Abstract
Background:
A growing number of ethnic minority populations in the United States are from the Middle East and North Africa (MENA) region, specifically Lebanon. This region is witnessing one of the highest expected increases in diabetes prevalence. However, limited data exists on how social determinants of health impact clinical care for diabetes in this population. The aim of this study was to assess the social determinants of health and their impact on clinical outcomes in Lebanese adults with type 2 diabetes (T2DM).
Methods:
A convenience sample of Lebanese patients with T2DM was recruited from primary health care centers in Lebanon. Data on demographics and social determinants of health, including socioeconomic status, neighborhood and built environment, as well as psychosocial variables were collected. Clinical outcomes including Hemoglobin A1c (A1C), systolic (SBP) and diastolic blood pressure (DBP) were measured. Unadjusted and adjusted linear regression models were used to test for associations between the independent variables and each of the outcomes.
Results:
Out of the 300 adults with T2DM, 52% were men, 73% were married and 64% had an education level below high school. Results from multivariate analyses showed that food insecurity (β = 0.16, p = 0.01), owning an air conditioner (β = −0.64, p = 0.01), and commuting by walking (β = −0.93, p = 0.01) were independently associated with A1C. Predictors of DBP were male gender (β = 3.59, p = 0.03), age (β = −0.19, p = 0.005) and lack of confidence in filling medical forms (β = −4.89, p = 0.007), while male gender was the only predictor of SBP (β = 7.41, p = 0.008).
Conclusions:
This is the first study to examine the relationship between social determinants of health and clinical outcomes for diabetes in the MENA region. Our findings suggest that living in an underprivileged neighborhood and built environment was significantly and independently associated with poor clinical outcomes among adults with T2DM in Lebanon. Findings from this study will inform care for immigrant populations with diabetes from the MENA region.
Keywords: Type 2 diabetes, Social determinants of health, Glycemic control, Hemoglobin A1c, Blood pressure, Lebanon
1. INTRODUCTION
The burden of diabetes is rising globally and is characterized with significant morbidity, mortality, and increased health care cost [1]. Middle East and North African (MENA) region is facing one of the highest expected increases in diabetes prevalence [2]. Lebanon is among the top 10 countries in the MENA region with the highest prevalence, whereby 12.9% were reported to suffer from diabetes [3]. Despite this tremendous burden, the risk factors that contribute to the development of diabetes and its serious microvascular and macrovascular complications are not well understood. While much of the attributable risk associated with diabetic complications is due to conventional risk factors such as hyperglycemia, hypertension and dyslipidemia, a growing evidence base has identified social determinants of health as contributors to health outcomes [4–5]. The importance of identifying these social determinants of health in chronic diseases including diabetes was highlighted by various international organizations, such as the American Diabetes Association [6]. As defined by the World Health Organization, social determinants of health (SDHs) are upstream factors of health that are outside of an individual’s control with significant subsequent effects on health outcomes [7]. These include the socioeconomic, psychological and other factors that were shown to impact glycemic control and other diabetes outcomes through self-care, access and processes of care [8]. Socioeconomic determinants encompass availability of resources to meet daily needs, access to educational, economic and job opportunities, access to health care services, as well as household size and family support. Studies in the United States and United Kingdom have found associations between increased incidence, prevalence and burden of disease and poverty, hunger, aging, gender, racism, home circumstances, income, education, and socioeconomic status [9–11]. Psychological factors include such determinants as depression and adverse childhood experiences that were shown to contribute to both disease etiology and self-management [12–14]. Others are rooted in the cultural background of individuals, such as in having fatalistic beliefs and attitudes towards disease and are also correlated with diabetes outcomes, including studies conducted in Lebanon [4–5, 15–17]. Additionally, living in underprivileged neighborhood/built environment with inadequate access to healthy food was also linked to poor diabetes care and adverse health outcomes [18].
The patient populations in emerging economies, such as Lebanon, are largely uncharacterized, and the current preventive and treatment strategies remain sub-optimal. In response to the increasing impact of diabetes on the nation, Lebanon has created an operational policy for diabetes; however, there is no comprehensive Lebanese registry of patients with diabetes, no national survey in which blood glucose was measured, and only partial implementation of evidence-based guidelines for diabetes treatment [19]. Also, there is paucity of information detailing the etiological factors of patients with type 2 diabetes (T2DM), including the social determinants that may influence outcomes. Such data is needed to guide interventions and health policies for more effective and efficient management of diabetes in Lebanon. Thus, this study aims to identify social determinants of health in a sample of Lebanese adults with T2DM and to examine the associations with diabetes control (Hemoglobin A1C) and CVD risk factors (Systolic and diastolic blood pressure, SBP & DBP). Of even greater significance is the fact that a growing number of ethnic minority populations in the United States are from the MENA region, specifically Lebanon [20]. Many of these individuals are recent immigrants and/or refugees [20]. A significant proportion of individuals from the MENA region reside in Detroit and Michigan and there is limited information on how their lived experience and social risk factors both prior to coming to the United States and since maintaining residence in the US impact their diabetes control as well as CVD risk factors [20]. Therefore, findings from this study will inform care for immigrant populations with diabetes from the MENA region.
2. MATERIALS AND METHODS
2.1. Study Population and Procedure
A convenience sample of Lebanese adults with T2DM (N = 300) was recruited from several primary health care centers (PHCs) in Beirut, Mount Lebanon and North Lebanon. Participants were included in the study if they were above the age of 18 years old, Lebanese, clinically diagnosed with T2DM and able to communicate in Arabic. Exclusion criteria included mental confusion on interview, or reported alcohol or drug abuse/dependency, dementia, active psychosis, or acute mental disorders. Subjects who agreed to participate were asked to provide written consent, followed by collection of survey data. After filling out all questionnaires, researchers measured Hemoglobin A1C (A1C) and waist circumference. Systolic (SBP) and diastolic blood pressure (DBP), as well as weight and height were assessed and reported by the registered nurse at the PHC. The study was approved by LAU institutional review board (IRB) (#LAU.SAS.MB2.24/Sep/2018).
2.2. Data Collection
Clinical Outcomes
A1C levels were obtained using the FDA approved portable A1C EZ 2.0 Glycohemoglobin Analysis System (BioHermes, Wuxi, China). The assay only requires about 3 microliters of capillary or venous blood sample and uses the boronate affinity chromatography technique that secures no interference from hemoglobin. It has received both NGSP (National Glycohemoglobin Standardization Program) & IFCC (International Federation of Clinical Chemistry) certificates. Blood pressure (mmHg) was measured and reported by the registered nurse at the PHCs using a sphygmomanometer, after 5 min of rest sitting on a chair. Weight (kg) and height (m) were collected with minimal clothing and without any shoes to calculate the Body Mass Index as weight (kg)/height (m2). Finally, waist circumference (cm) was measured by investigators via a measuring tape, using standard procedures.
Demographics
A self-reported questionnaire was used to collect participant demographic characteristics that included age in years (expressed both as a continuous variable and categorical 18 – 49, 50 – 64, 65 – 94), gender (as a dichotomous male/female), marital status (as a dichotomous married/not married), education level (as a dichotomous </≥ school graduate), employment status (as a dichotomous employed/not employed), monthly household income in USD (<$500; $500 – $1499; ≥$1500; not reported), availability of health insurance (as a dichotomous Yes/No), and confidence in filling medical forms (as a dichotomous Yes/No).
Social Determinants of Health
Socioeconomic Status Variables
Socioeconomic determinants were assessed using self-reported questions. Continuous variables included family size, number of individuals with financial independence, number of persons per bedroom and number of rooms in the household. Financial status (comfortable have more than enough to make ends meet, have enough to make ends meet, do not have enough to make ends meet), as well as accommodation status (as a dichotomous owned/rented) were also reported.
Psychosocial Variables
Depression was assessed using PHQ-9, which is a brief questionnaire that scores each of the 9 DSM-IV criteria for depression as “0” (not at all), to “3” (nearly every day). PHQ9 scores ranges are as follows: 0–4 non to minimal, 5–9 mild, 10–14 moderate, 15–19 moderately severe, 20–27 severe PHQ9 [21]. The items of the PHQ-9 Arabic translated scale were highly consistent based on reliability analyses (Cronbach’s alpha =.88) [22].
Adverse Childhood Experience (ACE) was assessed using the 10-item ACE scale that assesses the degree to which the respondent experienced childhood maltreatment [23]. The measures in the original scale are highly interrelated and correlated (Cronbach alpha of .88 for the 10 discrete binary items (no/yes)) [24]. In the present study, the questionnaire was translated and back translated from Arabic and then piloted for ease of comprehension.
Fatalism in diabetes was assessed using the Arabic version of the Diabetes Fatalism scale, a 12-item questionnaire. It has 3 subscales: emotional distress, perceived self-efficacy, and spiritual coping [25]. The Arabic version of the scale (DFS-Ar) was validated by the research team (Cronbach’s alpha of 0.86). The item analyses of the three subscales also revealed good reliability with Cronbach’s alphas of 0.87 for subscale 1 (emotional distress), 0.85 for subscale 2 (spiritual coping) and 0.89 for subscale 3 (perceived self-efficacy) [4].
Neighborhood/Built Environment Variables
Food insecurity was assessed using the US Household Food Security Survey Module: Six-Item Short Form [26]. Additive scores were categorized into: 0–1 high or marginal food security, 2–4 low food security, 5–6 very low food security. A scale that was derived from the same questionnaire was validated in Lebanon with good internal validity and reasonable reliability: item in-fits from 0.73 to 1.16 [27].
Other neighborhood/built environment variables were assessed using self-reported questions on presence of household facilities including electricity, drinking water, television, cable subscription, telephone, air conditioner, transportation, heater, wireless-internet subscription, computer, refrigerator (all as dichotomous Yes/No), as well as mean of transportation (public transportation, car, walking).
2.3. Statistical Analysis
Continuous variables (age, family size, financial independence, number of persons per bedroom, number of rooms, psychosocial variables, and food insecurity) were expressed as mean ± SD, while categorical variables (age, gender, marital status, education, employment, income health insurance, confidence in filling medical forms, and neighborhood/built environment variables) were expressed as counts and percentages. Three unadjusted linear regression models were run to test for correlations between the different variables and each of the clinical outcomes (A1C, SBP and DBP). Finally, three adjusted multiple linear regression models were conducted to understand the independent correlates of diabetes control (A1C) and CVD risk (SBP and DBP), while adjusting for all predictors. For both unadjusted and adjusted regression analyses, outcomes (A1C, SBP & DBP) were treated as continuous variables. All analyses were run using Stata V.16. Significance was determined based on a two tailed alpha of p<0.05.
3. RESULTS & DISCUSSION
3.1. Characteristics of study participants
Demographic characteristics of study participants are presented in table 1. The study population had a mean age of 60.29 (SD = 12.07) and was approximately equally distributed between women (48%) and men (52%). Most of the participants were married (73%), had less than high school education (64%), were unemployed (55%) with a household monthly income below 1500 USD (61%), and reported having no confidence in filling out medical forms (71%).
Table 1:
Demographic characteristics of Lebanese adults with T2DM recruited from PHCs in Beirut, Lebanon.
| M (SD) | |
|---|---|
| Age | 60.29 ± 12.07 |
| N (%) | |
| Age (years) | |
| 18 – 49 | (51, 17) |
| 50 – 64 | (149, 49.67) |
| 65 – 94 | (100, 33.33) |
| Sex | |
| Female | (144, 48) |
| Male | (156, 52) |
| Marital status | |
| Married | (219, 73) |
| Not married | (81, 27) |
| Education (years) | |
| < high school grad | (173, 63.84) |
| ≥ high school grad | (98, 36.16) |
| Employment | |
| Employed | (134, 44.67) |
| Unemployed | (166, 55.33) |
| Income | |
| <500$ | (91, 30.3) |
| 500$ – 1499$ | (92, 30.67) |
| ≥1500$ | (30, 10) |
| Not reported | (87, 29) |
| Health Insurance | |
| Yes | (139, 46.8) |
| No | (158, 53.2) |
| Confidence in filling out medical forms | |
| Yes | (86, 28.76) |
| No | (213, 71.24) |
Table 1 provides a summary description of the study population by demographic characteristics
Social Determinants of Health are presented in table 2. The average family size included 4.5 members with 2 reported financially independent individuals, on average. Mean number of rooms in the household was 3.5 with an average of 2 persons per bedroom. Most of the study population did not have enough to make ends meet (61.74 %), while 64% of the study population had rented rather than owned (34%) accommodation. The psychosocial variables included DFS12 mean total score of 35.73 (SD = 7.9), ACE total score with a mean of 0.6 (SD = 1.07) and PHQ9 total score with a mean of 7.19 (SD = 5.12). As for the neighborhood-built environment variables, food insecurity mean score was 1.04 (SD = 1.85), which is classified as “low food security”. All participants had electricity and the majority had drinking water (84%), television (99%), cable subscription (72%), telephone (97%) air conditioner (62%), heater (82%), and refrigerator (93%). On the other hand, 62% did not have a computer and 40% did not have wireless-internet subscription. Half of the study population commuted by car (51%), and the other half by public transportation (34%) or walking (12.67%).
Table 2:
Social Determinants of Health of Lebanese adults with T2DM recruited from PHCs in Beirut, Lebanon.
| Socioeconomic Status Variables | |
|---|---|
| M (SD) | |
| Family size | 4.48 ± 1.88 |
| Financial independence | 1.63 ± 1.92 |
| Number of persons per bedroom | 1.99 ± 0.91 |
| Number of rooms | 3.49 ± 1.51 |
| N (%) | |
| Financial status | |
| Comfortable; have more than enough to make ends meet | (31, 10.4) |
| Have enough to make ends meet | (83, 27.85) |
| Do not have enough to make ends meet | (184, 61.74) |
| Accommodation | |
| Owned | (108, 36) |
| Rented | (192, 64) |
| Psychosocial Variables | |
| M (SD) | |
| Diabetes Fatalism (Dfs12) | 35.73 ± 7.9 |
| Adverse childhood experience (ACE) Total Score | 0.6 ± 1.07 |
| Patient Health Questionnaire (PHQ9) Total Score | 7.19 ± 5.12 |
| Neighborhood/Built Environment Variables | |
| M (SD) | |
| Food Insecurity Total Score | 1.04 ± 1.85 |
| N (%) | |
| Electricity | |
| Yes | (300, 100) |
| Drinking water | |
| Yes | (252, 84) |
| No | (48, 16) |
| Television | |
| Yes | (298, 99.33) |
| No | (2, 0.67) |
| Cable subscription | |
| Yes | (216, 72) |
| No | (84, 28) |
| Telephone | |
| Yes | (292, 97.33) |
| No | (8, 2.67) |
| Air Conditioner | |
| Yes | (185, 61.67) |
| No | (115, 38.33) |
| Heater | |
| Yes | (246, 82) |
| No | (54, 18) |
| Wireless-internet subscription | |
| Yes | (180, 60) |
| No | (120, 40) |
| Computer | |
| Yes | (113, 37.67) |
| No | (187, 62.33) |
| Refrigerator | |
| Yes | (280, 93.33) |
| No | (20, 6.67) |
| Means of transportation | |
| Public transportation | (102, 34) |
| Car | (153, 51) |
| Walking | (38, 12.67) |
Table 2 presents detailed social determinants of health characteristics of the study population across socioeconomic status, psychosocial, and neighborhood/built environment
3.2. Unadjusted linear regression model of associations with A1C, SBP and DBP
Table 3 provides the unadjusted correlations between the independent variables and clinical outcomes. Regarding the socioeconomic status variables, significant negative correlation was found between the number of rooms and A1C (β = −0.13, p = 0.04). Moreover, with respect to neighborhood/built environment variables, the food insecurity total score showed a significant positive correlation with A1C (β = 0.15, p = 0.003), whereas participants who had a computer showed significantly lower A1C levels (β = −0.49, p = 0.014). In addition, SBP was found to be significantly lower among participants who had cable subscription (β = −6.8, p= 0.004), computer (β = −5.27, p = 0.017) and commuted by car (β = −7.07, p = 0.001). To add, SBP was found to be significantly higher among participants who commuted by public transportation (β = 5.94, p = 0.009) and by walking (β = 8.18, p = 0.01). Diastolic blood pressure was significantly lower among those who commuted by car (β = −3.34, p = 0.011). As for the socioeconomic status variables, a significant positive correlation was found between not having enough money to make ends meet and both SBP (β = 9.95, p = 0.05) and DBP (β = 4.62, p = 0.03) blood pressure. Finally, in the demographic characteristics section, only age showed a significant negative correlation with diastolic blood pressure (β = −0.17, p = 0.001).
Table 3:
Unadjusted linear regression model of associations with A1C, SBP and DBP showing the unadjusted correlations between the independent variables and clinical outcomes of Lebanese adults with T2DM recruited from PHCs in Beirut, Lebanon.
| Variables | A1C | SYSBP | DBP | |||
|---|---|---|---|---|---|---|
| β | β | β | ||||
| Demographic Characteristics | ||||||
| Age | −0.007 | −0.002 | −0.17 | |||
| Sex | ||||||
| Male (ref) | 0.19 | 2.33 | 1.63 | |||
| Marital status | ||||||
| Married (ref) | 0.292 | 3.12 | −0.51 | |||
| Education | ||||||
| ≥ high school grad (ref) | −0.30 | −1.007 | −1.24 | |||
| Employment status | ||||||
| Employed (ref) | −0.22 | 0.016 | 1.01 | |||
| Income | ||||||
| 500$ –1499$ | −0.13 | −1.82 | 1.87 | |||
| ≥1500$ | −0.48 | −4.66 | −1.24 | |||
| Not reported | −0.17 | 4.66 | 3.04 | |||
| Insurance | ||||||
| Yes | −0.35 | −3.44 | −1.23 | |||
| Confidence in filling out medical forms | ||||||
| No | −0.20 | −0.16 | −1.84 | |||
| Socioeconomic Status Variables | ||||||
| Financial status | ||||||
| Have enough to make ends meet | 0.009 | 5.72 | 1.29 | |||
| Do not have enough to make ends meet | 0.32 | 9.95 | 4.62 | |||
| Family size | 0.06 | 0.26 | 0.21 | |||
| Financial independence | −0.01 | −0.2 | 0.51 | |||
| Number of persons per bedroom | 0.14 | −0.64 | −0.94 | |||
| Number of rooms | −0.13 | −0.33 | −0.44 | |||
| Psychosocial Variables | ||||||
| Diabetes Fatalism (Dfs12) | 0.01 | 0.16 | 0.1 | |||
| Adverse Childhood Experience (ACE) total score | −0.12 | −0.59 | −0.27 | |||
| Patient Health Questionnaire (PHQ9) | 0.03 | −0.02 | 0.04 | |||
| Neighborhood/Built Environment Variables | ||||||
| Food insecurity total score | 0.15 | 0.77 | 0.49 | |||
| Drinking water | ||||||
| Yes | −0.09 | −0.22 | −1.55 | |||
| Television | ||||||
| Yes | −1.31 | 21 | −0.7 | |||
| Cable subscription | ||||||
| Yes | −0.3 | −6.8 | −2.13 | |||
| Telephone | ||||||
| Yes | −0.14 | 6.2 | −1.75 | |||
| Air conditioner | ||||||
| Yes | −0.39 | 1.08 | −0.34 | |||
| Heater | ||||||
| Yes | −0.08 | 0.31 | 0.36 | |||
| Wireless-internet subscription | ||||||
| Yes | −0.12 | −3.33 | 0.42 | |||
| Computer | ||||||
| Yes | −0.49 | −5.27 | −1.16 | |||
| Refrigerator | ||||||
| Yes | −0.05 | −2.55 | −0.64 | |||
| Means of transportation | ||||||
| Public | 0.07 | 5.94 | 1.78 | |||
| Car | −0.35 | −7.07 | −3.34 | |||
| Walking | −0.34 | 8.18 | 1.98 | |||
Significant values p<0.05
Table 3 presents detailed information on the relationship between social determinants of health variables (i.e., socioeconomic status, psychosocial, and neighborhood/built environment) and clinical outcomes including hemoglobin A1c, systolic blood pressure and diastolic blood pressure for unadjusted models. Significant associations are marked with asterisks.
3.3. Multiple linear regression model of associations with A1C, SBP and DBP
Table 4 provides the adjusted multivariate analyses with each of the study outcomes. Out of the neighborhood/built environment variables, food insecurity (β = 0.16, p = 0.01), owning an air conditioner (β = −0.64, p = 0.01), and commuting by walking (β = −0.93, p = 0.01) were associated with A1C after correcting for all other potential variables. None of the demographic characteristics were significantly linked to A1C, except for male gender that had borderline significance (β = 0.53, p = 0.05). As for SBP and DBP, male gender was the only predictor of SBP (β = 7.41, p = 0.008), while male gender (β = 3.59, p = 0.03), age (β = −0.19, p = 0.005) and lack of confidence in filling medical forms (β = −4.89, p = 0.007) were the demographic variables that were significantly and independently associated with DBP. None of the neighborhood/built environment variables predicted SBP or DBP. Also, all socioeconomic and psychosocial variables were not significantly associated with any of the study outcomes.
Table 4:
Adjusted multiple linear regression model with A1C, SBP and DBP showing the adjusted multivariate analyses by each study outcome for Lebanese dults with T2DM recruited from PHCs in Beirut, Lebanon.
| Variables | A1C | SYSBP | DBP | |||
|---|---|---|---|---|---|---|
| β | β | β | ||||
| Demographic Characteristics Variables | ||||||
| Age | −0.014 | −0.06 | −0.19 | |||
| Sex | ||||||
| Male (ref) | 0.533 | 7.41 | 3.59 | |||
| Marital status | ||||||
| Married (ref) | 0.373 | 2.58 | −1.74 | |||
| Education | ||||||
| ≥ high school grad (ref) | 0.16 | 0.96 | 0.8 | |||
| Employment status | ||||||
| Employed | −0.44 | 2.03 | −0.78 | |||
| Income | ||||||
| 500$ – 1499$ | 0.04 | −2.42 | 0.8 | |||
| ≥1500$ | −0.49 | −1.48 | −1.13 | |||
| Not reported | −0.05 | 3.64 | 2.67 | |||
| Insurance | ||||||
| Yes | −0.25 | −0.82 | 0.75 | |||
| Confidence in filling out medical forms | ||||||
| No | −0.1 | −1.88 | −4.89 | |||
| Socioeconomic Status Variables | ||||||
| Financial status | ||||||
| Have enough to make ends meet | −0.35 | 3.65 | −0.2 | |||
| Do not have enough to make ends meet | −0.51 | 6.31 | 2.18 | |||
| Family size | 0.08 | 0.33 | 0.56 | |||
| Financial independence | −0.05 | −0.6 | 0.04 | |||
| Number of persons per bedroom | −0.06 | −1.05 | −1.61 | |||
| Number of rooms | −0.04 | 1.47 | 0.26 | |||
| Psychosocial Variables | ||||||
| Diabetes Fatalism (Dfs12) | 0.01 | 0.087 | 0.11 | |||
| Adverse Childhood Experience (ACE) Total Score | −0.13 | −0.9 | −0.34 | |||
| Patient Health Questionnaire (PHQ9) | 0.03 | −0.27 | −0.1 | |||
| Neighborhood/Built Environment Variables | ||||||
| Food insecurity total score | 0.16 | 0.64 | 0.64 | |||
| Drinking water | ||||||
| Yes | 0.05 | 6.36 | 1.18 | |||
| Television | ||||||
| Yes | −0.63 | 19.07 | −3.57 | |||
| Cable subscription | ||||||
| Yes | −0.002 | −0.53 | 0.45 | |||
| Telephone | ||||||
| Yes | 0.02 | 2.12 | −2.71 | |||
| Air conditioner | ||||||
| Yes | −0.64 | 4.28 | 0.23 | |||
| Heater | ||||||
| Yes | 0.41 | 2.41 | 0.93 | |||
| Wireless-internet subscription | ||||||
| Yes | 0.38 | −1.81 | 1.43 | |||
| Computer | ||||||
| Yes | −0.31 | −4.42 | 0.11 | |||
| Refrigerator | ||||||
| Yes | −0.43 | −4.88 | −0.99 | |||
| Means of transportation | ||||||
| Public | −0.28 | 5.67 | 1.1 | |||
| Car | −0.48 | −3.44 | −3.99 | |||
| Walking | −0.93 | 2.52 | −2.79 | |||
Significant values p<0.05
Table 4 presents detailed information on the relationship between social determinants of health variables (i.e., socioeconomic status, psychosocial, and neighborhood/built environment) and clinical outcomes including hemoglobin A1c, systolic blood pressure and diastolic blood pressure for fully adjusted models. Significant associations are marked with asterisks.
Each model is adjusted for age, sex, marital status, education, employment status, monthly income, insurance, confidence in filling out health forms, family size, financial independence, number of persons per bedroom, number of rooms, diabetes fatalism, adverse childhood experience, depression (PHQ9), food insecurity, home availability of drinking water, television, cable subscription, air conditioner, heater, wireless-internet subscription, and refrigerator, and means of transportation.
In a convenience sample of Lebanese adults with type 2 diabetes and low socioeconomic status, correlates of diabetes control (A1C) included food insecurity, owning an air conditioner, and commuting by walking. Predictors of CVD risk in our sample, namely SBP and DBP, were male gender for both outcomes, as well as age and lack of confidence in filling medical forms for DBP.
Diabetes research and prevention is currently focused on tackling the social determinants of health that underlie the development and etiology of type 2 diabetes rather than solely treating its complications. International organizations, including the American Diabetes Association (ADA), recognize the importance of SDH and their effects on T2DM outcomes, which guide diabetes management and interventions.6 In the present study, we examined the effect of SDH (socioeconomic status, psychological variables and neighborhood/built environment) on clinical outcomes in T2DM in Lebanon, including A1C, SBP and DBP.
3.4. Discussion
Our results showed that food insecurity was significantly and independently associated with A1C, whereby patients with higher A1C levels had a higher food insecurity total score. This finding is consistent with the literature on food insecurity especially among people with diabetes, which was found to hinder proper diabetes self-management leading to higher levels of A1C [18]. Indeed, food insecurity is characterized by the inability to afford diabetes-appropriate foods, overeating during food adequacy, and impaired medication adherence [18]. Household food insecurity in Lebanon has been linked to inadequate dietary intakes and obesity [28], yet our study is the first to assess its effect on diabetes outcomes and highlights the need to address this construct in diabetes management. Conversely, a better built environment reflected by owning an air conditioner, was linked to lower levels of A1C in our sample. This is consistent with previous studies showing that living in neighborhoods with social advantages and improved living conditions had significantly lower A1C levels than those who lived in poorer conditions[29]. The findings emphasize the importance of a comprehensive approach in diabetes management taking into consideration the living conditions of patients with T2DM [29]. Commuting by walking was also inversely and independently related with A1C suggesting that the higher the levels of physical activity in patients with T2DM, the lower their A1C levels. It is well established that long term physical activity in patients with T2DM ameliorates A1C levels by increasing cardiovascular fitness and decreasing BMI levels and the pro- inflammatory status of obesity known to exacerbate A1C levels[30].
As for blood pressure and in line with reports from Lebanon and the world [31–33], male gender was found to be an independent risk factor for increased SBP and DBP. Apart from the genetic predisposition, women tend to have higher medical visits and access to health care services, which in turn increases their frequency of health screening. Accordingly, early detection of hypertension in women leads to early treatment and control of the disease. Men, on the other hand, tend to be driven by the communities’ guidelines of masculinity that often pressures them not to seek medical care or checkups, which in turn exacerbates underlying chronic diseases like hypertension [34].
Additionally, age was significantly and inversely associated with DBP in our study. In addition, lack of confidence in filling medical forms was inversely associated with DBP. This may be linked to compromised health literacy and self-efficacy, known to impact diabetes outcomes [35]. These findings can also be interpreted, in light of the fact that Middle eastern countries have strong family ties and social support [36]. Thus, older adults who are either illiterate or do not have enough knowledge to fill out medical forms usually have family members or caregivers taking care of their health status, medication compliance, lifestyle factors, and monitoring their blood pressure. As such, these will probably have more controlled DBP as compared to those living on their own. Indeed, family empowerment has been significantly associated with controlled blood pressure for the elderly that lived in supportive families [37].
3.5. Limitations
Limitations are inherent in the cross-sectional design of the study that does not allow for establishing causation. Secondly, many variables in this study are based on self-report and therefore may be susceptible to recall bias. Also, findings cannot be generalized as the recruitment of our convenience sample was from selected primary health care centers in Lebanon. Additionally, although we did adjust for possible confounders, residual confounding might still affect results.
3.6. Implications
In conclusion, this study is the first to provide preliminary data on social determinants of health and their impact on T2DM outcomes in Lebanon. These findings set the stage for population-based research to determine the burden of T2DM, and effective strategies for its treatment. Results revealed that living in underprivileged neighborhood/built environment in Lebanon characterized by food insecurity and lack of household facilities is associated with poor diabetes outcomes (higher A1C). Moreover, in addition to the classical determinants of blood pressure, lack of confidence in filling medical forms that might indicate poor health literacy and self-efficacy was linked to higher DBP.
Better understanding of causal pathways and upstream factors underlying diabetes control in Lebanon and their effect on A1C and CVD risk (SBP & DBP) is a novel and effective approach for better diabetes management. Studying social determinants of health in diabetes is a key factor to be incorporated in clinical based practice and correct clinical interventions to achieve optimal glycemic control and lower the risk of complications, including CVD [6, 38]. This would further guide policies and population-based interventions tackling these intertwined determinants and effectively managing diabetes instead of only focusing on the direct causes of this disease. Further research is warranted to investigate other social determinants and to inform the development and study of pertinent interventions addressing predictors of poor diabetes management in Lebanon.
Finally, given the fact that a significant proportion of individuals from the MENA region, a large proportion of who are Lebanese Americans, reside in Detroit and Michigan and there is limited information on how their lived experience and social risk factors both prior to coming to the United States and since maintaining residence in the US impact their diabetes control as well as CVD risk factors [20], findings from this study will inform care for immigrant populations with diabetes from the MENA region.
Acknowledgements
Effort for this study was partially supported by the National Institute of Diabetes and Digestive Kidney Disease (K24DK093699, R01DK118038, R01DK120861, PI: Egede), the National Institute for Minority Health and Health Disparities (R01MD013826, PI: Egede/Walker).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Financial Disclosures: No financial disclosures are reported by the authors of this paper.
Competing interests: The authors declare that they have no competing interests.
REFERENCES
- 1.Control CfD, Prevention. National diabetes statistics report, 2020. Atlanta, GA: Centers for Disease Control and Prevention, US Department of Health and Human Services. 2020:12–15. [Google Scholar]
- 2.Sherif S, Sumpio BE. Economic development and diabetes prevalence in MENA countries: Egypt and Saudi Arabia comparison. World journal of diabetes. 2015;6(2):304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.International Diabetes Federation (IDF). 2019.
- 4.Sukkarieh-Haraty O, Egede LE, Abi Kharma J, Bassil M. Predictors of Diabetes Fatalism Among Arabs: A Cross-Sectional Study of Lebanese Adults with Type 2 Diabetes. Journal of religion and health. 2018;57(3):858–868. [DOI] [PubMed] [Google Scholar]
- 5.Sukkarieh-Haraty O, Egede LE, Abi Kharma J, Bassil M. Diabetes fatalism and its emotional distress subscale are independent predictors of glycemic control among Lebanese patients with type 2 diabetes. Ethnicity & health. 2019;24(7):767–778. [DOI] [PubMed] [Google Scholar]
- 6.Hill-Briggs F, Adler NE, Berkowitz SA, et al. Social determinants of health and diabetes: a scientific review. Diabetes Care. 2021;44(1):258–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Organization WH. A conceptual framework for action on the social determinants of health. 2010.
- 8.Walker RJ, Gebregziabher M, Martin-Harris B, Egede LE. Relationship between social determinants of health and processes and outcomes in adults with type 2 diabetes: validation of a conceptual framework. BMC endocrine disorders. 2014;14(1):1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Martikainen PT, Marmot MG. Socioeconomic differences in weight gain and determinants and consequences of coronary risk factors. The American journal of clinical nutrition. 1999;69(4):719–726. [DOI] [PubMed] [Google Scholar]
- 10.Marmot MG, Stansfeld S, Patel C, et al. Health inequalities among British civil servants: the Whitehall II study. The Lancet. 1991;337(8754):1387–1393. [DOI] [PubMed] [Google Scholar]
- 11.Marmot M, Wilkinson R. Social determinants of health. Oup Oxford; 2005. [Google Scholar]
- 12.Shields ME, Hovdestad WE, Pelletier C, Dykxhoorn JL, O’Donnell SC, Tonmyr L. Childhood maltreatment as a risk factor for diabetes: findings from a population-based survey of Canadian adults. BMC public health. 2016;16(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Campbell JA, Farmer GC, Nguyen-Rodriguez S, Walker RJ, Egede LE. Using path analysis to examine the relationship between sexual abuse in childhood and diabetes in adulthood in a sample of US adults. Preventive medicine. 2018;108:1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Egede LE, Zheng D, Simpson K. Comorbid depression is associated with increased health care use and expenditures in individuals with diabetes. Diabetes care. 2002;25(3):464–470. [DOI] [PubMed] [Google Scholar]
- 15.Walker RJ, Smalls BL, Hernandez-Tejada MA, Campbell JA, Davis KS, Egede LE. Effect of diabetes fatalism on medication adherence and self-care behaviors in adults with diabetes. General hospital psychiatry. 2012;34(6):598–603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Educators AAoD. Cultural considerations in diabetes education. AADE Practice Synopsis Chicago, IL: American Association of Diabetes Educators. 2015. [Google Scholar]
- 17.Kueh YC, Morris T, Borkoles E, Shee H. Modelling of diabetes knowledge, attitudes, self-management, and quality of life: a cross-sectional study with an Australian sample. Health and quality of life outcomes. 2015;13(1):1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Seligman HK, Davis TC, Schillinger D, Wolf MS. Food insecurity is associated with hypoglycemia and poor diabetes self-management in a low-income sample with diabetes. Journal of health care for the poor and underserved. 2010;21(4):1227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.World Health Organization (WHO). Diabetes country profiles. 2016.
- 20.Green M, Resnicow K, Tariq M, Syed N, Alhawli A, Patel M. Risk and Protective Factors for Cost-Related Nonadherence Among Middle East and North African (MENA) Adults. Ethn Dis. 2022;32(1):11–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sawaya H, Atoui M, Hamadeh A, Zeinoun P, Nahas Z. Adaptation and initial validation of the Patient Health Questionnaire–9 (PHQ-9) and the Generalized Anxiety Disorder–7 Questionnaire (GAD-7) in an Arabic speaking Lebanese psychiatric outpatient sample. Psychiatry research. 2016;239:245–252. [DOI] [PubMed] [Google Scholar]
- 23.Felitti VJ, Anda RF, Nordenberg D, et al. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. American journal of preventive medicine. 1998;14(4):245–258. [DOI] [PubMed] [Google Scholar]
- 24.Murphy A, Steele M, Dube SR, et al. Adverse childhood experiences (ACEs) questionnaire and adult attachment interview (AAI): Implications for parent child relationships. Child abuse & neglect. 2014;38(2):224–233. [DOI] [PubMed] [Google Scholar]
- 25.Egede LE, Ellis C. Diabetes and depression: global perspectives. Diabetes research and clinical practice. 2010;87(3):302–312. [DOI] [PubMed] [Google Scholar]
- 26.Service USDoAER. US household food security survey module: Six-item short form. In: US Department of Agriculture, Economic Research Service; Washington, DC; 2012. [Google Scholar]
- 27.Sahyoun NR, Nord M, Sassine AJ, Seyfert K, Hwalla N, Ghattas H. Development and validation of an Arab family food security scale. The Journal of nutrition. 2014;144(5):751–757. [DOI] [PubMed] [Google Scholar]
- 28.Jomaa L, Naja F, Cheaib R, Hwalla N. Household food insecurity is associated with a higher burden of obesity and risk of dietary inadequacies among mothers in Beirut, Lebanon. BMC public health. 2017;17(1):1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kowitt SD, Donahue KE, Fisher EB, Mitchell M, Young LA. How is neighborhood social disorganization associated with diabetes outcomes? A multilevel investigation of glycemic control and self-reported use of acute or emergency health care services. Clinical diabetes and endocrinology. 2018;4(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Najafipour F, Mobasseri M, Yavari A, et al. Effect of regular exercise training on changes in HbA1c, BMI and VO2max among patients with type 2 diabetes mellitus: an 8-year trial. BMJ Open Diabetes Research and Care. 2017;5(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Singh JN, Nguyen T, Kerndt CC, Dhamoon AS. Physiology, blood pressure age related changes. StatPearls [Internet]. 2020. [PubMed] [Google Scholar]
- 32.Mouhtadi BB, Kanaan RMN, Iskandarani M, Rahal MK, Halat DH. Prevalence, awareness, treatment, control and risk factors associated with hypertension in Lebanese adults: A cross sectional study. Global cardiology science & practice. 2018;2018(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Noubani A, Nasreddine L, Sibai AM, Tamim H, Isma’eel H. Prevalence, awareness, and control of hypertension in greater Beirut Area, Lebanon. International journal of hypertension. 2018;2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Everett B, Zajacova A. Gender differences in hypertension and hypertension awareness among young adults. Biodemography and social biology. 2015;61(1):1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Osborn CY, Cavanaugh K, Wallston KA, Rothman RL. Self-efficacy links health literacy and numeracy to glycemic control. Journal of health communication. 2010;15(S2):146–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Tajvar M, Fletcher A, Grundy E, Arab M. Social support and health of older people in M iddle E astern countries: A systematic review. Australasian journal on ageing. 2013;32(2):71–78. [DOI] [PubMed] [Google Scholar]
- 37.Keshvari M, Hedayati B, Moeini M, Alhani F. A survey on the effect of implementation of a family-centered empowerment model on blood pressure and empowerment dimensions in the elderly people with hypertension. Journal of education and health promotion. 2015;4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Braveman PA, Egerter SA, Woolf SH, Marks JS. When do we know enough to recommend action on the social determinants of health? American journal of preventive medicine. 2011;40(1):S58–S66. [DOI] [PubMed] [Google Scholar]