker, Parker, & Nurss, 1998; Wolf, Gazmararian, & Baker, 2005). Health literacy in children is linked with general literacy and is a potential marker for health risk behaviors in teens. The importance of children and adolescents understanding basic health information is recognized by First Lady Michelle Obama’s Let’s Move (2016) campaign to raise a healthier generation of kids.
Unfortunately, there are very few health literacy assessments designed for younger age-groups and consequently very few data indicating the effects of their own health literacy on health outcomes in children and adolescents (Chari, Warsh, Ketterer, Hossain, & Sharif, 2014). Two health literacy assessments have been validated in adolescents over age 10, but they are very different: the Rapid Estimate of Adolescent Literacy in Medicine (REALM-Teen) and the Newest Vital Sign (NVS; Davis et al., 2006; Warsh, Chari, Badaczewski, Hossain, & Sharif, 2014; Weiss et al., 2005). The REALM-Teen takes about 3 minutes to complete and requires the adolescent to pronounce a series of words in ascending order by difficulty. Reading recognition is scored by the administrator, and the overall score relates to estimates of grade-level reading ability. Comprehension and interpretation of the words are not assessed. While the REALM-Teen tool is quick to administer, it requires one assessor for each person being tested and may cause stress for teens with low literacy because it requires them to read out loud. In addition, the REALM-Teen health literacy assessment cannot be self-administered and is only available for English speakers.
The second health literacy assessment tool for adolescents, the NVS, also takes about 3 minutes to complete but assesses both reading comprehension and numeracy with required skills approximately matching the educational level of fifth- to sixth-grade students. The NVS comprises a nutrition label from an ice cream container and a series of six questions that require the person to use information from the label to answer the questions. Each response is scored, and the overall score on the NVS correlates with the category of health literacy of the individual. A score of 4 to 6 almost always indicates adequate health literacy, a score of 2 to 3 indicates the possibility of limited health literacy, and a score of 0 to 1 suggests a high likelihood (50% or more) of limited health literacy. The NVS is appropriate for many settings in the United States due its ease of administration, availability in Spanish, and previous use in diverse groups of patients (e.g., Caucasians, African Americans, Hispanics, and several other ethnicities; Shealy & Threatt, 2016). It has been studied in pediatric clinics and in the community (Shealy & Threatt, 2016). Like the REALM-Teen, the NVS is typically administered by trained clinicians or study investigators. However, the NVS is more conducive to self-administration, as it is in a written form that can be scored separately from administration. Self-administration of the NVS could improve access and increase the utility of the tool as a means to assess nutrition-related knowledge and skills. In addition, a self-administered version of the NVS could potentially be incorporated into the public school system as a component of health education. We are unaware of any data using the NVS as a self-administered assessment of health literacy. Thus, we aimed to use a self-administered form of NVS to assess the health literacy of adolescents and to identify factors that might affect the results.
Method and Strategies
The NVS tool was used to assess the health literacy of sixth graders attending a public middle school in Aurora, Colorado, in the fall of 2015. Health literacy assessment surveys were provided to students by their science teachers as a voluntary part of a sixth-grade science project. The project was approved by the Liberty Middle School Institutional Review Board.
Surveys were composed of a brief introduction to the purpose of the assessment, a section for parental consent, a section for student assent, and a section to collect gender, age, grade, and previous elementary school data. On the opposite side of the survey were the NVS ice cream nutritional label and the six assessment questions (Figure 1). Instructions provided for the students included that the student was to use the ice cream label to answer the questions, that there was no time limit, that it was ok to use scratch paper but not a calculator, and that the answers should be provided in the spaces provided. Finally, the instructions stated in two separate places that the student should complete the questions on their own without help from anyone else.
Newest Vital Sign Self-Administered Health Literacy Assessment
Approximately 400 surveys were provided to the three sixth-grade science teachers for distribution to their 373 sixth-grade students. Instructions on the surveys provided a date to return them to the science teacher, which was approximately 1 week after distribution. Returned surveys were checked for both parent consent and student assent. Next, the consent/assent was cut from the portion with the NVS responses, making the surveys confidential. Finally, each survey was evaluated using grading criteria provided by the NVS tool kit (Pfizer, 2011). One point was awarded for each correct response, with 6 possible points in total.
The outcome measures of the study were total score on the NVS assessment, proportion of students with adequate health literacy, and proportion of students with limited health literacy. Outcome measures were further evaluated based on gender and elementary school. Descriptive statistics were used to evaluate health literacy scores of the overall group. Chi-square and students t tests (www.socscistatistics.com) were used to evaluate health literacy based on gender; a one-way analysis of variance test followed by Tukey’s multiple comparisons test (GraphPad Prism 6) was used to evaluate health literacy scores based on prior elementary school. A p < .05 defined significance.
Results
One hundred seventy surveys were returned. Three surveys did not include either consent or assent so they were excluded. In all, 167 surveys were included in the main analysis (Table 1). The majority of respondents were females (52.7%), and the average age was 11.1 years. Data from students were subdivided by the elementary school they attended prior to entering middle school. In total, 45% of students returned usable surveys, with 39%, 50%, 35%, and 94% returning surveys from elementary schools “A,” “B,” “C,” and Other, respectively.
Demographics and Results of Assessing Health Literacy Using a Self-Administered Newest Vital Sign in Sixth Graders
Difference not statistically significant based on students t test, p = .74. bDifference not statistically significant based on chi-square test, p = .81
Overall, the average health literacy score was 3.75 ± 1.70 (out of 6 points), with 62.9% of sixth graders assessed as having adequate health literacy (Scores 4-6) and 12.6% assessed as limited health literacy (Scores 0-1). When evaluated based on gender, the average health literacy score was similar in male (3.74 ± 1.67) and female students (3.80 ± 1.72), p = .74 (Table 1). The results were similar for the comparison of male and female students with adequate health literacy and limited health literacy, p = .81 (Table 1). However, in the analysis of NVS scores based on prior elementary school attendance (Figure 2), the average health literacy score of “School C” (2.94 ± 2.01) was significantly lower than that of “School A” (4.11 ± 1.55, p < .001) and “School B” (4.00 ± 1.43, p < .05). Almost three fourths of students from “School A” and two thirds of students from “School B” had adequate health literacy as assessed by the NVS, as compared to less than one half of students from “School C” (Table 1). Likewise, the number of students from “School C” who scored as limited health literacy was greater than those with limited health literacy from Schools “A,” “B,” and “Other” combined (Table 1).
Average Health Literacy Score of Sixth Graders Based on Prior Elementary School
Discussion and Conclusion
Our study found that adolescents in a public middle school provided with a written version of the NVS were able to successfully complete the health literacy screening as a self-administered assessment. Based on the results of our study, the majority of sixth graders was categorized as having adequate health literacy and the average overall health literacy score was 3.75 ± 1.70. This result is similar to that found by Warsh et al. (2014) where researchers administered the NVS to adolescents aged 10 to 17 years and the median score was 3.0.
Thirteen percent of the students surveyed in our study were categorized as consistent with having limited health literacy, and 25% had the possibility of limited health literacy (Scores 2-3; data not shown). This indicates that a significant number of students may be candidates for further education to improve their health literacy. Public school systems in the United States are an important resource for educating adolescents and could play a key role in helping adolescents with low health literacy improve their knowledge and skills prior to potentially suffering from poor health outcomes, such as obesity, or displaying health risk behaviors. Chari et al. (2014) surveyed 239 parent–child dyads using the NVS and included children and adolescents aged 7 to 19 years of age, with a median age of 11 years. Obesity in the adolescent group (defined as age 12-19 years) was highly correlated with adolescent health literacy, with the odds of obesity significantly higher for adolescents with limited health literacy according to the NVS (adjusted odds ratio = 5.26, 95% confidence interval: [1.26, 22.01], p = .02).
In adolescents with limited health literacy, it is very possible that the inability to read and comprehend nutrition labels contributes to obesity or makes it more challenging for them to combat existing obesity. Likewise, limited health literacy may contribute to adolescents engaging in activities adverse to their health. The students in the current study had not yet taken the school-required health course, offered only in the seventh and eighth grades. We are hopeful that this health course will provide more information to the low-performing students regarding reading, interpreting, and applying information from nutrition labels. School-based health education courses likely include many health topics that can positively affect and improve adolescent health literacy and the development of healthy lifelong behaviors.
The Centers for Disease Control and Prevention (CDC) provides resources for schools to implement health education in the classroom (CDC, 2016b). In addition, they discuss the Health Education Curriculum Analysis Tool, a comprehensive tool that can help schools and/or school districts conduct an analysis of health education curricula based on the National Health Education Standards and CDC’s Characteristics of an Effective Health Education Curriculum (CDC, 2016a). The Health Education Curriculum Analysis Tool can be customized to local needs and results can help schools develop or revise health education curricula. The Let’s Move (2016) campaign also includes recommendations for incorporating nutrition education into the classroom. They have innovative recommendations, such as teaching younger children their colors using fruits and vegetables. For older students, the campaign recommends instructing them about nutrition through calorie counts and daily nutritional value charts (Let’s Move, 2016).
A potential mechanism for improving the efficiency of school-based health education related to nutrition could be to incorporate self-administered NVS assessments into the classroom with immediate formative feedback provided through classroom work with additional nutrition labels. Incorporating nutrition labels from various ethnic foods may also help improve applicability for students of different ethnicities, retention of information outside of the classroom, and possible dissemination to family members. Schools with a high prevalence of low NVS scores could increase their time educating students about nutrition or health topics, or they could revisit the mathematical skills necessary to correctly answer the NVS questions.
Very little other data exist in children and adolescents regarding health literacy assessment and future health consequences of low health literacy. Although it is well known that parent health literacy can affect the health and welfare of their children, further research is needed regarding the health literacy of the children and adolescents themselves. With two health literacy assessments validated in adolescents, research can focus on assessing their health literacy, correlating it to health outcomes, and/or researching ways to increase the health literacy of our youth. Self-administering the NVS as a health literacy tool for adolescents may make this research more feasible.
One limitation of our study design is that, against the instructions, students self-administering the NVS could have received help in answering the NVS questions. However, there was no incentive for the students to do this as the survey was confidential and had no impact on their classroom grades. We do not believe this potential risk significantly affected our study results. First, we found similar results to that of Warsh et al. (2014). Second, we found no difference between health literacy scores for male and female students, indicating there was no bias based on gender. Third, we found a significant difference in the results based on prior elementary school attended, indicating that even in a confined geographic area, health education levels vary significantly. Although we did not collect socioeconomic or health literacy data from families of assessed students, “School C” has a different socioeconomic distribution and slightly different ethnic makeup compared to “School A” and “School B.” For example, in 2014 (the year the students would have started their last year of elementary school), 40.6% of students from “School C” were eligible for free or reduced-price meals while less than 20% of students from schools “A” and “B” were eligible; 21% of students from “School C” identified themselves as Hispanic or Latino while only 11% to 12% of students from schools “A” and “B” identified as Hispanic or Latino (Cherry Creek School District Office of Assessment and Evaluation, 2016).
Another limitation of our study design is that we relied on adolescents to independently take the surveys home, obtain parental permission, and return the survey to their science teachers within 1 week. As such, only 45% of students returned surveys, potentially biasing the results toward students who felt confident with their skills of reading and interpreting a nutritional label. Since we do not have access to baseline literacy data for the students, it is also possible that those who did not return surveys had lower literacy rates than the students who returned surveys.
Finally, the data gathered in this study may have limited applicability to students outside the Cherry Creek School District of Colorado. Students in other demographic areas of the United States and educated in different school districts may respond differently on a self-administered NVS assessment.
In conclusion, a self-administered NVS assessment of adolescents was successful and well-received and indicated that the majority of sixth graders in the study had adequate health literacy. However, the data indicate differences between levels of health literacy amongst sixth graders based on elementary school training. Middle school science or health courses could focus on health literacy, if desired, to improve health knowledge in adolescents.
Footnotes
Acknowledgements
The authors would like to thank Dr. Lorry Getz for his help with the project design and implementation.