Use of an online personal health record’s Track Health function to promote positive lifestyle behaviors in Veterans with prediabetes

Abstract

This pilot 3-month clinical trial investigated the feasibility, effectiveness, and acceptability of using the Track Health function of the Veterans Health Administration’s personal health record for eliciting a more positive physical activity and dietary intake lifestyle in a sample of 38 overweight and obese Veterans with prediabetes. Comparisons between baseline and 3 months post-intervention indicated significant improvements in weight, physical activity, abdominal circumference, and blood pressure. Use of a personal health record that users can identify with and find usable and useful coupled with instruction targeting critical functionalities could potentially promote healthy behavioral lifestyle changes.

Keywords

intervention lifestyle personal health record prediabetes Veterans

Introduction

Prediabetes has reached epidemic levels in the United States (Centers for Disease Control and Prevention, 2014), prompting a public service campaign launched by the American Medical Association (AMA) and other health advocacy groups to encourage people to be checked for prediabetes. In explaining the reason for the campaign, AMA President-Elect Dr Andrew W Gurman noted that “If you are at risk for prediabetes and lose 5%–10% of your body weight, change your diet, and exercise more regularly, you can markedly reduce your chance of developing diabetes” (MedPage Today, 2016). Unfortunately, 25–35 percent of Americans are essentially sedentary (American Psychological Association, 2009), which greatly increases the risk of diabetes and obesity (Tremblay et al., 2010), as can poor dietary habits (Hu et al., 2001). Conversely, positive lifestyle changes can be beneficial for diabetes prevention (Tuomilehto et al., 2001).

Veterans in the United States represent a particularly vulnerable population that could greatly benefit from such positive lifestyle changes as many of them, and especially older Veterans, generally have poor overall health status and maintain behavioral lifestyles that are conducive to these adverse health states (Selim et al., 2004; Vandenberg et al., 2010). The MOVE (Managing Overweight/Obesity for Veterans Everywhere) Program was created by the Veterans Health Administration (VHA) to address behavioral lifestyles related to physical activity (PA) and diet that could be placing Veterans at tremendous risk for diseases such as diabetes and hypertension (Kinsinger et al., 2009). However, the need for ongoing reinforcement and maintenance care makes such a professional-led program impractical and costly in the long-term or even infeasible for many patients, especially those living in rural areas.

The current pilot clinical trial explored the possible benefits of a type of computer-based intervention, the personal health record (PHR), as a potentially cost-effective means for promoting healthy PA and dietary lifestyles among Veterans with prediabetes. PHRs are electronic health record tools intended for empowering patients by increasing their access to and ownership over their healthcare information (Archer et al., 2011). The VHA’s PHR is known as My HealtheVet (MHV). It is a comprehensive and easily accessible online tool that includes Track Health (TH), a particularly powerful function. TH comprises two features that are especially relevant to the promotion of positive lifestyle changes: Journals, wherein users can record their daily activity and food intake through activity and food journals, and Vitals + Readings, where the user can access, edit, and monitor various physiological measures such as body weight, blood pressure, and glucose and display these data in tabular or graphical formats over selected time periods. The capability of these specialized tools for empowering Veterans to adopt more positive lifestyles, however, has been largely unknown as a recent national survey indicated that most Veterans do not exploit a number of this PHR’s functionalities, including TH (Nazi, 2010).

Psychosocial and cognitive considerations also argue for using this PHR as a conduit for promoting healthy lifestyle behaviors among Veterans. Because this PHR is designed for Veterans, it is more likely to elicit acceptance by this group as, consistent with social identity theory, people are likely to be more trustworthy of products with a shared group membership (Hogg and Terry, 2014). Also, its promotion of dynamic user–system interaction affords the opportunity for exploiting natural links between different sets of information (Domagk et al., 2010), for example, between the lifestyle data contained in TH’s Journals and the weight data contained in TH’s Vitals + Readings, which can facilitate understanding of cause–effect relationships between lifestyle and health outcomes.

This pilot clinical trial used a single-group design to test the feasibility, effectiveness, and acceptability of using the TH function of MHV as a tool for eliciting a more positive PA and dietary intake lifestyle in overweight and obese Veterans with prediabetes. Baseline and 3-month post-intervention outcomes were measured in two main domains: objective (PA as measured by an accelerometer, diet, weight, abdominal circumference, body mass index (BMI), and blood pressure) and subjective (self-efficacy, intent to adhere to lifestyle change, and patient activation). We hypothesized that this PHR-based intervention would demonstrate improvements and increases, respectively, in these objective and subjective domain measures, and that engagement with MHV would be demonstrated through meeting target criteria regarding making entries into this PHR’s TH function.

Methods

All participants were Veterans (n = 38, 29 males, and 9 females) who were recruited from the primary care clinics of a Veterans Administration Medical Center (VAMC) of a large city in the United States. Veterans were eligible to enroll in the study if they were diagnosed with prediabetes (HbA1c = 5.7%–6.4%, fasting plasma glucose = 100–125 mg/dL, or oral glucose tolerance test = 140–199 mg/dL); were older than 20 years of age; English speaking; had access to the Internet; had a BMI of 25–42 kg/m; were not physically active ⩾ 3 days per week for 20 minutes each time during the previous 6 months; were able to engage in a regular program of mild-to-moderate intensity PA but were not currently following a weight loss diet or participating in an intervention trial; were not cognitively impaired (Mini-Cog score < 3; Borson et al., 2000) or depressed (Patient Health Questionnaire (PHQ)-2 > 3; Rizzo et al., 2000); did not have vision or hearing disorders; were not diagnosed with diabetes mellitus, uncontrolled hypertension, or cardiovascular disease; and were not currently pregnant or nursing. The mean age of the participants was 57.7 (standard deviation (SD) = 7.7) years, and the sample was racially and ethnically diverse (24% Hispanic, 55% Black). The study was approved by the VAMC’s Institutional Review Board, including meeting all guidelines regarding obtaining informed consent by participants.

Assessment

Assessments took place during each of four visits to the VAMC. The first visit included screening on inclusion criteria and training on use of MHV’s TH function. Of the 56 participants who were recruited for the study, 9 were excluded based on not meeting the study criteria. After providing their informed consent, participants received training and practice on using MHV to promote positive PA and diet lifestyles as described in the “Intervention” section.

Participants were then asked to complete five questionnaires: a socio-demographics questionnaire; a PA self-efficacy questionnaire (the 18-item Exercise Self-Efficacy Scale; Bandura, 2006); a diet self-efficacy questionnaire (a single 7-point item; 1, true; 7, untrue; “I am confident that I can adhere to a healthy diet, daily, for the next three months”; Fishbein and Ajzen, 2011); and two single 7-point item (1, likely; 7, unlikely) questionnaires pertaining to intent to perform PA and intent to adhere to diet, respectively: “I intend to perform PA for at least 20 minutes, three times per week for the next three months” and “I intend to adhere to a healthy diet, every day, for the next three months” (Fishbein and Ajzen, 2011).

Next, participants were given an Actigraph GT3X accelerometer and asked to place the device on their hips in the morning upon awakening and to remove it before going to bed for seven consecutive days. These devices, which calculate total daily energy expenditure in kcals, provide more accurate and comprehensive measures of PA than other methods such as self-reports and pedometers (Butryn et al., 2016). To improve adherence, four evidence-based strategies were used (Sharpe et al., 2011), such as placement of reminder stickers near one’s bed. Participants were also instructed to make the following entries into the TH function of MHV: food intake information at least four times per week, PA information at least three times per week, and weight recordings at least once a week. Participants were informed that they would be sent secure messages to their MHV accounts three times a week to remind them to use the accelerometer and to make the requested entries into the PHR. They were informed that if they failed to meet accelerator use and information entry criteria for the first week, they would not qualify for the study. Participants received US$175 if they completed the entire study.

Baseline assessment occurred one week later during the participants’ second visit (week 1 of the study). The accelerometer data were downloaded by the experimenter and participants were asked to download their log data, which contained their use of MHV’s TH function, from their MHV accounts (which only they had access to). Five people did not meet the accelerometer use, and four people did not meet the information entry criteria. Participants then had their weight, abdominal circumference, pulse, and systolic and diastolic blood pressure measured. The Nutrition Data System for Research (NDSR; 2016), a dietary analysis computer program designed for the collection and analysis of 24-hour dietary recalls (from which a kcal dietary intake measure was computed), was then administered, followed by the administration of questionnaires for measuring health literacy, objective numeracy, graph literacy, patient activation, and perceived usability and usefulness of MHV. Health literacy was measured with the Newest Vital Sign (NVS; Weiss et al., 2005). Numeracy ability was measured using a scale consisting of 11 items developed by Schwartz et al. (1997) and by Lipkus et al. (2001). Graph literacy was measured using a graph literacy scale consisting of 13 items (Galesic and Garcia-Retamero, 2011). These skills were measured as PHRs can place demands on users to comprehend health information presented in text, numeric, or graphical formats. Patient activation was assessed using the Patient Activation Measure (PAM), a 22-item scale that captures various domains of patient activation such as taking action to maintain and improve one’s health (Hibbard et al., 2004). Participants’ perceived usefulness and usability of MHV was assessed with a questionnaire containing seven usefulness items and five usability items (Sharit et al., 2014). The baseline visit concluded with the experimenter supplying and reviewing several documents that the VHA has produced to educate Veterans regarding PA and dietary lifestyle modifications, and assisting participants in selecting among different types of physical activities to undertake.

Participants’ third visit occurred at week 12, during which the PA and diet self-efficacy and intention to adhere questionnaires, as well as the NDSR, were administered a second time. In addition, measures of weight, abdominal circumference, pulse, and systolic and diastolic blood pressure were also collected. Participants were again given an accelerometer to wear over the following week. The fourth and final visit occurred at week 13. During this visit, the experimenter downloaded the accelerometer data (of the 38 people still actively participating in the study, all adhered to the accelerometer use criteria) and participants downloaded their log data containing their use of MHV’s TH function over their 13-week period of participation in the study. The PAM and the MHV perceived usability and usefulness questionnaires were then administered to participants a second time. In addition, the System Usability Scale (SUS), a 10-tem questionnaire that provides a general assessment of the usability of a product (Brooke, 1996), and a questionnaire which gauged satisfaction with various aspects of the study were administered. The study concluded with an audio-recorded semi-structured interview which inquired about various perceptions participants had regarding the study.

Intervention

The primary study intervention involved the provision to study participants of targeted instruction and practice on use of TH’s Journals and its Vitals + Readings features as a basis for promoting positive PA and dietary lifestyles. Toward this end, seven interactive multimedia instructional modules were developed by the study team specifically for this pilot clinical trial using Adobe’s Captivate application (these training modules are available upon request). As noted, during their first visit to the study site, participants received training and practice on these modules. The instructional program was displayed on a Hewlett-Packard TouchSmart 9300 computer with a 23-inch diagonal widescreen liquid-crystal display at 1920 × 1080 resolution. These modules demonstrated the following: how to enter PA and diet information into the Journals feature of MHV’s TH function; how to add weight data into the Vitals + Readings feature of the TH function and view these and other measures in tabular and graphical formats over different time frames; the concept of possible cause–effect links between behavioral lifestyle and physiological outcome (e.g. weight) data; and how to use MHV’s Secure Messaging (SM) function to communicate with study team members if there were any concerns or questions. Participants received practice and feedback during training on each module and performed a set of basic tasks following training. They were required to meet certain training criteria, which all participants met, to ensure that they had the ability to perform basic tasks using this PHR’s TH function.

An important component of this intervention was the directive to participants to enter food intake information at least four times per week; PA information at least three times per week; and weight recordings at least once a week over the ensuing 13-week study period in order to maintain active engagement with the PHR and to communicate with study investigators through this PHR’s SM function. An identical weight scale was provided to all participants.

Data analysis

The primary outcome measures were changes in weight (and BMI), abdominal circumference, energy expenditure, dietary intake, and blood pressure. Secondary measures included changes in PA and diet self-efficacy, intention to adhere to PA and diet, and patient activation. Changes in perceptions of usefulness and usability of MHV were also examined. Pre- and post-intervention comparisons of these measures were based on paired t-tests. Pearson’s moment correlations were computed to determine whether health literacy, numeracy ability, and graph literacy were predictive of changes in the physiological, self-efficacy, and intention to adhere measures. For those difference measures that violated the assumption of normality, including the expected non-normal discrete questionnaire data, Wilcoxon signed rank tests were used. Violations of normality were found for the weight (and BMI) difference measures, which were attributed to one participant who actually gained 26 pounds over the 3-month study period; we opted to include this participant in the analysis.

Data were also collected on the percentages of participants who met target study criteria regarding differences between baseline and post-intervention measures of PA, dietary intake, weight loss, and abdominal circumference. The target PA criterion was defined as an average increase (>100 kcal/day) of energy expenditure over the 7-day period that the accelerometer was worn; the target dietary intake criterion was defined as a decrease (<100 kcal). These values were based on the 6-month Van Genugten et al.’s (2012) clinical trial that investigated an online computer-tailored weight management intervention for overweight and obese adults. The study criterion for weight loss was based on the findings from several Internet-based weight-loss studies: Collinson et al. (2011), in which the average weight loss over 6 months was found to be 7.71 lbs; Rothert et al. (2006), who found an average weight loss over 6 months of 6.17 lbs; and Womble et al. (2004), who reported an average weight loss of 1.54 lbs over 3 months. Based on these findings, we adopted a target criterion of > 6 lbs weight loss over 3 months. Collinson et al. (2011) also found an average decrease in abdominal circumference over 6 months of 1.89 inches, whereas Seo and Niu (2015), in a meta-analysis of Internet-based interventions spanning from 4 weeks to 2 years, found an average waist circumference loss of 0.94 inches. We chose a relatively stringent target criterion of a reduction in abdominal circumference of > 2 inches along with a less stringent criterion of > 1-inch reduction. We also adopted the criterion of a decrease in BMI greater than one unit.

Finally, percentages were computed of participants who met the criteria regarding the minimum number of entries that they were asked to make each week into the TH function with respect to PA, diet, and weight. The data related to meeting target PA, weight, and abdominal circumference criteria, as well as meeting weekly target criteria related to making PA, diet, and weight entries into the TH function of MHV, are reported using descriptive statistics. For the latter entry data, only the last 10 weeks—weeks 4 to 13—are reported due to difficulties that were encountered in recovering entry data over the first 3 weeks.

Results

Primary and secondary outcome measures

Table 1 presents the means and SDs of the primary (objective domain) and secondary (subjective domain) outcome measures at baseline and post-intervention, along with the corresponding results of paired comparison tests. Statistically significant differences associated with the intervention were found in the direction of lower weight and reduced abdominal circumference, increased PA, and lower systolic and diastolic blood pressure. In addition, a significant decrease in diet self-efficacy was found, implying less confidence in the ability to adhere to a healthy diet; the movement of the other subjective measures following the intervention were generally in the same direction. Correlations computed between the measures of health literacy, objective numeracy, and graph literacy and the outcome measures in Table 1 indicated significant correlations between objective numeracy and abdominal circumference r(38) = 0.384, p = 0.017; graph literacy and abdominal circumference, r(35) = 0.346, p = 0.042; and graph literacy and weight loss, r(35) = 0.357, p = 0.035.

Table 1.

Primary and secondary (subjective domain) outcome measures.

Measure	Baseline	Post-intervention	p-value
	mean (SD)	mean (SD)
Weight (lbs)	225.0 (36.3)	222.1 (36.0)	0.003^a
Body mass index	33.6 (3.9)	33.4 (3.9)	0.087^a
Abdominal circumference (inches)	44.2 (5.0)	43.5 (4.9)	0.007^b
Physical activity (kcal/day)	435.2 (208.5)	483.8 (221.0)	0.026^b
Dietary intake (kcal/day)	1612.2 (598.3)	1525.0 (777.3)	0.613^b
Systolic blood pressure (mmHG)	131.5 (16.6)	124.0 (13.7)	0.003^b
Diastolic blood pressure (mmHG)	89.2 (10.9)	81.0 (12.6)	0.001^b
Exercise self-efficacy (0–100)	57.4 (16.8)	56.4 (17.5)	0.981^a
Diet self-efficacy (1–7)	1.9 (1.2)	2.4 (1.4)	0.019^a
Intent to perform PA (1–7)	2.1 (1.3)	2.6 (1.5)	0.053^a
Intent to adhere to diet (1–7)	2.0 (1.2)	2.2 (1.3)	0.207^a
Patient activation (0–100)	67.5 (12.3)	67.8 (14.5)	0.243^a

SD: standard deviation; PA: physical activity.

Based on Wilcoxon signed rank tests.

Based on paired t-tests.

Meeting target PA, dietary intake, and weight-related criteria

The number and percentage of participants who met the 3-month PA, dietary intake, weight, abdominal circumference, and BMI criteria established for the study are presented in Table 2. In general, between about 24 and 45 percent of the participants met these criteria.

Table 2.

Achievement of target study criteria (n = 38).

Criterion	Met target	Did not meet target
Physical activity (>100 kcal increase/day over 7 days)	17 (44.7%)	21 (55.3%)
Weight (>5% weight loss)	2 (5.3%)	36 (96.7%)
Weight (>than 6 lbs weight loss)	10 (26.3%)	28 (73.7%)
Abdominal circumference (reduction > 1 inch)	20 (52.6%)	18 (47.4%)
Abdominal circumference (reduction > 2 inches)	9 (23.7%)	29 (76.3%)
Body mass index (reduction > 1)	14 (36.8%)	24 (63.2%)
Dietary intake (reduction in kcal/day > 100 kcal)	17 (44.7%)	21 (55.3%)

Meeting target engagement with TH function criteria

Participants were asked to enter into the TH function information regarding their food intake at least four times per week, information regarding their PA at least three times per week, and their weight at least once a week. For each participant, the percentage of weeks (of the last 10 weeks) that each of these criteria was met was computed. The mean percentages of participants achieving these specified engagement criteria across the last 10 weeks were, respectively, 53 percent for PA, 49 percent for dietary intake, and 58 percent for weight.

Perceived usefulness and usability of MHV and study satisfaction

The mean ratings (on a scale from 0–4) of the five usefulness items and the seven usability items from the MHV Perceived Usefulness and Usability questionnaire at baseline were 3.72 and 3.38, and at 13 weeks, they were 3.78 and 3.32, respectively. No statistically significant differences were found between baseline and post-intervention ratings; however, these scores suggest very strong perceptions of usefulness of this PHR and moderately strong perceptions of its usability. Ratings based on the SUS indicated a mean score of 76.97 (SD = 15.02); the national norm mean (across all products) is 68; this mean SUS value places its usability above products such as MS Word and the Wii, and just below the iPhone (Kortum and Bangor, 2013).

Generally, participants’ reactions regarding their satisfaction with the study were positive based on their responses to a 23-item questionnaire. For example, 89.2 percent either agreed or strongly agreed that the training on the TH function of MHV was easy to understand; 70.2 percent either agreed or strongly agreed that they were more aware of diabetes as a result of participating in this study; and 97.3 percent either agreed or strongly agreed that they were satisfied with participating in this study. Participants’ responses to audio-taped semi-structured interview questions are not reported here.

Discussion

The primary objective of this pilot clinical trial was to determine whether tailored instruction on the use of powerful but underused interactive health-management features of a PHR designed by the VHA could promote, among a sample of overweight and obese Veterans with prediabetes, engagement with this PHR and healthy lifestyle behaviors. From a statistical perspective, the data overall indicated significant improvement associated with the intervention on a number of measures, including weight, abdominal circumference, and energy expenditure (PA). Although these improvements were accompanied by significantly decreased diet self-efficacy and a trend in that same direction with regard to intent to perform PA, the post-intervention values of these measures were still relatively heavily centered on the “likely” portion of the response scale (Table 1). This suggests that participants were still determined to engage in positive lifestyle behaviors but perhaps were acknowledging the challenge or difficulty in maintaining such goals.

The findings with respect to meeting the target PA, dietary intake, weight loss, and abdominal circumference criteria established for this 3-month study period (Table 2), however, may be considered to be more compelling than the statistically significant findings. First, complying with these criteria, which in some cases were relatively stringent for a 3-month intervention, was expected to be challenging for this sample of Veterans. Yet, about 24–45 percent of the sample met these criteria which, when extrapolated over a large population, represents an important finding with respect to the number of individuals who can potentially be positively impacted. It should be noted that it is unreasonable, even under the most optimistic prospects, to expect very large proportions of a population, let alone older, mostly minority Veterans, to comply with changes to lifestyle that are relatively difficult to undertake for many typical adults. Considering that a number of target criteria were based on very rigorous Internet-based programs that were designed exclusively for promoting weight loss and PA, these findings are very encouraging.

Overall, these findings are consistent with the data which indicated that many of the study participants adhered to the instructions for engaging with the TH function of the PHR, had positive perceptions of MHV’s usefulness and usability, and were generally satisfied with the study. We believe that the customized multimedia instructional program, which not only introduced participants to the TH function of MHV but also provided insights into how to best use it and understand possible cause–effect relationships that underlie this functionality’s features, as well as possible underlying psychosocial considerations related to increased potential for buy-in of this PHR, played critical roles among the participants who demonstrated engagement with MHV and positive outcomes in lifestyle behaviors.

However, we also need to treat these findings with caution in light of the possible influence of skill factors such as objective numeracy and graph literacy as implied by the findings of significant positive correlations between these variables and weight loss and reduction in abdominal circumference. Thus, it may be imperative if we are to pursue interventions using PHRs to promote healthy lifestyle behaviors that we address the fact that effective use of their powerful health-management functionalities, such as the TH function of MHV, may require modifications in order to accommodate users who may have weaker numeracy and graph literacy skills. In this regard, future studies will need to evaluate the benefits of shifting the face-to-face multimedia instructional program to PHR-contained modules, which could provide greater reach to people for whom transportation to health clinics is a burden. It may also be necessary to incorporate aids within such instructional modules that can promote understanding of numeric and graphical information that PHRs depend on.

An obvious limitation of this study was its relatively short length, limited size, and inclusion of a convenience sample of Veterans. However, given the focus on Veterans with prediabetes and on investigating the feasibility of this intervention, the study was informative. We also discovered shortcomings associated with the NDSR program used to evaluate dietary intake, which we feel may have contributed to the lack of significant findings related to kcal intake. Despite its rigor, it was tedious to administer and was based solely on the participant’s dietary intake from the previous day, which in many instances may not have been sufficiently representative of the participant’s general dietary regimen. Future studies may benefit from sacrificing specificity for a more global representation that extends over the past week of the person’s dietary regimen.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by a grant through the US Department of Veterans Affairs, Health Service Research & Development Service (PPO 13-135).

References

American Psychological Association (2009) Physical inactivity poses greatest health risk to Americans, research shows. Available at: www.apa.org/news/press/releases/2009/08/physical-inactivity.aspx (accessed 24 June 2016).

Archer

Fevrier-Thomas

Lokker

et al . (2011) Personal health records: A scoping review. Journal of the American Medical Informatics Association 18(4): 515–522.

Bandura

(2006) Guide for constructing self-efficacy scales. In: Pajares

Urdan

(eds) Self-Efficacy Beliefs of Adolescents, vol. 5. Greenwich, CT: Information Age Publishing, pp. 307–337.

Borson

Scanlan

Brush

et al . (2000) The mini-cog: A cognitive “vital signs” measure for dementia screening in multi-lingual elderly. International Journal of Geriatric Psychiatry 15(11): 1021–1027.

Brooke

(1996) SUS: A “quick” and “dirty” usability scale. In: Jordan

Thomas

Weerdmeester

et al . (eds) Usability Evaluation in Industry. London: Taylor & Francis, pp. 189–194.

Butryn

Arigo

Raggio

et al . (2016) Enhancing physical activity promotion in midlife women with technology-based self-monitoring and social connectivity: A pilot study. Journal of Health Psychology 21(8): 1548–1555.

Centers for Disease Control and Prevention (2014) National diabetes fact sheet: National estimates and general information on diabetes and prediabetes in the United States 2014. Available at: www.cdc.gov/diabetes/data/statistics/2014statisticsreport.html (accessed 21 July 2016).

Collinson

Lindley

Campbell

et al . (2011) An evaluation of an Internet-based approach to weight loss with low glycaemic load principles. Journal of Human Nutrition and Dietetics 4(2): 192–195.

Domagk

Schwartz

Plass

(2010) Interactivity in multimedia learning: An integrated model. Computers in Human Behavior 26(5): 1024–1033.

10.

Fishbein

Ajzen

(2011) Predicting and Changing Behavior: The Reasoned Action Approach. London: Taylor & Francis.

11.

Galesic

Garcia-Retamero

(2011) Graph literacy: A cross-cultural comparison. Medical Decision Making 31(3): 444–457.

12.

Hibbard

Stockard

Mahoney

et al . (2004) Development of the patient activation measure (PAM): Conceptualizing and measuring activation in patients and consumers. Health Services Research 39(4): 1005–1026.

13.

Hogg

Terry

(2014) Social Identity Processes in Organizational Contexts. Hove: Psychology Press.

14.

Van Dam

Liu

(2001) Diet and risk of type II diabetes: The role of types of fat and carbohydrate. Diabetologia 44(7): 805–817.

15.

Kinsinger

Jones

Kahwati

et al . (2009) Design and dissemination of the MOVE! Weight-management program for veterans. Preventing Chronic Disease 6(3): A98.

16.

Kortum

Bangor

(2013) Usability ratings for everyday products measured with the system usability scale. International Journal of Human-Computer Interaction 29(2): 67–76.

17.

Lipkus

Samsa

Rimer

(2001) General performance on a numeracy scale among highly educated samples. Medical Decision Making 21(1): 37–44.

18.

MedPage Today (2016) Groups launch campaign to raise diabetes awareness. Available at: www.medpagetoday.com/endocrinology/diabetes/55789 (accessed 4 September 2016).

19.

Nazi

(2010) Veterans’ voices: Use of the American Customer Satisfaction Index (ACSI) Survey to identify My HealtheVet personal health record users’ characteristics, needs, and preferences. Journal of the American Medical Informatics Association 17(2): 203–211.

20.

Nutrition Data System for Research (2016) Nutrition Coordinating Center, Epidemiology & Community Health, University of Minnesota, Minneapolis, MN. Available at: www.ncc.umn.edu (accessed 14 June 2016)

21.

Rizzo

Piccinelli

Mazzi

et al . (2000) The Personal Health Questionnaire: A new screening instrument for detection of ICD-10 depressive disorders in primary care. Psychological Medicine 30(4): 831–840.

22.

Rothert

Strecher

Doyle

et al . (2006) Web-based weight management programs in an integrated health care setting: A randomized, controlled trial. Obesity 14(2): 266–272.

23.

Schwartz

Woloshin

Black

et al . (1997) The role of numeracy in understanding the benefit of screening mammography. Annals of Internal Medicine 127(11): 966–972.

24.

Selim

Berlowitz

Fincke

et al . (2004) The health status of elderly veteran enrollees in the Veterans Health Administration. Journal of the American Geriatrics Society 52(8): 1271–1276.

25.

Seo

Niu

(2015) Evaluation of Internet-based interventions on waist circumference reduction: A meta-analysis. Journal of Medical Internet Research 17(7): e181.

26.

Sharit

Lisigurski

Andrade

et al . (2014) The roles of health literacy, numeracy, and graph literacy on the usability of the VA’s personal health record by veterans. Journal of Usability Studies 9(4): 173–193.

27.

Sharpe

Wilcox

Rooney

et al . (2011) Adherence to accelerometer protocols among women from economically disadvantaged neighborhoods. Journal of Physical Activity and Health 8(5): 699–706.

28.

Tremblay

Colley

Saunders

et al . (2010) Physiological and health implications of a sedentary lifestyle. Applied Physiology Nutrition and Metabolism 35(6): 725–740.

29.

Tuomilehto

Lindström

Eriksson

et al . (2001) Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. New England Journal of Medicine 344: 1343–1350.

30.

Van Genugten

Van Empelen

Boon

et al . (2012) Results from an online computer-tailored weight management intervention for overweight adults: Randomized controlled trial. Journal of Medical Internet Research 14(2): e44.

31.

Vandenberg

Bergofsky

Burris

(2010) The VA’s systems of care and the veterans under care. Generations 34(2): 13–19.

32.

Weiss

Mays

Martz

et al . (2005) Quick assessment of literacy in primary care: The newest vital sign. Annals of Family Medicine 3(6): 514–522.

33.

Womble

Wadden

McGuckin

et al . (2004) A randomized controlled trial of a commercial internet weight loss program. Obesity Research 12(6): 1011–1018.