This observational, longitudinal study was undertaken to better understand how changes in health-related behaviors during a societal disruption such as that caused by the COVID-19 pandemic might alter cardiometabolic risk in Latinos of Mexican Ancestry who live in the border region of the Southwest United States, in Tucson, Arizona. Since members of this community are at elevated risk of type 2 diabetes, we recruited participants in El Banco por Salud, a Latino diabetes biobank focused on type 2 diabetes and cardiometabolic risk [12]. Genetically, such individuals on average have a 45% contribution of genes derived from Indigenous American ancestry on average, with the remainder primarily being European ancestry (unpublished results). Participants were either healthy or had prediabetes or type 2 diabetes mellitus, defined according to criteria of the American Diabetes Association based on HbA1c [19]. The present study focused on the impact of voluntary or forced changes in health-related habits during the pandemic, rather than COVID-19 infection itself. This longitudinal study provides the first evidence on the changes of nutritional habits and physical activity in Latino patients of Mexican Ancestry who are at high risk for type 2 diabetes mellitus during the COVID-19 pandemic.

Regarding anthropometric and clinical characteristics of the participants at the pre-pandemic, baseline timepoint, participants with healthy HbA1c levels were younger, had lower HbA1c, BMI, and plasma triglycerides but had higher total and LDL cholesterol levels. This could be due to the fact that subjects with abnormal glucose tolerance also usually receive cholesterol lowering medications in order to reach therapeutical targets for patients with diabetes. Patients with type 2 diabetes had the lowest HDL cholesterol levels (controlled for sex). A major finding of this study is that HbA1c did not change in any group during the period of the pandemic covered by this study, indicating that diabetes care likely was not affected to the point that HbA1c was altered. Since a time period of 3 months elapsed during the period of study that fell within the pandemic, major changes in glucose control in patients with type 2 diabetes likely would have been detected. This finding suggests that well-managed Community Health Centers maintained their usual level of diabetes care despite the problems in healthcare delivery during this initial part of the pandemic when uncertainties resulted in lockdowns and a fall in patient visits. To mitigate this, El Rio Community Health Center engaged in systematic telephonic outreach to patients with elevated risk for complications (including patients with diabetes) to ensure they were taking medications, had refills and to offer telemedicine visits, making over 40,000 outreach telephone calls to our patients in the first year of the pandemic. The COVID-19 pandemic has been associated with an increased risk of developing or worsening cardiometabolic risk factors [20] that have been ascribed to several factors. First, there was a marked decline in hospital-based clinic visits for non-COVID-related conditions and this was due to patient fear of contracting COVID-19, public health messages to avoid the outpatient clinic for non-COVID-19 related conditions, and limited access to emergency medical services (because of reduced staffing from illness or isolation requirements) [21,22,23]. This also resulted in a reduction of routine laboratory testing that is an important part of cardiovascular risk reduction, since medical treatment for diabetes, hypertension and dyslipidemia involves the routine measurement of laboratory parameters, both for safety and for appropriate modification of therapies. In addition, the loss of employment, for many, may have resulted in the loss of insurance benefits and thus difficulties affording glucose-lowering, antihypertensive and lipid-lowering medications [24]. However, because many patients in El Banco por Salud have insurance coverage through Medicaid or other needs-based programs, this may not have been a larger problem during the pandemic than what is usually experienced by these patients. Nevertheless, total cholesterol levels fell in all participants during the pandemic. This may have been partially due to dietary changes during the pandemic (see below). Perhaps counterbalancing this potential cardioprotective change, HDL levels also declined. This could be related to physical activity changes that are described below.

The present study revealed a number of sex differences in diet that were present before the pandemic and were sustained during the pandemic period. Men reported being significantly more likely than women to consume flour tortillas, refried beans, hamburgers or cheeseburgers, fried potatoes, fried chicken, tacos/burritos/enchiladas, other dishes with meat, pizza, fruit juice, any potatoes, roast pork/beef/steak. On the other hand, women reported consuming more green salads and vegetables than men. Sex hormones have been reported to have an impact on eating behavior, in the so-called “homeostatic” control of energy intake as well as “the hedonic” control of food intake [25, 26]. Our finding is in agreement with the current evidence in literature. A recent study by Stea et al. investigating the sex-based differences in food choices with a cross-sectional study in 21 European countries (n = 37,672 individuals) found that women (n = 19,815) were more prone to have a higher consumption of fruit and vegetables than men (n = 17,857) [27]. In line with these results, the study by Wardle et al. in 19,298 individuals (8482 men and 10,816 women) highlighted that fruit and fiber intake was higher in women than in men that preferred to consume more high-fat foods and salt [28]. The results of the current study extend those findings to individuals from an unrelated population, Southwest U.S. Latinos of Mexican Ancestry. The present findings indicate that there were few changes in reported diet during the first year of the pandemic. However, LDL cholesterol levels fell in all participants during the pandemic. This could be explained [29] by the fact that egg consumption fell significantly during 2020. Regarding food security, there were no significant changes in food security during the pandemic, which is consistent with the lack of major changes in reported diet described above. There was a marginally significant increase in patients with type 2 diabetes in their reported ability to afford to eat balanced meals.

Regarding physical activity, there were sex and diabetes status-related differences in self-reported voluntary and job-related physical activity before the pandemic. Men reported engaging in more job-related walking and heavy manual labor, consistent with individuals with lower economic resources and education level attainment, characteristic of El Banco por Salud [12]. Most striking, however, was the overall low level of voluntary physical activity, with the vast majority of participants reporting little to no voluntary activity lasting at least 15 min every week. In particular, participants with prediabetes and type 2 diabetes reported significantly lower levels of moderate and vigorous voluntary exercise, even below the already low levels of activity in individuals with “healthy” HbA1c levels. In this regard, it is worth noting that participants with lower HbA1c were also overweight, while patients with prediabetes and diabetes on average suffered from obesity. Low levels of physical activity are consistent with obesity. Interestingly, when participants were asked about their feelings regarding physical activity, during the pandemic people overall were less likely to report that they exercised because they were told to do so, but patients with diabetes had the opposite effect, where independent of the pandemic or sex they were more likely to engage in exercise because they were told to do so. This may have been a result of clinical recommendations to these patients. Moreover, independent of the pandemic or sex, participants with diabetes or prediabetes were less likely to report they enjoyed exercise or felt restless if they refrained from exercise. A portion of the decline in HDL during the pandemic may have been attributable to lower physical activity levels, but this is not entirely clear from these data.

This study had several limitations. First, the sample size, which was only 98 for the first assessment, fell due to difficulties in recontacting patients during the pandemic. If those patients who did not participate at the second time point were somehow different from those who did participate, this may have affected the results. Statistical techniques used account for dropout but assume this is random. Second, since COVID-19 infection was not assessed and participants did not always know if they had been affected, it is not possible to dissect the effect of pandemic conditions of hardships from those of COVID-19 infection. In addition, because diet, physical activity and food security depended on self-reported measures, these variables have inherent limitations. In fact, although our study primarily focuses on voluntary and work-related physical activity due to data availability, we acknowledge that this decision may have constrained our comprehensive assessment of physical activity patterns during the pandemic. Finally, it is important to note that while the study focused on the impact of changes in health-related habits during the pandemic, rather than COVID-19 infection itself, the potential role of factors such as vitamin D intake in influencing immune response and COVID-19 outcomes should also be considered [30].

In conclusion, the findings of this study describe information regarding COVID-19 pandemic-related changes in eating habits and physical activity and their consequences on cardiometabolic risk factors in Latino patients of Mexican Ancestry who are at high risk for type 2 diabetes mellitus. Although analysis of the data revealed a community with deficits in physical activity and food security, there were surprisingly few pandemic-induced changes in these parameters or cardiometabolic risk factors that may be connected to health behaviors. Major differences in the diets of men and women suggest that behavioral interventions in diet and physical activity should be not only culturally appropriate but designed to have a sex-specific component.