Free
Behavioral Health  |   November 2020
Associations Between Social Support and Diabetes-Related Distress in People With Type 2 Diabetes Mellitus
Author Notes
  • From the Touro University California College of Osteopathic Medicine, Vallejo, California (Drs. Young and Shubrook); the Touro University California College of Health Sciences and Education, Vallejo, California (Dr. Dugan); the Veterans Health Administration Sierra Nevada Health Care System, Reno, Nevada (Dr. Valencerina); the University of the Pacific Thomas J. Long School of Pharmacy and Health Sciences, Stockton, California (Ms. Wong); and the Ohio Health Grant Medical Center, Columbus, Ohio (Dr. Lo). 
  • Disclaimer: Dr Shubrook, a JAOA Associate Editor at the time of acceptance, was not involved in the editorial review or decision to publish this article. 
  • Financial Disclosures: None reported. 
  • Support: Funding was provided by Touro University California College of Pharmacy and College of Osteopathic Medicine to cover the cost of purchasing gift cards used to compensate participants for their time. 
  •  *Address correspondence to Clipper F. Young, PharmD, MPH, Touro University California College of Osteopathic Medicine, 1310 Club Drive, Vallejo, CA 94592-1119. Email: clipper.young@tu.edu
     
Article Information
Behavioral Health   |   November 2020
Associations Between Social Support and Diabetes-Related Distress in People With Type 2 Diabetes Mellitus
The Journal of the American Osteopathic Association, November 2020, Vol. 120, 721-731. doi:https://doi.org/10.7556/jaoa.2020.145
The Journal of the American Osteopathic Association, November 2020, Vol. 120, 721-731. doi:https://doi.org/10.7556/jaoa.2020.145
Abstract

Context: Diabetes is a complex, chronic condition and managing it can have psychosocial implications for patients, including an impact on relationships with their loved ones and physical wellness. The necessary modifications to daily behaviors can be very overwhelming, thus leading to diabetes-related distress.

Objective: To investigate the association between diabetes-related distress and perceived social support among people with type 2 diabetes.

Methods: This cross-sectional study surveyed a population with a lower socioeconomic status (Medi-Cal recipients, which are only given to low-income individuals) in Solano County, California. Patients who had type 2 diabetes mellitus, who were between 40 and 80 years old, and who had a medical appointment in the clinic(s) at least once between December 2015 and December 2016 were included. Patients who could not understand or speak English and patients whose primary care clinicians declined their participation in the study were excluded from the study. Each study participant was recruited at the end of their medical appointment, and the survey instrument in paper form was administered. The Problem Areas in Diabetes (PAID) scale, which indicates diabetes-related distress, and Multidimensional Scale of Perceived Social Support (MSPSS) with 3 subscales (family, friends, and significant others) were used in this study. Multiple linear regression models were used to analyze the associations between PAID and MSPSS surveys.

Results: For the 101 participants included in our study, multiple linear regression models showed statistically significant association between total MSPSS scores and total PAID scores (β = −.318; 95% CI, .577, −.0581; P=.017) as well as between MSPSS family subscale scores and total PAID scores (β= −.761; 95% CI, −1.35, −.168; P=.012). Among the 3 MSPSS subscales, higher perceived support from family members was found to be significantly associated with lower total PAID scores (β= −.761; 95% CI, −1.35, −.168; P=.012).

Conclusion: Our findings suggest that a higher level of perceived social support experienced was associated with lower diabetes-related distress among patients with type 2 diabetes. Osteopathic physicians have a central role in providing comprehensive, patient-centered, holistic care, and the attention to social support in chronic disease management can help remove barriers in providing optimal care.

Diabetes is a complex, chronic condition that can be daunting to manage physically and mentally; it can affect not only patients’ physical wellness, but also their relationships with loved ones. People with diabetes are advised to modify diet patterns, lead active lifestyles, adhere to medication regimens, perform blood glucose self-monitoring, and keep clinical appointments to minimize the risks of diabetes-related complications. Modifying daily behaviors to effectively manage diabetes can be very overwhelming, thus leading to diabetes-related distress.1 The Diabetes Attitudes, Wishes, and Needs study2 was the first international study to focus on the psychosocial aspects of diabetes management, addressing the needs for psychosocial support as an essential component for achieving glycemic goals and improving long-term health outcomes. In 2016, the American Diabetes Association (ADA) published the first position statement on guidelines for psychosocial care, which the ADA suggested be incorporated into routine care to improve health outcomes and quality of life.3 The ADA also recommended assessing affected individuals’ psychosocial factors during the initial encounter, at recurrent intervals during follow-up appointments, or when changes occurred in their diabetes status.3 Furthermore, the guidelines recommended that clinicians monitor psychosocial effects on self-management capabilities and evaluate the individuals’ self-care competencies.3 Patients’ diabetes distress is inversely associated with their blood glucose control; the lower the degree that a patient perceives their glycemic levels are managed, the higher the level of their diabetes-related emotional distress.4 
There has been substantial improvement in the management of type 2 diabetes, but the overwhelming increase in disease incidence means that there are still more people developing diabetes-related complications.5,6 Diabetes-related mortality and morbidity are the highest in people with lower socioeconomic status (SES). For example, patients from lower SES are less likely to achieve glycemic targets.7,8 A study9 of vulnerable people who lack social support and have a low SES showed that these patients experienced higher levels of distress, which resulted in lower levels of medication adherence and higher levels of blood glucose, blood pressure, and lipid profile. 
A systematic review10 of 37 studies with sample sizes ranging from 12 to 3535 participants found that social support positively affects a person’s medication adherence and disease management. Having social support also enhanced the patients’ quality of life, reduced stress, and helped them accept their diabetes.10 In addition, this review10 showed that people who lacked social support tended to have more complications and higher mortality. In the Diabetes Attitudes, Wishes, and Needs-2 study, psychosocial issues negatively affected self-management behaviors among patients with diabetes and their families.11-13 The study12 observed that 760 of 2,057 family members whose loved ones had diabetes (37.1%) did not know how to assist people with diabetes.12 
Collectively, these studies identified a strong link between psychosocial issues relating to self-management and the presence of significant gaps in the psychosocial aspects of diabetes management. Such recognition further confirms the important role of social support in empowering individuals with diabetes to perform self-care and disease management more effectively.14,15 Although hemoglobin A1c levels are an important clinical outcome, complications associated with uncontrolled diabetes can serve as another important indicator in guiding clinicians to evaluate patients’ risks for death. A recent study16 examined the association between diabetes burden and diabetes stress, including social support as a moderating factor.16 Few studies have investigated the nature of diabetes distress and social support pertaining to underserved, diverse populations with type 2 diabetes. Our study aimed to decipher the associations between perceived social support and diabetes-related distress; we hypothesized that higher perceived social support (captured by the total Multidimensional Scale of Perceived Social Support [MSPSS] scores) would be associated with lower diabetes-related distress (exhibited by the Problem Areas In Diabetes [PAID] scores) while controlling for age, HbA1c, SES, Diabetes Complications Severity Index (DCSI), ethnicity, and gender. 
Methods
This cross-sectional, survey-based study was conducted at Solano County Family Health Services Clinics in Vallejo and Fairfield, California. This health care system comprises 3 federally-qualified health centers (FQHCs) and serves people from lower socioeconomic backgrounds as they are Medi-Cal recipients (Medi-Cal benefits—California's version of the Medicaid program—are only given to low-income individuals). The study was approved by Touro University California Institutional Review Board and the Steering Committee at Solano County Family Health Services. 
Patients who had type 2 diabetes mellitus, were between 40 and 80 years old, and had a medical appointment in the clinic(s) at least once between December 2015 and December 2016 were included in this study. Patients were excluded if they could not understand or speak English and patients whose primary care clinicians declined their participation in the study. Consent forms were given to participants in the clinical setting at the time of care and before starting the surveys. 
Most of the recruited participants presented to the clinics for diabetes consultation services; some were recruited via primary care visits. Each potential participant was screened via the electronic medical record before they presented to the clinic for their medical appointment, and those who fitted the inclusion criteria were approached at the end of their medical visits to be included. The survey instrument in paper form was then administered. After PAID and MSPSS surveys were administered, clinical data (eg, HbA1c, serum creatinine, and urine protein) for each participant were retrieved from electronic medical records (EMR). The dates of clinical data noted in the EMR were within 1 year of the survey data collected. 
Survey Instrument
All the surveys used in this study have been validated in previous publications described here. 
The PAID questionnaire is a clinical assessment that has been used to identify an individual’s diabetes-specific emotional distress.16,17 The questions include treatment-related issues (3 items), food-related problems (3 items), social support-related problems (2 items), and diabetes-related emotional distress (12 items). The results of this 20-item survey are calculated based on a Likert scale from 0 to 4, indicating no problem (0), minor problem (1), moderate problem (2), somewhat serious problem (3), or serious problem (4).17 Once all the scores are added for each item, the sum of the total raw score is then multiplied by 1.25 to adjust the total PAID score to 100, which ranges from 0 to 100.16 
The MSPSS questionnaire is a widely used, self-reported measure that assesses 3 dimensions of perceived social support: family, friends, and significant others.18 This 12-item survey contains 4 items per source of support, using a Likert scale from 1 to 7 with 1 being “very strongly disagree” and 7 being “very strongly agree” with the predetermined statement. To capture the support from each source, a mean score is calculated for each subscale by summing 4 items and dividing the total by 4. A high total mean score on any subscale would indicate a high level of perceived social support from that source. For the total perceived social support score, each individual score from the 12 questions is added together and then divided by 12 to obtain a mean score. A mean score ranging from 1 to 2.9 would be considered low support; a score of 3 to 5 would be considered moderate social support; a score of 5.1 to 7 would be considered high support.19 
The DCSI incorporates laboratory data and diagnostic codes using the International Classification of Diseases, 9th or 10th Revision (ICD-9 or ICD-10) to quantify the long-term complications resulting from consistently elevated A1c levels.20 This index is composed of diabetes complications in 7 body systems: retinopathy, neuropathy, nephropathy, cerebrovascular, cardiovascular, peripheral vascular, and metabolic.20 The diagnoses were confirmed with ICD-9 and ICD-10 diagnostic codes and laboratory results before quantifying diabetes complication severity. Both ICD-9 and ICD-10 codes were used because updates in the EMR system occurred during our data collection period. The number of diagnoses, however, does not indicate the number of complications or severity level. For instance, if a participant had background retinopathy, diabetic ophthalmologic disease, and diabetic nephropathy, it would be considered a severity level of 2, even though there were 3 diagnoses, because background retinopathy and diabetic ophthalmologic disease are both classified under retinopathy and therefore were only counted as 1 point on the diabetes complication severity index. Thus, this index has a possible score ranging from 0 to 8. 
The Nam-Powers-Boyd Occupational Status Scores, developed in 2000, were used as a proxy to measure and represent an individual's socioeconomic status, which is defined by occupation, income, and education.21 The score spectrum ranges from the highest of 100, representing dentists, physicians, and surgeons, to the lowest of 1, representing counter attendants in a cafeteria and dish washers.21 Each participant's occupation, annual income, and education level was asked during data collection as part of demographics, and the occupational status score—using the reported occupation from each participant—was obtained from the Nam-Powers-Boyd Occupational Status Scores publication.21 
Statistical Analyses
To achieve an 80% power with α=.05, sample size calculation for linear regression models was performed and generated a sample size of 101.22 Multiple linear regression models were constructed via STATA version 14.2 (STATA Software) to determine the multiple levels of outcomes. Primary outcomes examined the associations between perceived social support (captured by MSPSS questionnaire, total, and subscales) and diabetes-related distress (reflected by PAID questionnaire). Secondary outcomes evaluated the associations between perceived social support and HbA1c, between perceived social support and diabetes complication severity (captured by the diabetes complication severity index), and between diabetes-related distress and diabetes complication severity. For multiple linear regression models, testing primary outcomes (ie, age, HbA1c levels, DCSI scores, SES [via occupational index scores], ethnicity, and gender) were included to adjust for variability in study participants. These 6 baseline characteristics were considered potential confounding factors, possibly affecting both psychologic and social measures captured by the PAID survey and the MSPSS questionnaire, respectively. The rationale behind including HbA1c levels and DCSI scores was that suboptimal social support might hinder the outcomes of diabetes self-management behaviors, which may elevate A1c levels and increase the risks of hospitalization, all of which can intensify diabetes-related distress. 
Results
The PAID and MSPSS survey instruments were completed by 101 participants. Sociodemographic, psychologic, and clinical data of the study participants were analyzed using descriptive statistics and are summarized in Table 1. The mean (SD) age of patients was 56.7 (7.8) years with a relatively equal distribution between genders (46 [45.5%] men and 55 [54.5%] women). Seventy-five participants (74.3%) reported an annual income of less than $20,000. The mean (SD) occupational status score was 40.0 (24.0), which was equivalent to a score representing an office clerk, a paper machine operator, or a construction worker.21 The race and ethnicity distribution of our study population was 30.7% African American (31), 23.8% Asian (24), 23.8% white (24), 20.8% Hispanic/Latino (21), and 0.9% uncategorized (1). As a whole, the study participants lived in diverse communities; according to the United States Census Bureau in 2019, the population of Solano County, California, was 59.6% white, 14.8% black, 16.2% Asian, and 27.3% Hispanic/Latino.23 Sixty-three of the study participants (62.4%) were recruited from the clinic in Vallejo, California, and 38 participants (37.6%) were recruited from the Fairfield clinic within the same safety-net healthcare system. 
Table 1.
Characteristics of Study Participants (n=101)
Sociodemographics
Age
 Mean (range), years 56.7 (41-76); SD, 7.8
Gender
 Female 55 (54.5%)
 Male 46 (45.5%)
Ethnicity/race
 African American 31 (30.7%)
 Asian 24 (23.8%)
 Caucasian 24 (23.8%)
 Hispanic/Latino 21 (20.8%)
 Other 1 (0.9%)
Nam-Power-Boyd Occupational Status score
 Mean (range) 40.0 (1-96); SD, 24.0
Annual income [frequency (percentage)]
 Less than $20,000 75 (74.3%)
 $20,000 to $34,999 12 (11.9%)
 $35,000 to $49,999 9 (8.9%)
 $50,000 to $75,000 3 (3.0%)
 More than $75,000 2 (2.0%)
Clinic site (frequency [%])
 Vallejo, California 63 (62.4%)
 Fairfield, California 38 (37.6%)
Psychosocial data
MSPSS scores (mean [range])
 Total 5.6 (2.2-7); SD, 1.2
 Family subscale 5.6 (1-7); SD, 1.6
 Friend subscale 5.4 (1.3-7); SD, 1.5
 Significant other subscale 5.9 (2-7); SD, 1.2
PAID questionnaire (diabetes-related distress)
 Mean for all participants (range) 23.1 (0-70); SD, 18.2
   Total score: 40 or above [frequency (percentage)] 21 (20.8%)
 Mean for males 22.6; SD, 18.3
 Mean for females 23.5; SD, 18.1
Clinical data
DSCI (diabetes complication severity index; median) 3
Hemoglobin A1c (mean [range]) 8.6% (5.5-14.7), SD, 2.0
70 mmol/mol (36.5-137)
Table 1.
Characteristics of Study Participants (n=101)
Sociodemographics
Age
 Mean (range), years 56.7 (41-76); SD, 7.8
Gender
 Female 55 (54.5%)
 Male 46 (45.5%)
Ethnicity/race
 African American 31 (30.7%)
 Asian 24 (23.8%)
 Caucasian 24 (23.8%)
 Hispanic/Latino 21 (20.8%)
 Other 1 (0.9%)
Nam-Power-Boyd Occupational Status score
 Mean (range) 40.0 (1-96); SD, 24.0
Annual income [frequency (percentage)]
 Less than $20,000 75 (74.3%)
 $20,000 to $34,999 12 (11.9%)
 $35,000 to $49,999 9 (8.9%)
 $50,000 to $75,000 3 (3.0%)
 More than $75,000 2 (2.0%)
Clinic site (frequency [%])
 Vallejo, California 63 (62.4%)
 Fairfield, California 38 (37.6%)
Psychosocial data
MSPSS scores (mean [range])
 Total 5.6 (2.2-7); SD, 1.2
 Family subscale 5.6 (1-7); SD, 1.6
 Friend subscale 5.4 (1.3-7); SD, 1.5
 Significant other subscale 5.9 (2-7); SD, 1.2
PAID questionnaire (diabetes-related distress)
 Mean for all participants (range) 23.1 (0-70); SD, 18.2
   Total score: 40 or above [frequency (percentage)] 21 (20.8%)
 Mean for males 22.6; SD, 18.3
 Mean for females 23.5; SD, 18.1
Clinical data
DSCI (diabetes complication severity index; median) 3
Hemoglobin A1c (mean [range]) 8.6% (5.5-14.7), SD, 2.0
70 mmol/mol (36.5-137)
×
For psychosocial characteristics, the mean (SD) total MSPSS score was 5.6 (1.2), indicating perceived high social support.18 The mean (SD) scores for family, friends, and significant others subscales were 5.6 (1.6), 5.4 (1.4), and 5.9 (1.2). The mean (SD) total PAID score was 23.1 (18.2), which indicated a moderate level of diabetes-related distress—a cutoff of 40 was used to determine the presence of severe diabetes-related psychologic problems.24 The mean (SD) total PAID score for men was 22.6 (18.3), whereas the mean (SD) for women was 23.5 (18.1). Twenty-one study participants (20.8%) expressed diabetes-related distress above a total score of 40 or above. 
In terms of clinical characteristics, the mean HbA1c of participants was 8.6% or 70 mmol/mol, and the median DCSI score, reflecting diabetes-related complications among the participants, was 3. 
Primary Outcomes
Table 2A shows the association between the total MSPSS scores and the total PAID scores determined by a multiple linear regression model (controlling for age, HbA1c, DCIS, SES, ethnicity, and gender). This regression model showed a statistically significant relationship between total perceived social support and diabetes-related distress (β= −.318; 95% CI, .577, −.0581; P=.017); β is the unstandardized regression coefficient. For every unit of increase in perceived social support captured by the MSPSS, there was a 0.3 unit decrease in diabetes-related distress captured by the PAID questionnaire. This statistical significance extended to the relationship between perceived family support and diabetes-related distress (β= −.761, 95% CI [−1.35, −.168], P=.012), meaning 1 unit of increase on the family subscale of the MSPSS corresponded to a 0.8 unit decrease in diabetes-related distress (Table 2C). No such associations were observed between the friend or significant other subscale scores and the total PAID scores; those results are shown in Table 2B and Table 2D, respectively. 
Primary Outcomes
Variable/covariatesa Unstandardized regression coefficient (β) P value 95% confidence interval
(a) Association between PAID score and total MSPSSb
 Total MSPSS −.318 .017 −.577, −.0581
 DCSI −.134 .753 −.972, .705
 HbA1c 1.65 .102 −.332, 3.62
 Ethnicity 1.05 .519 −2.16, 4.26
 SES .00722 .926 −.147, .161
 Age −.0384 .870 −.503, .426
 Gender −3.27 .380 −10.6, 4.11
(b) Association between PAID score and friends subscaleb
 Friends subscale −.553 .086 −1.19, .0802
 DCSI −.00689 .987 −.850, .836
 HbA1c 1.44 .159 −.573, 3.444
 Ethnicity 1.27 .443 −2.00, 4.54
 SES .0291 .709 −.125, .184
 Age −.0260 .913 −.498, .446
 Gender −2.85 .451 −10.3, 4.63
(c) Association between PAID score and family subscaleb
 Family subscale −.761 .012 −1.35, −.168
 DCSI −.147 .729 −.984, .690
 HbA1c 1.78 .078 −.204, 3.75
 Ethnicity .688 .671 −2.52, 3.90
 SES −.00328 .967 −.158, .152
 Age −.0652 .780 −.528, .398
 Gender −3.03 .415 −10.4, 4.31
(d) Association between PAID score and significant other subscaleb
 Significant other subscale −.635 .100 −.578, −.0581
 DCSI −.124 .775 −.972, .705
 HbA1c 1.67 .104 −.332, 3.62
 Ethnicity 1.06 .521 −2.16, 4.26
 SES .0199 .800 −.147, .161
 Age −.0457 .848 −.503, .426
 Gender −2.93 .441 −10.7, 4.11

a Covariates included in this multiple linear regression model are age, HbA1c, SES, DCSI, ethnicity, and gender.

b PAID score=dependent variable; MSPSS or subscale=independent variable.

Abbreviations: DCSI, Diabetes Complications Severity Index; MSPSS, Multidimensional Scale of Perceived Social Support; PAID, Problem Ares in Diabetes; SES, socioeconomic status.

Table 2.
Primary Outcomes
Variable/covariatesa Unstandardized regression coefficient (β) P value 95% confidence interval
(a) Association between PAID score and total MSPSSb
 Total MSPSS −.318 .017 −.577, −.0581
 DCSI −.134 .753 −.972, .705
 HbA1c 1.65 .102 −.332, 3.62
 Ethnicity 1.05 .519 −2.16, 4.26
 SES .00722 .926 −.147, .161
 Age −.0384 .870 −.503, .426
 Gender −3.27 .380 −10.6, 4.11
(b) Association between PAID score and friends subscaleb
 Friends subscale −.553 .086 −1.19, .0802
 DCSI −.00689 .987 −.850, .836
 HbA1c 1.44 .159 −.573, 3.444
 Ethnicity 1.27 .443 −2.00, 4.54
 SES .0291 .709 −.125, .184
 Age −.0260 .913 −.498, .446
 Gender −2.85 .451 −10.3, 4.63
(c) Association between PAID score and family subscaleb
 Family subscale −.761 .012 −1.35, −.168
 DCSI −.147 .729 −.984, .690
 HbA1c 1.78 .078 −.204, 3.75
 Ethnicity .688 .671 −2.52, 3.90
 SES −.00328 .967 −.158, .152
 Age −.0652 .780 −.528, .398
 Gender −3.03 .415 −10.4, 4.31
(d) Association between PAID score and significant other subscaleb
 Significant other subscale −.635 .100 −.578, −.0581
 DCSI −.124 .775 −.972, .705
 HbA1c 1.67 .104 −.332, 3.62
 Ethnicity 1.06 .521 −2.16, 4.26
 SES .0199 .800 −.147, .161
 Age −.0457 .848 −.503, .426
 Gender −2.93 .441 −10.7, 4.11

a Covariates included in this multiple linear regression model are age, HbA1c, SES, DCSI, ethnicity, and gender.

b PAID score=dependent variable; MSPSS or subscale=independent variable.

Abbreviations: DCSI, Diabetes Complications Severity Index; MSPSS, Multidimensional Scale of Perceived Social Support; PAID, Problem Ares in Diabetes; SES, socioeconomic status.

×
Secondary Outcomes
The associations between MSPSS (total and subscales) and HbA1c, between MSPSS (total and subscales) and DCSI, and between DCSI and total PAID scores were also examined (Table 3). There were no statistically significant associations found in any of these multiple linear regression models except the association between DCSI and PAID (p<.001). 
Secondary Outcomesa
Relationship Unstandardized regression coefficient (β) P value 95% CI
(a) Association between social support (total MSPSS and subscales) and HbA1cb
 Total vs. HbA1c .00628 .643 −.0206, .0331
 Family subscale vs. HbA1c .0289 .350 −.0322, .0899
 Friends subscale vs. HbA1c −.0169 .603 −.0814, .0475
 Significant other subscale vs. HbA1 .0303 .437 −.0467, .107
(b) Association between social support (total MSPSS and subscales) and DCSIc
 Total vs. DCSI −.0469 .185 −.1166, .0229
 Family subscale vs. DCSI −.0933 .250 −.253, .0665
 Friends subscale vs. DCSI −.0471 .584 −.217, .123
 Significant other subscale vs. DCSI −.171 .090 −.369, .0270
(c) Association between DCSI and PAIDd
 DCSI vs. PAID .187 <.001 .110, .263

a Covariates included in this multiple linear regression model are age, SES, DCSI, ethnicity, and gender.

b H1bA1c=dependent variable; MSPSS or subscales=independent variable.

c DCSI=dependent variable; total MSPSS or subscales=independent variable.

d PAID=dependent variable; DCSI=independent variable.

Abbreviations: DCSI, Diabetes Complications Severity Index; MSPSS, Multidimensional Scale of Perceived Social Support; PAID, Problem Areas in Diabetes; SES, socioeconomic status.

Table 3.
Secondary Outcomesa
Relationship Unstandardized regression coefficient (β) P value 95% CI
(a) Association between social support (total MSPSS and subscales) and HbA1cb
 Total vs. HbA1c .00628 .643 −.0206, .0331
 Family subscale vs. HbA1c .0289 .350 −.0322, .0899
 Friends subscale vs. HbA1c −.0169 .603 −.0814, .0475
 Significant other subscale vs. HbA1 .0303 .437 −.0467, .107
(b) Association between social support (total MSPSS and subscales) and DCSIc
 Total vs. DCSI −.0469 .185 −.1166, .0229
 Family subscale vs. DCSI −.0933 .250 −.253, .0665
 Friends subscale vs. DCSI −.0471 .584 −.217, .123
 Significant other subscale vs. DCSI −.171 .090 −.369, .0270
(c) Association between DCSI and PAIDd
 DCSI vs. PAID .187 <.001 .110, .263

a Covariates included in this multiple linear regression model are age, SES, DCSI, ethnicity, and gender.

b H1bA1c=dependent variable; MSPSS or subscales=independent variable.

c DCSI=dependent variable; total MSPSS or subscales=independent variable.

d PAID=dependent variable; DCSI=independent variable.

Abbreviations: DCSI, Diabetes Complications Severity Index; MSPSS, Multidimensional Scale of Perceived Social Support; PAID, Problem Areas in Diabetes; SES, socioeconomic status.

×
Discussion
This study evaluated the associations between perceived social support and diabetes-related distress in a population of patients with type 2 diabetes and a low SES. The mean age among all study participants was 56.7 years, indicating the potential for a considerable amount of life experience and knowledge. Study participants in this age group may have been responsible for taking care of multiple generations within the family, contributing to the high total and subscale scores from the MSPSS questionnaire. A strong sense of connection with people in their lives might have been translated into the strong perceived social support from family, friends, and significant others. 
The ADA position statement on psychosocial care of people with diabetes expresses the linkage between psychosocial factors and people’s psychologic/medical well-being relating to diabetes.25 Inadequate social support interferes with effective diabetes self-management behaviors which, in turn, may increase the risks of hospitalization and death.26 A 2018 study that revealed the relationship between diabetes distress and glycemic control further confirmed that glycemic control could be improved with effective approaches in reducing diabetes distress and enhancing social support from family and friends.27 Although evidence supporting and emphasizing the importance of social support and how family or friends influence the outcomes of diabetes self-management, some participants in the current study perceived having received inadequate social support, reflected by the mean scores at the lower end of the range in each subscale of the MSPSS questionnaire (Table 1). Clinicians are encouraged to support the needs of people with diabetes by assisting them in adjusting their diabetes self-management behaviors either by obtaining specific training in this aspect or by referring them to health care professionals who have been trained to address patients’ psychologic and emotional well-being. Incorporating psychosocial assessments, such as the PAID questionnaire and the MSPSS, and integrating the family and support systems into diabetes management would be a model for individualized treatment approaches. Acknowledging and recognizing the essential role of social support in diabetes management encourages clinicians to select appropriate interventions when interacting with both people with diabetes and their family members.28,29 The results of the present study confirmed the findings of Baek et al,30 where the relationship between diabetes distress and diabetes burden was significantly moderated by social support; the difference between the 2 studies was that social support was an independent variable in our study and a moderator in Baek et al.30 
According to the mean PAID score indicated in Table 1, the study participants as a group expressed a moderately low level of diabetes-related distress (PAID score average, 23.1; SD, 18.2), which might have been due to the relatively high levels of perceived social support received from the various sources. The relatively strong support from family, friends, or significant others perceived by participants might have been a neutralizing factor for diabetes-related distress. 
Despite the moderately low level of mean PAID scores relating to the management of and coping with diabetes, a wide range of PAID scores was expressed and captured, ranging from 0 to 70. Twenty-one study participants (20.8%) expressed diabetes-related distress above a total score of 40, indicating that they might need psychologic interventions provided by specialized healthcare professionals from multiple disciplines (eg, mental health) to help alleviate the perceived burden of managing and coping with diabetes. Previous studies have shown that diabetes distress scores were generally higher in an ethnically diverse sample.31,32 
With the use of the DCSI instrument and the ICD-10 codes, validity may have been affected because the DCSI was initially designed for use with ICD-9 codes; however, in a recently published study,20 the researchers reported that neither the original nor new DCSI models included all the possible complications or comorbidities associated with diabetes. Inconsistencies in charting (eg, documenting medical histories) and medical code selection were potential limitations of the present study. 
Despite the possible varying interpretations of what constituted family, friends, or significant others, statistically significant associations were discovered in multiple linear regression models—controlling for covariates—between total perceived social support scores and levels of diabetes-related distress. Furthermore, these significant associations extended between the family subscale scores from the MSPSS questionnaire and the data collected by the PAID questionnaire (Table 2). In this current study, these statistically significant associations indicate that participants’ higher perceived social support is associated with their lower diabetes-related distress30; additionally, diabetes-related distress has been shown to be associated with support received from family with and without considering age, HbA1c, SES, and DCSI. As the perceived total and family social support captured by the MSPSS questionnaire went up, the diabetes-related distress was reduced by the indicated number of points measured by the PAID questionnaire, signified by the negative coefficients. In our study, data from the friends and the significant other subscales did not show equally significant associations with the PAID scores collected. The PAID scores and other data should have been monitored during the data collection process as a means to obtain the most immediate feedback while the study was being conducted. 
This study had many limitations. A potential limiting factor in this study was the timing in which HbA1c levels were collected. The HbA1c level used for each study participant was the most recent result collected in the time period before the collection of psychosocial data via the survey instrument, which collected mostly self-reported data. Some of the latest HbA1c levels were collected within a 3-month interval after the last HbA1c check, whereas others were collected within a 6-month interval, depending on the previous glycemic control. HbA1c levels used in this cross-sectional study created a problem because the levels were included as a snapshot in time without considering the trend and the time interval in which the levels were collected. This discrepancy might have been another reason why no statistically significant associations were identified in secondary outcomes involving HbA1c levels (Table 3). The use of HbA1c levels might have been more appropriate if the trends were included instead of the latest levels. As in any cross-sectional study, causality cannot be established between studied variables from the results. Because depression is often a comorbid condition with diabetes distress, our survey instrument did not include a validated instrument to capture any levels of depression in study participants. Also, the results from this study can only be applied to people aged 40 to 80 years with type 2 diabetes who have a lower SES. 
As mentioned earlier, the definition of family may have varied greatly among participants. For example, 1 participant may have defined “family” as immediate relatives/blood-related individuals, and others may have viewed “family” as close friends. Future studies are needed to standardize the definition of each category (ie, family, friends, and significant others on the MSPSS questionnaire) to maintain the consistency and quality of data, even though this survey instrument has been validated.18,19 
Despite the limitations of this study, our findings demonstrate the importance of integrating patients’ social support circle into their diabetes self-management behaviors. Given the strong role of social support has on diabetes-related distress, clinicians are highly encouraged to focus not only on people with diabetes but also on their support system to optimize diabetes management outcomes and reduce the risk of diabetes-related complications. Educating the support team and identifying their roles can positively affect health outcomes. Evaluations of diabetes-related distress and social support are critical in achieving optimal diabetes self-management and should be integrated into routine diabetes care as suggested by the psychosocial care position statement. This is especially important in the context of an osteopathic medical model of care, where it has been well-documented that osteopathic approaches to patient care include biomechanical, respiratory-circulatory, metabolic-energy, neurologic, and behavioral.33 The behavioral approach/model pertains to the psychologic and social activities, where the results of this study can be integrated and applied. When Seffinger and colleagues discussed osteopathic philosophy,33 they describe the patient-care domain pertaining to the behavioral model as to “assess and treat the whole person—physical, psychological, social, cultural, behavioral, and spiritual aspects; collaborative partnership; individualized patient care and self-responsibility for healthy lifestyle choices.”33 Osteopathic physicians thus have a central role in providing comprehensive, patient-centered, holistic care, and the attention to social support and psychologic components in chronic disease management can help remove barriers in providing optimal care. 
Conclusion
The statistical significance of the amount of perceived social support experienced by people aged 40 to 80 years with type 2 diabetes and lower SES in our study can provide insights into understanding and helping to identify levels of diabetes-related distress. Among the 3 MSPSS subscales, perceived support received from family members was found to be significantly associated with PAID scores. Future longitudinal research is needed to determine whether higher levels of perceived social support help people with type 2 diabetes reduce their diabetes-related distress. Along with the results from previous studies, osteopathic physicians are encouraged to incorporate psychosocial assessments into clinical practice for strengthening type 2 diabetes management and improving outcomes clinically, psychologically, and socially. 
Author Contributions
Drs Young, Shubrook, and Lo and Ms Wong provided substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; Drs Young, Shubrook, Valencerina, and Dugan drafted the article or revised it critically for important intellectual content; all authors gave final approval of the version of the article to be published; and all authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. 
References
Baek RN, Tanenbaum ML, Gonzalez JS. Diabetes burden and diabetes distress: the buffering effect of social support. Ann Behav Med. . 2014;48:145-155. doi: 10.1007/s12160-013-9585-4 [CrossRef] [PubMed]
Peyrot M, Rubin RR, Lauritzen T, Snoek FJ, Matthews DR, Skovlund SE. Psychosocial problems and barriers to improved diabetes management: results of the Cross-National Diabetes Attitudes, Wishes and Needs (DAWN) Study. Diabet Med. . 2005;22(10):1379-1385. doi: 10.1111/j.1464-5491.2005.01644.x. [CrossRef] [PubMed]
Young-Hyman D, de Groot M, Hill-Briggs F, Gonzalez JS, Hood K, Peyrot M. Psychosocial care for people with diabetes: a position statement of the American Diabetes Association. Diabetes Care. 2016;39:2126-2140. doi: 10.2337/dc16-2053 [CrossRef] [PubMed]
Gonzalez JS, Shreck E, Psaros C, Safren, SA. Distress and type-2 diabetes treatment adherence: a mediating role for perceived control. Health Psychol. 2015;34(5):505-513. doi: 10.1037/hea0000131 [CrossRef] [PubMed]
Centers for Disease Control and Prevention. Diabetes Report Card 2014. CDC.gov website. http://www.cdc.gov/diabetes/pdfs/library/diabetesreportcard2014.pdf. Accessed February 2, 2018.
Guariguata L. By the numbers: new estimates from the IDF Diabetes Atlas Update for 2012. Diabetes Res Clin Pract. . 2012;98(3):524-525. doi: 10.1016/j.diabres.2012.11.006 [CrossRef] [PubMed]
Grintsova O, Maier W, Mielck A. Inequalities in health care among patients with type 2 diabetes by individual socio-economic status (SES) and regional deprivation: a systematic literature review. Int J Equity Health. . 2014;13:43. doi: 10.1186/1475-9276-13-43
Dalsgaard E-M, Skriver MV, Sandbaek A, Vestergaard M. Socioeconomic position, type 2 diabetes and long-term risk of death. PloS One. . 2015;10(5):e0124829. doi: 10.1371/journal.pone.0124829 [CrossRef] [PubMed]
Pandit AU, Bailey SC, Curtis LM, et al. Disease-related distress, self-care and clinical outcomes among low-income patients with diabetes. J Epidemiol Community Health. . 2014;68:557-564. doi: 10.1136/jech-2013-203063 [CrossRef] [PubMed]
Strom JL, Egede LE. The impact of social support on outcomes in adult patients with type 2 diabetes: a systematic review. Curr Diab Rep. 2012;12(6):769-781. doi: 10.1007/s11892-012-0317-0 [CrossRef] [PubMed]
Nicolucci A, Kovacs Burns K, Holt RIG, et al Diabetes Attitudes, Wishes and Needs second study (DAWN2TM): cross-national benchmarking of diabetes-related psychosocial outcomes for people with diabetes. Diabet Med. . 2013;30(7):767-777. doi: 10.1111/dme.12245 [CrossRef] [PubMed]
Burns K Kovacs, Nicolucci A, Holt RIG, et al Diabetes Attitudes, Wishes and Needs second study (DAWN2TM): cross-national benchmarking indicators for family members living with people with diabetes. Diabet Med. . 2013;30(7):778-788. doi: 10.1111/dme.12239 [CrossRef] [PubMed]
Holt RIG, Nicolucci A, Kovacs Burns K, et al Diabetes Attitudes, Wishes and Needs second study (DAWN2TM): cross-national comparisons on barriers and resources for optimal care--healthcare professional perspective. Diabet Med. . 2013;30(7):789-798. doi: 10.1111/dme.12242 [CrossRef] [PubMed]
Rosland AM, Heisler M, Choi HJ, Silveira MJ, Piette JD. Family influences on self-management among functionally independent adults with diabetes or heart failure: do family members hinder as much as they help? Chronic Illn. . 2010;6(1):22-33. doi: 10.1177/1742395309354608 [CrossRef] [PubMed]
Rintala TM, Jaatinen P, Paavilainen E, Astedt-Kurki P. Interrelation between adult persons with diabetes and their family: a systematic review of the literature. J Fam Nurs. . 2013;19(1):3-28. doi: 10.1177/1074840712471899 [CrossRef] [PubMed]
Reddy J, Wilhelm K, Campbell L. Putting PAID to diabetes-related distress: the potential utility of the problem areas in diabetes (PAID) scale in patients with diabetes. Psychosomatics. . 2013;54(1):44-51. doi: 10.1016/j.psym.2012.08.004 [CrossRef] [PubMed]
Polonsky WH, Anderson BJ, Lohrer PA, et al. Assessment of diabetes-related distress. Diabetes Care. . 1995;18(6):754-760. doi: 10.2337/diacare.18.6.754 [CrossRef] [PubMed]
Zimet GD, Powell SS, Farley GK, Werkman S, Berkoff KA. Psychometric characteristics of the Multidimensional Scale of Perceived Social Support. J Pers Assess. . 1990;55(3-4):610-617. doi: 10.1080/00223891.1990.9674095 [CrossRef] [PubMed]
Zimet GD, Dahlem NW, Zimet SG, Farley GK. The Multidimensional Scale of Perceived Social Support. J Pers Assess. . 1988;52:30-41. doi: 10.1207/s15327752jpa5201_2 [CrossRef]
Glasheen WP, Renda A, Dong Y. Diabetes Complications Severity Index (DCSI) -- Update and ICD-10 translation. J Diabetes Complicat. . 2017;31(6):1007-1013. doi: 10.1016/j.jdiacomp.2017.02.018 [CrossRef]
Nam CB, Boyd M. Occupational status in 2000: over a century of census-based measurement. Popul Res Policy Rev. . 2004;23(4):327-358. [CrossRef]
Tables of Sample Size Requirement for Multiple Regression. StataToDo. https://www.statstodo.com/SSizMultipleRegression_Exp.php. Accessed on September 21, 2020.
United States Census Bureau. QuickFacts. https://www.census.gov/quickfacts/solanocountycalifornia. Access on September 20, 2020.
Hermanns N, Kulzer B, Krichbaum M, Kubiak T, Haak T. How to screen for depression and emotional problems in patients with diabetes: comparison of screening characteristics of depression questionnaires, measurement of diabetes-specific emotional problems and standard clinical assessment. Diabetologia. 2006;49:469-477. doi: 10.1007/s00125-005-0094-2 [CrossRef] [PubMed]
Young-Hyman D, de Groot M, Hill-Briggs F, Gonzales JS, Hood K, Peyrot M. Psychosocial care for people with diabetes: a position statement of the American Diabetes Association. Diabetes Care. 2016;39:2126-2140. doi: 10.2337/dc16-2053 [CrossRef] [PubMed]
Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Med. 2010;7:e1000316. doi: 10.1371/journal.pmed.1000316 [CrossRef] [PubMed]
Lee AA, Piette JD, Heisler M, Rosland A-M. Diabetes distress and glycemic control: the buffering effect of autonomy support from important family members and friends. Diabetes Care. 2018;41(6):1157-1163. doi: 10.2337/dc17-2396. [CrossRef] [PubMed]
Rosland AM, Heisler M, Choi HJ, Silveira MJ, Piette JD. Family influences on self-management among functionally independent adults with diabetes or heart failure: do family members hinder as much as they help? Chronic Illn. . 2010;6(1):22-33. doi: 10.1177/1742395309354608 [CrossRef] [PubMed]
Rosland AM, Kieffer E, Israel B, et al When is social support important? The association of family support and professional support with specific diabetes self-management behaviors. J Gen Intern Med. 2008;23(12):1992-1999. doi: 10.1007/s11606-008-0814-7 [CrossRef] [PubMed]
Baek RN, Tanenbaum ML, Gonzalez JS. Diabetes burden and diabetes distress: the buffering effect of social support. Ann Behav Med. 2014;48(2):145-155. doi: 10.1007/s12160-013-9585-4 [CrossRef] [PubMed]
Ozcan B, Rutters F, Snoek FJ, et al High diabetes distress among ethnic minorities is not explained by metabolic, cardiovascular, or lifestyle factors: findings from the Dutch Diabetes Pearl cohort. Diabetes Care. . 2018;41(9):1854-1861. doi: 10.2337/dc17-2181 [CrossRef] [PubMed]
Schmidt CB, van Loon BJ Potter, Torensma B, Snoek FJ, Honig A. Ethnic minorities with diabetes differ in depressive and anxiety symptoms and diabetes-distress. J Diabetes Res. 2017;2017:1204237. doi: 10.1155/2017/1204237 [CrossRef] [PubMed]
Seffinger MA, King HH, Ward RC, et al Osteopathic philosophy. In: Chila AG, ed. Foundations of Osteopathic Medicine, 3rd ed. Lippincott Williams & Wilkins; 2011:3-22.
Table 1.
Characteristics of Study Participants (n=101)
Sociodemographics
Age
 Mean (range), years 56.7 (41-76); SD, 7.8
Gender
 Female 55 (54.5%)
 Male 46 (45.5%)
Ethnicity/race
 African American 31 (30.7%)
 Asian 24 (23.8%)
 Caucasian 24 (23.8%)
 Hispanic/Latino 21 (20.8%)
 Other 1 (0.9%)
Nam-Power-Boyd Occupational Status score
 Mean (range) 40.0 (1-96); SD, 24.0
Annual income [frequency (percentage)]
 Less than $20,000 75 (74.3%)
 $20,000 to $34,999 12 (11.9%)
 $35,000 to $49,999 9 (8.9%)
 $50,000 to $75,000 3 (3.0%)
 More than $75,000 2 (2.0%)
Clinic site (frequency [%])
 Vallejo, California 63 (62.4%)
 Fairfield, California 38 (37.6%)
Psychosocial data
MSPSS scores (mean [range])
 Total 5.6 (2.2-7); SD, 1.2
 Family subscale 5.6 (1-7); SD, 1.6
 Friend subscale 5.4 (1.3-7); SD, 1.5
 Significant other subscale 5.9 (2-7); SD, 1.2
PAID questionnaire (diabetes-related distress)
 Mean for all participants (range) 23.1 (0-70); SD, 18.2
   Total score: 40 or above [frequency (percentage)] 21 (20.8%)
 Mean for males 22.6; SD, 18.3
 Mean for females 23.5; SD, 18.1
Clinical data
DSCI (diabetes complication severity index; median) 3
Hemoglobin A1c (mean [range]) 8.6% (5.5-14.7), SD, 2.0
70 mmol/mol (36.5-137)
Table 1.
Characteristics of Study Participants (n=101)
Sociodemographics
Age
 Mean (range), years 56.7 (41-76); SD, 7.8
Gender
 Female 55 (54.5%)
 Male 46 (45.5%)
Ethnicity/race
 African American 31 (30.7%)
 Asian 24 (23.8%)
 Caucasian 24 (23.8%)
 Hispanic/Latino 21 (20.8%)
 Other 1 (0.9%)
Nam-Power-Boyd Occupational Status score
 Mean (range) 40.0 (1-96); SD, 24.0
Annual income [frequency (percentage)]
 Less than $20,000 75 (74.3%)
 $20,000 to $34,999 12 (11.9%)
 $35,000 to $49,999 9 (8.9%)
 $50,000 to $75,000 3 (3.0%)
 More than $75,000 2 (2.0%)
Clinic site (frequency [%])
 Vallejo, California 63 (62.4%)
 Fairfield, California 38 (37.6%)
Psychosocial data
MSPSS scores (mean [range])
 Total 5.6 (2.2-7); SD, 1.2
 Family subscale 5.6 (1-7); SD, 1.6
 Friend subscale 5.4 (1.3-7); SD, 1.5
 Significant other subscale 5.9 (2-7); SD, 1.2
PAID questionnaire (diabetes-related distress)
 Mean for all participants (range) 23.1 (0-70); SD, 18.2
   Total score: 40 or above [frequency (percentage)] 21 (20.8%)
 Mean for males 22.6; SD, 18.3
 Mean for females 23.5; SD, 18.1
Clinical data
DSCI (diabetes complication severity index; median) 3
Hemoglobin A1c (mean [range]) 8.6% (5.5-14.7), SD, 2.0
70 mmol/mol (36.5-137)
×
Variable/covariatesa Unstandardized regression coefficient (β) P value 95% confidence interval
(a) Association between PAID score and total MSPSSb
 Total MSPSS −.318 .017 −.577, −.0581
 DCSI −.134 .753 −.972, .705
 HbA1c 1.65 .102 −.332, 3.62
 Ethnicity 1.05 .519 −2.16, 4.26
 SES .00722 .926 −.147, .161
 Age −.0384 .870 −.503, .426
 Gender −3.27 .380 −10.6, 4.11
(b) Association between PAID score and friends subscaleb
 Friends subscale −.553 .086 −1.19, .0802
 DCSI −.00689 .987 −.850, .836
 HbA1c 1.44 .159 −.573, 3.444
 Ethnicity 1.27 .443 −2.00, 4.54
 SES .0291 .709 −.125, .184
 Age −.0260 .913 −.498, .446
 Gender −2.85 .451 −10.3, 4.63
(c) Association between PAID score and family subscaleb
 Family subscale −.761 .012 −1.35, −.168
 DCSI −.147 .729 −.984, .690
 HbA1c 1.78 .078 −.204, 3.75
 Ethnicity .688 .671 −2.52, 3.90
 SES −.00328 .967 −.158, .152
 Age −.0652 .780 −.528, .398
 Gender −3.03 .415 −10.4, 4.31
(d) Association between PAID score and significant other subscaleb
 Significant other subscale −.635 .100 −.578, −.0581
 DCSI −.124 .775 −.972, .705
 HbA1c 1.67 .104 −.332, 3.62
 Ethnicity 1.06 .521 −2.16, 4.26
 SES .0199 .800 −.147, .161
 Age −.0457 .848 −.503, .426
 Gender −2.93 .441 −10.7, 4.11

a Covariates included in this multiple linear regression model are age, HbA1c, SES, DCSI, ethnicity, and gender.

b PAID score=dependent variable; MSPSS or subscale=independent variable.

Abbreviations: DCSI, Diabetes Complications Severity Index; MSPSS, Multidimensional Scale of Perceived Social Support; PAID, Problem Ares in Diabetes; SES, socioeconomic status.

Table 2.
Primary Outcomes
Variable/covariatesa Unstandardized regression coefficient (β) P value 95% confidence interval
(a) Association between PAID score and total MSPSSb
 Total MSPSS −.318 .017 −.577, −.0581
 DCSI −.134 .753 −.972, .705
 HbA1c 1.65 .102 −.332, 3.62
 Ethnicity 1.05 .519 −2.16, 4.26
 SES .00722 .926 −.147, .161
 Age −.0384 .870 −.503, .426
 Gender −3.27 .380 −10.6, 4.11
(b) Association between PAID score and friends subscaleb
 Friends subscale −.553 .086 −1.19, .0802
 DCSI −.00689 .987 −.850, .836
 HbA1c 1.44 .159 −.573, 3.444
 Ethnicity 1.27 .443 −2.00, 4.54
 SES .0291 .709 −.125, .184
 Age −.0260 .913 −.498, .446
 Gender −2.85 .451 −10.3, 4.63
(c) Association between PAID score and family subscaleb
 Family subscale −.761 .012 −1.35, −.168
 DCSI −.147 .729 −.984, .690
 HbA1c 1.78 .078 −.204, 3.75
 Ethnicity .688 .671 −2.52, 3.90
 SES −.00328 .967 −.158, .152
 Age −.0652 .780 −.528, .398
 Gender −3.03 .415 −10.4, 4.31
(d) Association between PAID score and significant other subscaleb
 Significant other subscale −.635 .100 −.578, −.0581
 DCSI −.124 .775 −.972, .705
 HbA1c 1.67 .104 −.332, 3.62
 Ethnicity 1.06 .521 −2.16, 4.26
 SES .0199 .800 −.147, .161
 Age −.0457 .848 −.503, .426
 Gender −2.93 .441 −10.7, 4.11

a Covariates included in this multiple linear regression model are age, HbA1c, SES, DCSI, ethnicity, and gender.

b PAID score=dependent variable; MSPSS or subscale=independent variable.

Abbreviations: DCSI, Diabetes Complications Severity Index; MSPSS, Multidimensional Scale of Perceived Social Support; PAID, Problem Ares in Diabetes; SES, socioeconomic status.

×
Relationship Unstandardized regression coefficient (β) P value 95% CI
(a) Association between social support (total MSPSS and subscales) and HbA1cb
 Total vs. HbA1c .00628 .643 −.0206, .0331
 Family subscale vs. HbA1c .0289 .350 −.0322, .0899
 Friends subscale vs. HbA1c −.0169 .603 −.0814, .0475
 Significant other subscale vs. HbA1 .0303 .437 −.0467, .107
(b) Association between social support (total MSPSS and subscales) and DCSIc
 Total vs. DCSI −.0469 .185 −.1166, .0229
 Family subscale vs. DCSI −.0933 .250 −.253, .0665
 Friends subscale vs. DCSI −.0471 .584 −.217, .123
 Significant other subscale vs. DCSI −.171 .090 −.369, .0270
(c) Association between DCSI and PAIDd
 DCSI vs. PAID .187 <.001 .110, .263

a Covariates included in this multiple linear regression model are age, SES, DCSI, ethnicity, and gender.

b H1bA1c=dependent variable; MSPSS or subscales=independent variable.

c DCSI=dependent variable; total MSPSS or subscales=independent variable.

d PAID=dependent variable; DCSI=independent variable.

Abbreviations: DCSI, Diabetes Complications Severity Index; MSPSS, Multidimensional Scale of Perceived Social Support; PAID, Problem Areas in Diabetes; SES, socioeconomic status.

Table 3.
Secondary Outcomesa
Relationship Unstandardized regression coefficient (β) P value 95% CI
(a) Association between social support (total MSPSS and subscales) and HbA1cb
 Total vs. HbA1c .00628 .643 −.0206, .0331
 Family subscale vs. HbA1c .0289 .350 −.0322, .0899
 Friends subscale vs. HbA1c −.0169 .603 −.0814, .0475
 Significant other subscale vs. HbA1 .0303 .437 −.0467, .107
(b) Association between social support (total MSPSS and subscales) and DCSIc
 Total vs. DCSI −.0469 .185 −.1166, .0229
 Family subscale vs. DCSI −.0933 .250 −.253, .0665
 Friends subscale vs. DCSI −.0471 .584 −.217, .123
 Significant other subscale vs. DCSI −.171 .090 −.369, .0270
(c) Association between DCSI and PAIDd
 DCSI vs. PAID .187 <.001 .110, .263

a Covariates included in this multiple linear regression model are age, SES, DCSI, ethnicity, and gender.

b H1bA1c=dependent variable; MSPSS or subscales=independent variable.

c DCSI=dependent variable; total MSPSS or subscales=independent variable.

d PAID=dependent variable; DCSI=independent variable.

Abbreviations: DCSI, Diabetes Complications Severity Index; MSPSS, Multidimensional Scale of Perceived Social Support; PAID, Problem Areas in Diabetes; SES, socioeconomic status.

×