Diabetes self management: are Cuban Americans receiving quality health care?
Abstract: Objectives: We investigated the relationship among factors predicting inadequate glucose control among 182 Cuban-American adults (Females=110, Males=72) with type 2 diabetes mellitus (CAA).

Study Design: Cross-sectional study of CAA from a randomized mailing list in two counties of South Florida.

Methods: Fasted blood parameters and anthropometric measures were collected during the study. BMI was calculated (kg/[m.sup.2]). Characteristics and diabetes care of CAA were self-reported Participants were screened by trained interviewers for heritage and diabetes status (inclusion criteria: self-reported having type 2 diabetes; age [greater than or equal to] 35 years, male and female; not pregnant or lactating; no thyroid disorders; no major psychiatric disorders). Participants signed informed consent form. Statistical analyses used SPSS and included descriptive statistic, multiple logistic and ordinal logistic regression models, where all CI 95%.

Results: Eighty-eight percent of CAA had BMI of [greater than or equal to] 25 kg/ [m.sup.2]. Only 54% reported having a diet prescribed/told to schedule meals. We found CAA told to schedule meals were 3.62 more likely to plan meals (1.81, 7.26), p<0.001) and given a prescribed diet, controlling for age, corresponded with following a meal plan OR 4.43 (2.52, 7.79, p<0.001). The overall relationship for HbA1c < 8.5 to following a meal plan was OR 9.34 (2.84, 30.7. p<0.001).

Conclusions: The advantage of having a medical professional prescribe a diet seems to be an important environmental support factor in this sample's diabetes care, since obesity rates are well above the national average. Nearly half CAA are not given dietary guidance, yet our results indicate CAA may improve glycemic control by receiving dietary instructions.
Article Type: Report
Subject: Diabetes therapy (Methods)
Diabetes therapy (Health aspects)
Self-care, Health (Methods)
Self-care, Health (Usage)
Type 2 diabetes (Diagnosis)
Type 2 diabetes (Care and treatment)
Type 2 diabetes (Control)
Type 2 diabetes (Demographic aspects)
Authors: Huffman, Fatma G.
Vaccaro, Joan A.
Nath, Subrata
Zarini, Gustavo G.
Pub Date: 12/22/2009
Publication: Name: Journal of Health and Human Services Administration Publisher: Southern Public Administration Education Foundation, Inc. Audience: Academic Format: Magazine/Journal Subject: Government; Health Copyright: COPYRIGHT 2009 Southern Public Administration Education Foundation, Inc. ISSN: 1079-3739
Issue: Date: Winter, 2009 Source Volume: 32 Source Issue: 3
Product: Product Code: 8000431 Diabetes Therapy NAICS Code: 621 Ambulatory Health Care Services
Geographic: Geographic Scope: United States Geographic Code: 1USA United States
Accession Number: 250034407
Full Text: BACKGROUND

Minorities tend to have less access to and receive a lower quality of healthcare, even when controlling for insurance status and income. Adjusting for socioeconomic status reduces the effects of race and ethnicity, but the effects are still apparent; the lack of cultural competency and communication by health care providers can have many negative health outcomes (ACP, 2000). According to Department of Health and Human Services (2002), the prevalence of persons under the age of 65 without health insurance for all Americans was 13.3% whereas for all Hispanic Americans was 28% (Cuban, 27.0% , Puerto Rican, 14.8%, Mexican 30.2%). In addition, Hispanics are 1.5 times more likely to have diabetes than non-Hispanic whites (NDSS, 2005) and for Hispanics over 20 the age-adjusted rates are Cuban, 1.24; Mexican 1.80; Puerto Rican, 1.90 (CDC, 2008a). Despite the evidence of higher diabetes and uninsured rates among Hispanics, no studies, to our knowledge, have been available to assess the quality of healthcare and its effectiveness in diabetes self-management for each Hispanic ethnic group.

Prevalence for a particular chronic disease among ethnic groups is frequently used to determine funding for medical intervention programs within the healthcare system; yet, these statistics may not be appropriate indicators for the underserved populations. Health disparities for Cuban Americans are likely to be a product of lack of coverage compounded by a lower quality of care as compared to Non-Hispanic whites. For instance, according to the Centers for Disease Control and Prevention (2008a), other Hispanic American groups have a higher age-adjusted prevalence of diabetes than Cuban Americans (12.6% for Puerto Ricans, 11.9% for Mexicans and 8.2% for Cuban), yet Cuban Americans had the highest proportion of diabetes as the underlying cause of death (44%) as compared to Puerto Ricans (39%) and Mexican Americans (37%); the analysis of Hispanics as a homogeneous group masks the variation in the risk for diabetes mortality (Smith and Barnett, 2005).

Persons of Hispanic origins in the United States are from countries (Mexico, Puerto Rico, Central and South America, Cuba) with distinctively different cultures, yet have been almost exclusively combined for research. Over 15 years ago, Furino & Munoz (1991) suggested the groups be studied separately and that lack of distinction has been a deterrent toward improving health status and policies. The last significant epidemiological study separating Hispanics by ethnic groups was performed over 20 years ago by the Department of Health and Human Services (DHHS, 1985).

Hispanics, constituting 12.5% of the United States population experienced more age-adjusted years of potential life lost, higher prevalence of several chronic diseases and 41% more for diabetes than non-Hispanic whites, in 2001 (MMWR, 2004). The age-adjusted prevalence of diabetes for Hispanics was 69 per 1000, whereas for non-Hispanic whites it was 47 per 1000 in 2005. Hispanics are 1.5 times more likely to have diabetes than non-Hispanic whites (NDSS, 2005). Only 66% of Hispanics under the age of 65 years have health insurance as opposed to white non-Hispanics, of whom, 87% have health insurance. Only 77% of Hispanics have a regular source of healthcare as opposed to 90% of non-Hispanic Whites (MMWR, 2004; DHHS, 2002). Although Hispanics are the largest ethnic minority population in the United States, they are underserved by the healthcare system (Grieco & Cassidy, 2001). Hispanics as well as Asian Americans with diabetes, high blood pressure or heart disease are less likely to receive clinical services for monitoring these chronic conditions, than other groups (ACP, 2000). Affordable health care is one of the biggest barriers that minorities face. Insurance status more than any other demographic or economic factor determines the timeliness and quality of health care (ACP, 2000).

There is a gap in the current literature distinguishing health care coverage, quality and leading health indicators among the Hispanic cultures (ACP, 2000; AACE, 2007). Census data from 15 years ago (1992) differentiates Hispanic groups with respect to health care coverage (Ramirez et al, 1995) yet the demographics of many of these cultures have evolved. The 1992 census from the National Cancer Institute reported 24% uninsured and in the year 1994 the American College of Physicians (2000) reported an uninsured rate of 27.4%. The United States Census merges all Hispanic groups and reported a 33% uninsured rate in 2005 (U.S. Census, 2006). The last reported uninsured rates for Hispanic subgroups was nine years ago (DHHS, 2002) and may not reflect the current rates.

Type 2 diabetes, the most common form (90-95% of all cases) has increased among the general population (Harris et al, 1995; NIDDK, 2008) and disproportionately among minorities (particularly African Americans and Hispanics) (NDSS, 2005). Cultural differences among Hispanic ethnicities in the United States can influence health behaviors and access to healthcare; therefore it is essential to examine the interaction of ethnicity and disease to devise appropriate treatments leading to the elimination of health disparities (Furino & Munoz; Novello et al, 1991; Burchard et al, 2003). Specifically lacking are studies concerning the interaction of culture and Type 2 diabetes among Cuban Americans (AACE, 2007).

Cuban Americans represent 5.2% of the Hispanic population where 78% of Cubans live in the southern region (U.S. Census, 2004 a.). More than 830,000 Cubans (over half) reside in Florida (U.S. Census, 2004 b.). The US Census, 2000 indicates 57% of the population from Miami-Dade County are Hispanic and 50% of those Hispanics are Cuban (U.S. Census, 2000). We have a unique opportunity to study this minority group with high prevalence of diabetes.

Conceptualization of the Problem

The associations among diabetes self-management, health biomarkers and healthcare referral and participation in diabetes education are indicative of the level of quality of healthcare for a population with type 2 diabetes. We examined some of the variables of dietary care and health indicators for a Cuban American adult population (CAA) with Type 2 diabetes in Miami-Dade and Broward Counties. Specifically, in this paper, Hemoglobin A1c (HbA1c) and fasted plasma glucose (FPG) were the primary health outcomes. Diabetes education and social support were considered major predictors in accordance with the American Diabetes Association (ADA) recommendation for structured programs, which emphasize lifestyle change and ongoing nutrition self-management education (ADA, 2007 a.). Based upon the findings for minority groups, in particular the Hispanic population, we hypothesized that participants with Type 2 diabetes who reported having received diabetes education at least once will have adequate HbA1c and significantly lower fasted glucose levels than those participants with Type 2 diabetes who reported never receiving diabetes education. Further, we hypothesized that there will be differences between the group with diabetes education and the group without diabetes education with respect to the extent of following a meal plan and FPG level. We also hypothesized that those who follow a meal plan will have lower levels of FPG than those who do not follow a meal plan. Lastly, we hypothesized that those participants with Type 2 diabetes and social support will be more apt to follow a meal plan.

RESEARCH DESIGN AND METHODS

Data was part of an exploratory cross-sectional study to generate hypotheses using Cuban Americans with and without Type 2 diabetes matched for age and gender. Control subjects (without diabetes) were selected randomly from a community-based population. This paper focuses on an aspect of health disparity: the quality of health care as reported by Cuban Americans with Type 2 diabetes and its relationship to glucose control.

Subjects:

Inclusion criteria for participants with type 2 diabetes were self-reported Cuban or Cuban-American ethnicity; age [greater than or equal to] 35 years, male and female; self-report of a medical diagnosis of Type 2 diabetes; able to understand and complete all of the study protocol in English or Spanish; and willing and able to read and sign an informed consent form. Exclusion Criteria: pregnant or lactating, thyroid disorders, self-reported, major psychiatric disorders. The present study N=182 (72 males, 110 female)

The participants were initially recruited by random selection (every tenth address) from a randomly generated mailing list; those who responded and qualified were matched for age and gender to those with diabetes. Lists of addresses and phone numbers were purchased from Knowledge Base Marketing, Inc., Richardson, TX 75081. This company provided mailing lists of Cuban Americans from Miami-Dade and Broward Counties, Florida. Approximately ten thousand letters were mailed; Three percent (N=300) were returned for unknown addresses and 4% (N=388) responded by telephone. Subjects who responded to the mailings were screened for eligibility. The screening involved a standard questionnaire administered by telephone in either Spanish or English. This initial assessment determined eligibility based on self-reported Cuban heritage and diabetes status. Once eligibility was determined, one interview appointment was set. A series of questions based on language indicators were asked by a trained interviewer to establish Cuban heritage. Of the 388 candidates, 18 did not qualify and were excluded for the following reasons: two were not of Cuban heritage; seven had other chronic illnesses; and nine could not be matched by age.

We used "intent to treat" in the data analysis for the present study, where N=182, self-reported persons with type 2 diabetes represented data from the complete sample set of a case control, single point study of aged-matched, Cuban Americans with and without T2D where N=370. After blood analysis, 7 additional participants, who reported a non-diabetic status, were re-classified as diabetic, given their reports and advised to see their physician. Additionally, we considered persons informed by our study of their new diabetic classification as not qualified to answer diabetes self management questions.

We investigated the relationship among factors predicting inadequate glucose control among 182 Cuban-Americans (Females=110, Males=72) (CAA) with Type 2 diabetes mellitus (T2DM). This paper examined some of the variables deemed necessary for dietary T2DM self-management. Therefore, we conducted the study with the following objectives: to determine demographics of participants with T2DM; to determine the relationships among diabetes education, receiving a diet, following a meal plan or diet, level of social support, and glucose control for participants with T2DM; determine the association of diabetes self-management and quality factors of healthcare.

Procedures

Appointments were made for groups of subjects biweekly until a quota, based on a power analysis, was reached and all data were collected. The interviewer read, explained and disclosed the protocol. The acknowledgement, understanding of potential hazard and acceptance was documented by signature of informed consent. After informed consent forms were signed, height and weight were measured, fasting blood was drawn and a light breakfast was served. All independent variables were measured at one time-point during the initial interview. The participants received a modest compensation after venous, fasted blood was drawn.

Each participant then completed a series of questionnaires containing demographic information. All written materials were provided in English or Spanish except for the food frequency questionnaire (FFQ). Trained bilingual interviewers were available to aid in the translation of the FFQ. Each participant was assigned a unique personal identification number (PIN) to ensure his or her confidentiality. The log of PIN numbers as well as participants' files were kept in locked cabinets in the primary investigator (PI)'s office. Participants' data were identified by PIN numbers only.

Demographics

Data were collected using the socio-demographic questionnaire: Cuban American Type 2 Diabetes Study (Nath & Huffman, 2003) which included questions related to age, gender, education, income, medications for cholesterol, hypertension and diabetes and smoking habit. Health insurance was coded as a binary variable: all responses for any insurance was coded as having insurance and those that responded "I have not had an insurance plan in the past 12 months" was coded as not having health insurance.

We selected variables from the Cuban American Type 2 Diabetes Study questionnaire of Florida International University, Miami, Fl. 33199 (Nath & Huffman, 2003) that focused on intervention and dietary management of diabetes. The questionnaire was administered during the single point data collection and interpretation was aided by structured interview with trained personnel. Diabetes-related health behaviors (diabetes education received, insurance status, social support, dietary and self-care adherence) were predictor variables (primarily), ordinal scale and self reported. Following a meal plan was measured as both a predictor (of blood glucose control) and outcome variable (for receiving a diet and diabetes education)

Dietary adherence was measured using both of the following questions (in separate models, due to colinearity) from the Diet Adherence Scale: "How often do you follow a meal plan or diet?" and "How often do you follow the schedule for your meals and snacks?" These diet adherence variables were considered outcome measures of level of receiving diabetes education. Both responses are categorical following the Likert Scale (1-5; never-always). The social-support question chosen was, "My family or friends help and support me a lot to follow my meal plan." This ordinal support variable was categorized in the Likert Scale (1-5; strongly disagree -strongly agree).

Blood Collection and Analysis

Venous blood was collected from each subject after an overnight fast (at least 8 hours) by a certified phlebotomist in the PI's lab using standard laboratory techniques. The following tubes were used: SST tubes for blood lipids; sodium fluoride for fasted plasma glucose; and K2EDTA to analyze glycated hemoglobin. For fasted plasma glucose determination, blood was centrifuged 2500 rpm for 10 minutes (Eppendorf 5403 Centrifuge) and the plasma separated within 30 minutes of collection in the PI's lab. Plasma glucose was assayed by an automated glucose oxidase method by Laboratory Corporation of America (LabCorp [R])within 30 minutes of separation. Lipids analysis began after coagulation took place (approximately 30-45 minutes). The blood was centrifuged (2500 rpm for 10 minutes) in the PI's lab and the enzymatic assay was performed by LabCorp[R]. Glycated hemoglobin was drawn fully and tubes were inverted 8x to prevent clotting in the PI's lab and then submitted to LabCorp[R] for analysis by ion-exchange HPLC. Fasted plasma glucose (FPG) was measured by hexokinase enzymatic methods. For logistic regression, blood glucose was coded into binary variables: >126 mg/dL inadequate glucose control and <126 mg/dL adequate glucose control. Glucose level was measured as a continuous outcome variable for univariate analysis (AACE, 2007;ADA, 2007 b.).

Anthropometric Measures

Height was measured to the nearest 0.1 cm without shoes using a SECA standard stadiometer. Body weight was measured using a SECA balance beam scale (Seca Corp, Columbia, MD) to the nearest lb and converted to kg. Weight included light indoor clothing without shoes. Body mass index (BMI) was calculated as body weight (kg)/height ([m.sup.2]). Overweight and obesity were defined as BMI 25-29.9 kg/[m.sup.2] and [greater than or equal to] 30 kg/[m.sup.2], respectively, as defined by the National Institutes of Health (NIH, 2007). To determine central obesity, waist circumference (WC) was measured to the nearest 0.1 cm with a non-stretchable tape measure placed midway between the 12th rib and iliac crest at minimal respiration to determine central obesity (Calloway et al, 1988). As defined by NIH, WC> 102 cm for men and > 88 cm for women was considered at risk for cardiovascular disease (NIH, 2007).

Statistical Analysis

Statistical analyses included descriptive statistic, multiple logistic and ordinal logistic regression models and multivariate linear regression. Continuous variables were presented as a mean [+ or -] SD, and age, FPG were normally distributed. Ordinal outcome variables from the Likert 5-scales were analyzed by ordinal logistic regression. Binary variable outcomes: insurance, diabetes education, inadequate HbA1c were analyzed by multiple logistic regression models. A multivariate analysis was performed on predictors of FPG. Significance was indicated by a p-value of [less than or equal to] 0.05, hence confidence intervals (CI) of 95%. All analysis was conducted using the statistical software SPSS version 14 (Chicago, Illinois).

RESULTS

Selected characteristics of the study population are shown in Table 1. The participants (N=182) were primarily older adults with nearly two-thirds female. Triacylglycerides, a reflection of diet, were above the normal range.

Approximately 88% of participants with type 2 diabetes were found to be in the combined overweight and obese categories, where half were classified as obese BMI [greater than or equal to] 30 kg/[m.sup.2]. Approximately 15% of participants had no health insurance. Age differences in the study sample were significant with regards to health insurance level. Participants over 65 were more likely to have public insurance than those under 65, [OR 22 (CI: 9-52), p=0.05] (data not shown). Sixty two percent of CAA with type 2 diabetes were on either Medicare or Medicaid (Table 1). Approximately half of CAA with type 2 diabetes reported never receiving diabetes education regardless of health insurance status (data not shown).

"Reporting scheduling meals" and "following a meal plan" were found to predict fasted blood glucose levels in univariate analyses. The responses were collinear (Spearman's Rho = 0.546, p = 0.01, two tailed) necessitating the use of separate models. The best model for predicting fasted blood glucose by scheduling meals were F = 2.96; p = 0.014; [R.sup.2] = 7.7% (controlling for age). The results for following a meal plan which included an interaction with age was the best model, where (F=2.34, p =0.013, [R.sup.2] = 12%). Gender did not improve any of these models (data not shown). Pearson's correlation is negative for age and FPG (-1.94, p=0.01). Linear regression of CAA confirms the inverse relationship, where F = 7.0; adjusted [R.sup.2] (0.032) indicated only 3.2% of age explains FPG levels. Together the findings suggest following a meal plan, which improves with age, may counteract the biological increase in blood glucose levels for participants with type 2 diabetes (data not shown).

We found, by ordinal logistic regression of 5-scale Likert, that participants with inadequate glucose control (HbA1c> 8.5) were more than four-fold likely to never follow a meal plan , OR 5.39 (1.40, 20.7). The overall relationship of adequate HbA1c < 8.5 to following a meal plan was OR 9.34 (2.84, 30.7; p<0.001). Only 54% reported having a diet prescribed/told to schedule meals. Those told to schedule meals were 3.62 more likely to schedule (CI: 1.81, 7.26) p<0.001 (data not shown). Controlling for age, WC, BMI and gender did not improve the model.

The relationship of receiving diabetes education to scheduling meals and following a meal plan was not significant even when controlling for age. However, CAA reporting to "ever have a healthcare provider prescribe a diet" was significantly more likely to follow a meal plan [OR 6.09 (CI 3.23-11.47), p<0.001] (data not shown). Age was a significant factor in following a meal plan, whereas gender was not. Although controlling for gender was not significant, we provided this model for its clinical relevance. We found a significant three-fold, positive relationship to a similar question asked in the medical history section, "Are you currently (in the last 6 months) prescribed for any special diet by your health care provider (nurse, dietitian, diabetes educator, physician)?" (Table 2).

We suspected there may be a different relationship for level of receiving a prescription for a diet and HbA1c adequacy. We performed a logistic regression on a stratified sample of prescribed a diet within the last 6 months (yes/no) with following a meal plan on the outcome of HbA1c (inadequate > 8.5). Those who were prescribed a diet (in the past 6 months) and had HbA1c> 8.5 formed a significant bimodal relationship with following a meal plan. They reported either never followed OR 8.25 (1.15, 59.0) p=0.046 or almost always following OR 17.25 (2.53, 117), p=0.004 at the 95% CI. The model fit (classification table) predicts 78.8% of the cases correctly. The overall relationship of following a meal plan on adequacy of HbA1c was significant for the level that reported receiving a prescribed diet in the past 6 months (Wald statistic 9.68 (4 df), p=0.046), whereas there is no significant relationship for the level reporting no prescribed diet (Table 3).

Models fitting level of following a meal plan on level of family helping with diet, using both ordinal and nominal logistic regression were of poor fit, consequently, the relationship of family support and following a diet was not ascertained by this study. Even though the p-values of models controlling for age and gender were significant, the classification of correct cases was overall 38% for nominal with sometimes being classified 63% and rarely and mostly classified as 9 and 0% correctly, respectively. The high degrees of freedom, using Likert to Likert comparisons, may necessitate redefining the acceptable level of significance. Certainly, examining the correctness of fit is imperative prior to interpretation of any model and is particularly necessary with complex relationships among categories.

DISCUSSION

Our results indicated CAA received a poor quality of healthcare (45% not being prescribed a diet). A significant portion of those with inadequate HbA1c who were told followed a meal plan were likely to almost always follow; whereas those not told with inadequate HbA1c had no relationship with following a meal plan. The univariate analysis of clinically significant variables did not lead to a singular, multivariate model. The hope was diabetes education, social support and following a meal plan would be predictors, controlling for age and/or gender, for adequate glucose control. Only following a meal plan predicted level of glucose control. The discrepancy may be due to the exploratory nature of the study, the use of a cross-sectional design and self-reported data. In addition, a key variable, diabetes education, was not a consistent factor (the quantity, quality and temporality of the diabetes education received by the participants were indeterminable) and held a high degree of subjectivity. "Ever having a series of classes" could mean 3 months ago to one participant and 30 years ago to another. The quality of the nutritional intervention component of a diabetes education program could range from admonishments as "stay away from sweets" to helpful solutions such as recipes and meal suggestions. An intervention study of standardized, quality diabetes education would render a more consistent "diabetes education variable". These limitations may account, in part, for the lack of a significant relationship between diabetes education and adequate glucose control.

This study contributes toward some of the important aspects of diabetes management issues of Cuban Americans with type 2 diabetes. We investigated some of the clinically important variables in predicting successful diabetes management (diabetes education, following a meal plan, told to follow a diet, receiving a diet prescription, scheduling meals and social support). The outcome measurements of HbA1c, glucose level and glucose control were selected since CAA were found to have high percentages of overweight and obesity as well as poor lipid control and low physical activity levels (non-diabetic participants of our study not shown) (Table 1). This is consistent with other studies of Hispanic groups; an assessment of risk factors for Cuban American women indicated 24.4% smoking rate, 31.6% overweight and 75.5% ate junk food daily (Glanz et al, 2003). Hispanics and Hispanic subpopulations trail non-Hispanic whites in measures of persons aged 18 or over who participate in regular moderate physical activity (23% verses 35%) (CDC, 2008b).

We found poor glucose control by nearly half of the participants with type 2 diabetes. More than half of those participants with type 2 diabetes, on Medicare or Medicaid, did not receive diabetes education (62% of the participants report being covered in the last year by Medicare or Medicaid). Cuban Americans have a higher proportion covered by health insurance than other Hispanics at all ages. Non-Hispanic whites without medical insurance were 15.1% in 2002 (DHHS, 2002). Our sample with type 2 diabetes had a 14.8% uninsured rate. The problem does not appear to be health insurance coverage; but, rather the quality and their access to programs in diabetes self-management education and quality of medical care. Diet is a key factor in diabetes management. Our findings, 45% were never told to follow a meal plan or diet, is indicative of a significant proportion of individuals receiving less than adequate health care.

Even though Hispanics in the United States have a higher prevalence of type 2 diabetes and experience more complications than non-Hispanic whites, (NDSS, 2005; AACE, 2007) self-management practices may explain, in part, these differences (Coronado et al, 2007). Recent pilot intervention studies suggest the Hispanic population may be receptive to diabetes self-management as evidenced by improvements in related outcomes (lipid profile, glycemic control, HbA1c, BMI) (Glimer et al, 2005; Mauldon et al, 2006; Rosal et al, 2005). In particular, culturally appropriate and relevant interventions with a refresher course 6 to 9 months following the intervention were recommended (Glimer et al, 2005; Caballero et al, 2002).

Although this study advances our characterization of Cuban Americans it has several limitations. The sample, primarily from Miami-Dade County, may not reflect attitudes and practices of Cuban Americans in other regions. Despite efforts to recruit from a randomized mailing list, CAA were eligible volunteers who responded to mailings. Demographics, social characteristics and diabetes self-management data were self-reported at a single time point. The nutritional component, duration and effectiveness of diabetes education received was not assessed. The type and quality of diabetes education programs reimbursed by public health insurance may be different than what is provided by private health care.

Pre-intervention evaluation of environmental, personal, social, cultural and gender-specific barriers to following a meal plan and scheduling meals are warranted for the development of a successful intervention for this population. The results support the American Diabetes Association evidenced-based recommendation for ongoing, structured, culturally relevant diabetes education (ADA, 2007 a.). Diabetes self-management education (defined as an interactive, collaborative, ongoing process involving the person with diabetes and the educator(s) is essential to improve health outcomes for individuals with diabetes (Mensing et al, 2007). An effective program requires implementation and revision of standards congruent with evidenced-based knowledge (Mensing et al, 2007). The standards need to address the following broad areas: the structure of the program; processes of individual assessment; and program evaluation (Mensing et al, 2007). Efforts in this direction would, in turn help ameliorate the gap in the literature concerning Cuban Americans and health disparities.

CONCLUSIONS

While the quality, focus and frequency of diabetes education was not assessed in this study, key findings were the significant positive relationship between reporting receiving a prescribed diet to level of following a meal plan and HbA1c control. The advantage of having a medical professional prescribe a diet seems to be an important environmental support factor in this sample's diabetes care. Ongoing nutritional education is warranted for CAA since obesity rates are well above the national average. Nearly half CAA are not given dietary guidance, yet our results indicate CA may improve glycemic control by receiving dietary instructions. Future studies assessing quality of healthcare by the relationship of diabetes self-management skill and health outcome differentiating among the major Hispanic ethnic group in the United States are warranted.

ACKNOWLEDGEMENTS

The project described was supported by Award Number SO6 GM8205 from the National Institute Of Diabetes And Digestive And Kidney Diseases. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute Of Diabetes And Digestive And Kidney Diseases or the National Institutes of Health. The authors thank, Michele Swink, Jenny Esteves, Lizabeth Norell and Gariella Brissi for their help with data collection and data entry.

REFERENCES

AACE (2007). Diabetes Mellitus Clinical Practice Guideline Task Force: American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the Management of Diabetes Mellitus. Endocrine Practice, 13 (Suppl 1), 4-9.

ACP (2000) American College of Physicians--American Society of Internal Medicine: No Health Insurance? Its enough to make you sick. Latino community at great risk. White Paper, Philadelphia, PA, ACP ASIM.

ADA (2007 a.) American Diabetes Association: Nutrition Recommendations and Interventions for Diabetes: A position statement of the American Diabetes Association. Diabetes Care, 30, S48-S65.

ADA (2007 b.) American Diabetes Association. Position Statement: Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care, 30 (Suppl 1), S42-47.

Burchard, E.G., Ziv, E., Coyle, N., Gomez, S.L., Tang , H., Karter, A.J., Mountain, J.L., Perez-Stabel, E.J., Sheppard, D., & Risch, N. (2003). The Importance of Race and Ethnic Background in Biomedical Research and Clinical Practice. New England Journal of Medicine, 348, 1170-1175.

Caballero, A.C., Yohai, F., DeVillar, N.G., Monzillo, L., Romero, A., Lerman, I., Herrera-Acena, G. (2002). The Short and Long Term Effect of a Non Traditional Culturally Oriented Diabetes Education Program on Metabolic Control of Hispanic Patients with Type 2 Diabetes. Diabetes, 51 (Suppl 2), 77A.

Callaway, C.W., Chumlea, W.C. & Bouchard, C. (1988). Circumferences. In: Anthropometric Standardization Reference Manual. Lohman, T.G., Roche, A.F., & Martorell, R. ( Eds). Champaign, IL: Human Kinetics Books, 28-80.

Centers for Disease Control and Prevention (2008a). National Diabetes Fact Sheet: General Information and National Estimates on Diabetes in the United States, 2007. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention.

Center for Disease Control and Prevention (2008b). Family Support Project: Division of Diabetes Translation (DDT). Retrieved April, 6 2007 from http://www.cdc.gov/diabetes/projects/racial_init.htm

Coronado, G.D., Thompson B., Tejeda, S., Godina, R., & Chen, L.(2007). Sociodemographic Factors and Self-management Practices Related to Type 2 Diabetes among Hispanics and Non-Hispanic Whites in a Rural Setting. The Journal of Rural Health, 23(1):49-54.

DHHS (2002). Racial and Ethnic Differences in Health Insurance Coverage and Usual Source of Health Care (DHHS Chartbook #14) Retrieved June 14, 2007 from: http://www.meps.ahrq.gov/mepsweb/data_files/publications/cb14/cb14.shtml

DHHS (1985). Vital and Health Statistics Series 1, No. 19. In: National Center for Health Statistics: Plan and Operation of the Hispanic Health and Nutrition Examination Survey, 1982-1984. Washington, DC, U.S. Govt. Printing Office. Publication (PHS) 851-321.

Furino, A. & Munoz, E. (1991). Health Status Among Hispanics: Major Themes and New Priorities. JAMA, 265, 255-257.

Glanz K., Croyle, R.T., Chollette, V.Y., Pinn, W.V. (2003). Cancer-Related Health Disparities in Women American Journal of Public Health, 93(2), 292-298.

Glimer, T.P., Philis-Tsimikas, A., & Walker, C. (2005). Outcomes of Project Dulce: A Culturally Specific Diabetes Management Program. Annals of Pharmacotherapy, 39(5), 817-822.

Grieco, E.M. & Cassidy, R.C. (2001). Overview of race and Hispanic origin: Census 2000 Brief. United States Census 2000. Washington, DC: US Department of Commerce, US Census Bureau. Online article. Retrieved February 11, 2007 From: http://www.census.gov/prod/2001pubs/c2kbr011.pdf.

Harris, M.I., Cowie, C.C., Stern, M.P., Boyko, E.J., Reiber, G.E., & Bennet, P.H. (Eds.) (2005): Diabetes in America. Washington DC, U.S. Govt. Printing Office, DHHS Publication: NIH, 95-1468.

Mauldon, M., Melkus, G.D., Cagganello, M, (2006). Tomando Control: A Culturally Appropriate Diabetes Education Program for Spanish-speaking Individuals with Type 2 Diabetes Mellitus-Evaluation of a Pilot Project. The Diabetes Educator, 32(5), 751-760.

Mensing C, Boucher J, Cypress M et al. (2007). National Standards for Diabetes Self-Management Education Diabetes Care, 30 (Suppl. 1), S96-S103.

MMWR (2004) Center for Disease Control and Prevention: Health Disparities Experienced by Hispanics-United States. 53:935-937. [article online]. Retrieved April 5, 2007 from: http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5340a1.htm

NDSS (2005). Age-adjusted Prevalence of Diagnosed Diabetes per 100 Population (2005) Online Table..

National Diabetes Surveillance System. Retrieved March 17, 2007 from: http://www.omhrc.gov/templates/content.aspx?ID=3324.

National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (2008). National Diabetes Statistics, 2007 Fact Sheet. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health.

NIH (2007)National Heart, Lung, and Blood Institute. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults. Publication No. 98-4083 and Classification of Overweight and Obesity by BMI, Waist Circumference and Associated Disease Risk. Retrieved April 6, 2007 from: http://www.nhlbi.nih.gov/guidelines/obesity/prctgd_c.pdf

Nath, S. D. & Huffman, F. G. (2003). Lifestyle and Coronary Heart Disease Risk Factors of Cuban Americans with Type 2 Diabetes. FASEB, 17(4), 319A.

Novello, A.C., Wie, P.H. & Kleinman, D.V.(1991). Hispanic Health: Time for Data, Time for Action. JAMA, 265, 253-255.

Ramirez, A.C., Villarreal, R., Suarez, L. & Flores, E.T. (1995). The Emerging Hispanic Population: A foundation for Cancer Prevention and Control. J National Cancer Institute Monograph 18.

Rosal, M.C., Olendzki, B., Reed, G.W., Gumieniak, O., Scavron, J. & Ockene, I.L. (2005). Diabetes Self Management among Low-Income Spanish-Speaking Patients: A Pilot Study. Annals of Behavioral Medicine, 29(3), 225-235.

Smith C.A. & Barnett, E. (2005). Diabetes-related mortality among Mexican Americans, Puerto Ricans and Cuban Americans in the United States. Rev Panam Salud publica, 18(6), 381-387.

U.S. Census Bureau (2006). Health Insurance Coverage Status by Nativity, Citizenship, and Duration of Residence for Hispanic Population: 2005. Current Population Survey, Annual Social and Economic Supplement. Retrieved March 18, 2007 from: http://pubdb3.census.gov/macro/032006/health/h09a_000.htm

U.S. Census Bureau (2004 a.) Population by Region, sex, and Hispanic Origin Type, with percent Distribution by Region: 2004. Current Population Survey, Annual Social and Economic Supplement, Ethnicity and Ancestry Statistics Branch, Population Division. Retrieved March 18, 2007 from: http://www.census.gov/population/socdemo/hispanic/ASEC2004/ 2004CPS_tab18.2.html

U.S. Census Bureau (2004 b.), Profile of General Demographic Characteristic: Geographic Area: Florida. Update 2004. Retrieved March 18, 2007 from: http://factfinder.census.gov/servlet/QTTable?_bm=y&- geo_id=04000US12&qr_name=DEC_2000_SF1_U_DP1&ds_name=DEC_2000_SF1_U

U.S. Census Bureau (2000). QT-P9: Hispanic or Latino by Type: 2000. Geographic area: Miami-Dade County, Florida. Retrieved March 18, 2007 from: http://factfinder.census.gov/servlet/QTSubjectShowTablesServlet ?_ts=192279284343

FATMA G. HUFFMAN

JOAN A. VACCARO

Florida International University

SUBRATA NATH

University Of Texas Health Sciences Center At San Antonio

GUSTAVO G. ZARINI

Florida International University
Table 1. Descriptive Characteristics of Participants N =182 (a)

Characteristic                     Mean [+ or -] SD       N (5)

Age, yrs                          65.2 [+ or -] 12.1
  Female                          66.1 [+ or -] 12.4     114 (63)
  Male                            63.7 [+ or -] 11.4     68 (37)
BMI (kg/[m.sup.2])                 31.6 [+ or -] 6.6
  Underweight <18.5                                       0 (0)
  Normal 18.5-24.9                                      22 (12.1)
  Overweight 25-30                                      69 (37.9)
  Obese >30                                              91 (50)
Education
  HS or less                                             116 (64)
  Some College                                           31 (17)
  Bachelor's or Graduate degree                          35 (19)
Employment
  Working                                                53 (29)
  Homemaker                                              10 (18)
  Retired                                                72 (40)
  Disabled                                               24 (13)
  Unemployed                                              11 (6)
  Other                                                   4 (2)
Income
  < 20,000/yr                                            122 (67)
Health Insurance
  Public (Medicare or Medicaid)                         113(62.1)
  No health insurance in                                27 (14.8)
    the past 12 months
Received Diabetes Education
  Yes                                                   86 (47.3)
  No                                                    92 (50.5)
  Not sure                                               4 (2.2)
[HbA.sub.1c] (%) (b)               7.2 [+ or -] 2.9
  > 8.5                                                 39 (21.9)
  [less than or equal to] 8.5                           139 (78.1)
  FPG (mg/dL)
  Adequate ([less than                                  82 (45.1)
    or equal to] 126) (c)
  Inadequate (> 126)                                    100 (54.9)
FPG (mg/dL)                       143.9 [+ or -] 69.9
TG (mg/dL)                         176 [+ or -] 107
TC (mg/dL)                          189 [+ or -] 50

Abbreviations: BMI = body mass index, [HbA.sub.1c] = hemoglobin
A1c (glycated hemoglobin, FPG = fasted plasma glucose, TG =
triacylglycerides, TC = total serum cholesterol.

(a) Diabetic participants from study N=370 of diabetic and non
diabetic Cuban Americans. (b) N=178 participants with complete
data. (c) The American Diabetic Association's Statement 2003
was used at the time of the study; the 2007 statement classifies
adequate < 126 mg/dL. The difference is not statistically
significant.

Table 2 Ordinal Logistic Regression Models (a) Parameters
Diabetes Self-Management N=182 CAA

Variables                  OR      SE             df    CI

Predictor: Rx Diet (b)     N (% yes) 99 (54.4%)
Outcome: Meal Plan (c)     4.43    1.33           1    2.54, 7.79
Control: Age               1.03    1.01           1    1.01, 1.05
Predictor: Ever Told (d)   N (% yes) 7 (54.5%)
Outcome: Schedule (e)      .62     1.43           1    1.81, 7.26

Variables                    P

Predictor: Rx Diet (b)
Outcome: Meal Plan (c)     <.001
Control: Age               .014
Predictor: Ever Told (d)
Outcome: Schedule (e)      <.001

(a) Predictor and outcome variables were reversed for binary
logistic regression and confirmed the relationship found by for
ordinal logistic regression (data not shown)

(b) Are you currently (in the last 6 months) prescribed for any
special diet by your healthcare provider (nurse, dietitian diabetes
education, physician)? Yes/No

(c) How often do you follow a meal plan or diet? 1-5 Likert Scale
1=Never, 3= Sometimes, 5 = Always (n=178)

(d) Has your health care provider (nurse/dietitian/diabetes
educator/physician) ever told you to follow a meal plan or diet?
Yes, No, Not Sure95% Confidence intervals

(e) How often do you follow the schedule for your meals and snacks?
1-5 Likert Scale 1=Never, 3= Sometimes, 5 = Always

Table 3
Likelihood of Following a Meal Plan for CAA with HbA1c>
8.5 Stratified by Reporting Having a Prescribed Diet
within the past 6 months

Level                 FMP (a)   OR      SE     df       CI
No
Rx [Diet.sup.2] 1-5             ns             --       --
N=83(45.6%)
Yes
Rx Diet (b)           Wald=9.68          --    4        --
N=99(54.4%)           1         8.25    2.73   1    1.15, 59.0
                      2         3.75    2.15   1    .830, 16.9
                      3         5.40    2.27   1    1.08, 26.9
                      4         17.25   1.91   1    2.53, 117.7
Reference             5

Level                   P
No
Rx [Diet.sup.2] 1-5   0.917
N=83(45.6%)
Yes
Rx Diet (b)           .046
N=99(54.4%)           .036
                      .086
                      .040
                      .004
Reference

(a) How often do you follow a meal plan or diet? 1-5 Likert
Scale 1=Never, 3= Sometimes, 5 = Always (n=178)

(b) Are you currently (in the last 6 months) prescribed for
any special diet by your healthcare provider (nurse,
dietitian diabetes education, physician)?
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.