Ethnic differences in psychological outcomes among people with diabetes: USA results from the second Diabetes Attitudes, Wishes, and Needs (DAWN2) study

Figures & data

Table 1. Description of measures.

Measure	Definition/explanation
Disease, demographic and socioeconomic factors
Diabetes duration	Natural log of the number of years since diagnosis (log taken after adding 1 to the number so that minimum value returned is zero)
Income	1 = less than $15,000; 2 = $15,000 to $24,999; 3 = $25,000 to $34,999; 4 = $35,000 to $49,999; 5 = $50,000 to $74,999; 6 = $75,000 to $99,999; 7 = Over $100,000
Education	1 = less than high school; 2 = high school graduate; 3 = some college; 4 = college graduate; 5 = graduate school
Health status
BMI	Natural log of calculated BMI
Health index	EuroQol-5D health utility score18
Self-rated health	Visual analog item of the EuroQol-5D18
Complications	Count of the following conditions: ‘a stroke; an open wound on your foot that did not heal without medical care; some or all of your foot/leg amputated; your kidneys not working properly; damage to your eyesight; nerve damage; problems with your sexual functioning; heart disease or attack’, possible range = 0 to 8
Frequency of hypoglycemia	Hypoglycemia in last 12 months: 0 = none; 20 = less than monthly; 40 = monthly; 60 = several per month; 80 = weekly; 100 = daily
Frequency of severe hypoglycemia	Natural log of the number of times in the last month respondent had ‘severely low blood sugar such that you experienced problems with memory, language, thinking, or judgment and were unable to treat yourself and needed help from someone to restore your blood sugar levels’ (log taken after adding 1 to the number so that minimum value returned is zero)
Healthcare access/utilization
See physician	Count of healthcare professionals seen in the past 12 months for diabetes (primary care physician/general practitioner/family practitioner; doctor who specializes in diabetes; other specialist for your diabetes (e.g., eye care/foot care/kidney doctor), possible range = 0 to 3
See other healthcare professional	Count of healthcare professionals seen in the past 12 months for diabetes (a nurse or educator; dietitian/nutritionist; mental health professional; other type of care professional), possible range = 0 to 4
Diabetes education	Count of whether patient has received (a) group or (b) diabetes education in last 12 months and whether patient has received (a) group or (b) diabetes education prior to the last 12 months, possible range = 0 to 4
Access barriers	Count of the number of items that respondent indicated are currently a problem in best managing their diabetes (‘it is difficult for you to pay for the diabetes medications you need to best manage your diabetes; the medications you need to best manage your diabetes are not always available where you live; it is difficult for you get the blood sugar testing supplies and devices you need to best manage your diabetes’), possible range = 0 to 3
Patient-centered care	Mean of 12 items from PACIC-DSF scored 0–1008
Subjective burden of diabetes
Hypoglycemia worry	‘I am very worried about the risk of hypoglycemic (low blood sugar) events’ (1 = fully disagree; 2 = mainly disagree; 3 = mainly agree; 4 = fully agree)
Weight worry	‘I feel very anxious about my weight’ (1 = fully disagree; 2 = mainly disagree; 3 = mainly agree; 4 = fully agree)19
Dietary restrictions	‘My diabetes restricts my ability to eat the way I want’ (1 = fully disagree; 2 = mainly disagree; 3 = mainly agree; 4 = fully agree)
Diabetes-related discrimination	‘I have been discriminated against because I have diabetes’ (1 = fully disagree; 2 = mainly disagree; 3 = mainly agree; 4 = fully agree)8
Social support/burden
Diabetes support network	Natural log of the number of people that respondent could talk to about diabetes if s/he needed or wanted to besides the healthcare professionals seen for diabetes (log taken after adding 1 to the number so that minimum value returned is zero)
Diabetes-related family arguments	‘My family argues with me about how I choose to take care of my diabetes’ (1 = fully disagree; 2 = mainly disagree; 3 = mainly agree; 4 = fully agree)
Diabetes support gap	Count of the number of items to which respondent did not answer ‘very supportive’. ‘Overall, how supportive have the following people been in helping you manage your diabetes over the past 12 months?: your family; friends or people close to you; people at work or school; healthcare team; other people in your community’ (response options = not supportive; somewhat supportive; very supportive; not applicable), possible scores = 0 to 5
Psychological outcome
Well-being	WHO-5 Wellbeing Index, scored 0–10020
Quality of life	WHOQOL-BREF, single item scored 0–10021
Diabetes distress	PAID 5-item scale, scored 0–10022
Diabetes impact	Mean of six DIDP items (physical health; financial situation; relationship with family, friends, peers; leisure activities; work or studies; emotional well-being), scored 0–100 with higher scores representing more negative impact (response options = very positive impact; positive impact; slightly positive impact; no impact; slightly negative impact; negative impact; very negative impact)8
Diabetes empowerment scale	Mean of five DES-DSF items, scored 0–1008

BMI, body mass index; DES-DSF, Diabetes Empowerment Scale–DAWN Short Form; DIDP, DAWN Impact of Diabetes Profile; PACIC-DSF, Patient Assessment of Chronic Illness Care–DAWN Short Form; PAID-5, Problem Areas in Diabetes Scale 5; WHO-5, World Health Organization 5-item Well-Being Index; WHOQOL-BREF, World Health Organization Quality of Life (an abbreviated version of the WHOQOL-100).

Table 2. Raw data for design-related and weighting factors by ethnicity.

Sample descriptor	White non-Hispanic n = 447	African American n = 241	Hispanic n = 194	Chinese American n = 173
Diabetes type/treatment (n, %)**
Type 1 diabetes	73, 16.3	34, 14.1	32, 16.5	20, 11.6
Type 2 diabetes (no medication)	88, 19.7	28, 11.6	15, 7.7	17, 9.8
Type 2 diabetes (non-insulin medication)	155, 34.7	103, 42.7	87, 44.8	93, 53.8
Type 2 diabetes (insulin medication)	131, 29.3	76, 31.5	60, 30.9	43, 24.9
Gender (n, %)**
Male	191, 42.7	105, 43.6	93, 47.9	109, 63.0
Female	256, 57.3	136, 56.4	101, 52.1	64, 37.0
Education (n, %)**
Less than high school graduate	18, 4.0	25, 10.4	50, 25.8	18, 10.4
High school graduate	243, 54.4	48, 19.9	29, 14.9	18, 10.4
Some college	88, 19.7	87, 36.1	55, 28.4	23, 13.3
College graduate	52, 11.6	43, 17.8	37, 19.1	61, 35.3
Graduate school	46, 10.3	38, 15.8	23, 11.9	53, 30.6
Income (n, %)**
Less than $15,000	47, 12.1	44, 19.7	35, 19.0	18, 11.8
$15,000 to $24,999	58, 14.9	39, 17.5	30, 16.3	18, 11.8
$25,000 to $34,999	57, 14.7	19, 8.5	19, 10.3	14, 9.2
$35,000 to $49,999	82, 21.1	46, 20.6	41, 22.3	14, 9.2
$50,000 to $74,999	74, 19.0	39, 17.5	29, 15.8	27, 17.6
$75,000 to $99,999	35, 9.0	21, 9.4	14, 7.6	29, 19.0
Over $100,000	36, 9.3	15, 6.7	16, 8.7	33, 21.6
Diabetes duration, years (M ± SD)*	16.3 ± 12.1	14.4 ± 10.6	13.8 ± 11.0	14.5 ± 11.8
Age, years (M ± SD)**	59.0 ± 13.5	56.6 ± 14.6	53.1 ± 17.0	58.3 ± 17.6
Body mass index (M ± SD)**	33.7 ± 9.1	32.6 ± 8.3	32.0 ± 7.8	26.4 ± 5.5

*Overall P < 0.05 for ethnicity.

**Overall P < 0.001 for ethnicity.

M, Mean.

Table 3. Health status, healthcare access/utilization, subjective burden of diabetes and social support/burden by ethnicity.

Predictor	White non-Hispanic n = 447 mean (95% CI)	African American n = 241 mean (95% CI)	Hispanic n = 194 mean (95% CI)	Chinese American n = 173 mean (95% CI)
Health status
BMI (log)***	3.45†‡ (3.34, 3.48)	3.41‡|| (3.38, 3.45)	3.41‡ (3.38, 3.45)	3.22†§|| (3.18, 3.26)
Self-rated health*	70.83 (68.85, 72.80)	73.04‡|| (70.28, 75.81)	67.79† (64.84, 70.75)	68.46† (64.94, 71.99)
Health index*	0.73§ (0.70, 0.76)	0.74§ (0.70, 0.77)	0.67†‡|| (0.63, 0.72)	0.76§ (0.72, 0.81)
Complications	1.12 (0.99, 1.24)	1.22 (1.04, 1.39)	1.14 (0.96, 1.33)	1.22 (1.00, 1.44)
Frequency of hypoglycemia***	31.56†§ (28.84, 34.29)	38.33|| (34.52, 42.14)	39.53|| (35.46, 43.61)	33.77 (28.92, 38.63)
Frequency of severe  hypoglycemia (log)***	0.32†‡§ (0.24, 0.40)	0.57|| (0.46, 0.68)	0.48|| (0.36, 0.59)	0.50|| (0.36, 0.64)
Healthcare access/utilization
See physician**	1.36†‡ (1.29, 1.44)	1.59§|| (1.49, 1.69)	1.44† (1.33, 1.56)	1.55|| (1.42, 1.68)
See other healthcare professional***	0.37†‡§ (0.28, 0.47)	0.77|| (0.64, 0.90)	0.68|| (0.54, 0.82)	0.72|| (0.55, 0.89)
Diabetes education	1.00 (0.92, 1.09)	1.10 (0.98, 1.22)	1.05 (0.92, 1.17)	0.92 (0.77, 1.07)
Access barriers***	0.40†‡§ (0.33, 0.47)	0.55|| (0.45, 0.66)	0.68|| (0.58, 0.79)	0.68|| (0.55, 0.81)
Patient-centered care***	35.67†‡§ (32.87, 38.48)	55.49|| (51.62, 59.35)	52.03|| (47.85, 56.20)	52.46|| (47.42, 57.51)
Subjective burden of diabetes
Hypoglycemia worry***	2.22†‡§ (2.12, 2.32)	2.58‡|| (2.43, 2.72)	2.62‡|| (2.47, 2.77)	2.89†§|| (2.71, 3.07)
Weight worry	2.71 (2.60, 2.82)	2.71 (2.56, 2.87)	2.74 (2.58, 2.90)	2.75 (2.55, 2.94)
Dietary restrictions***	2.55†‡§ (2.45, 2.66)	2.92‡|| (2.77, 3.07)	3.08|| (2.92, 3.24)	3.21†|| (3.03, 3.40)
Diabetes-related discrimination	1.50†‡ (1.40, 1.59)	1.67|| (1.54, 1.81)	1.58 (1.45, 1.72)	1.70|| (1.54, 1.86)
Social support/burden
Diabetes support network***	1.11†§ (1.04, 1.90)	1.43‡|| (1.33, 1.54)	1.46‡|| (1.34, 1.58)	1.20†§ (1.06, 1.34)
Diabetes-related family arguments***	1.63†‡§ (1.52, 1.74)	1.94‡§|| (1.79, 2.09)	2.40†|| (2.24, 2.56)	2.29†|| (2.10, 2.48)
Diabetes support gap*	3.26†§ (3.11, 3.41)	3.00|| (2.79, 3.21)	2.86|| (2.64, 3.08)	3.08 (2.82, 3.35)

Metrics for all measures described in Table 1.

*Overall P < 0.05 for ethnicity from ANCOVA controlling diabetes type/treatment, diabetes duration, gender, age, income, education.

**Overall P < 0.01 for ethnicity from ANCOVA controlling diabetes type/treatment, diabetes duration, gender, age, income, education.

***Overall P < 0.001 for ethnicity from ANCOVA controlling diabetes type/treatment, diabetes duration, gender, age, income, education.

†Mean is significantly (P < 0.05) different from African American mean.

‡Mean is significantly (P < 0.05) different from Chinese American mean.

§Mean is significantly (P < 0.05) different from Hispanic mean.

||Mean is significantly (P < 0.05) different from White non-Hispanic mean.

Table 4. Adjusted means of psychological outcomes for ethnic groups.

Psychological outcome	White non-Hispanic n = 447	African American n = 241	Hispanic n = 194	Chinese American n = 173
Well-being (overall SD = 22.87)
Model 1*	57.12†	64.36||	60.89	59.96
Model 2**	58.81†‡§	66.48||	62.95||	63.84||
Quality of life (overall SD = 23.24)
Model 1**	65.11†	72.49‡§||	62.47†	63.98†
Model 2*	67.25†	72.42§||	65.95†	69.53
Diabetes impact (overall SD = 18.19)
Model 1**	57.20†§	48.95‡||	51.01||	53.85†
Model 2**	56.43†§	48.31||	49.85||	52.96
Diabetes distress (overall SD = 26.80)
Model 1**	22.92†‡§	32.33§||	37.51†||	36.98||
Model 2	27.13§	28.93	32.04||	26.82
Diabetes empowerment (overall SD = 22.52)
Model 1**	34.34†‡§	51.13‡§||	42.43†||	45.60†||
Model 2**	38.83†	47.58§||	39.23†	43.14

Model 1 is an ANCOVA controlling diabetes type/treatment, diabetes duration, gender, age, income, education. Model 2 is from ANCOVA controlling all factors reported in Table 5.

*Overall P < 0.01 for ethnicity.

**Overall P < 0.001 for ethnicity.

†Mean is significantly (P < 0.05) different from African American mean.

‡Mean is significantly (P < 0.05) different from Chinese American mean.

§Mean is significantly (P < 0.05) different from Hispanic mean.

||Mean is significantly (P < 0.05) different from White non-Hispanic mean.

Table 5. Regression analysis of psychological outcomes.

	Well-being		Quality of life		Diabetes impact		Diabetes distress		Diabetes empowerment
Predictor	Beta at entry	Final beta	Beta at entry	Final beta	Beta at entry	Final beta	Beta at entry	Final beta	Beta at entry	Final beta
Disease, demographic and socioeconomic factors
Diabetes duration	0.020	0.008	0.003	−0.012	−0.016	−0.027	−0.034	−0.039	0.013	0.019
Type 1 diabetes¶	0.018	0.046	−0.066	−0.021	−0.023	−0.007	0.122**	0.015	0.059	−0.071
Type 2 diabetes  (oral medication)¶	0.050	0.029	0.008	−0.019	−0.043	−0.025	−0.012	−0.005	0.010	0.028
Type 2 diabetes (insulin medication)¶	−0.083	−0.003	−0.085	−0.007	0.021	0.005	0.145***	0.005	0.082	−0.013
Female	−0.094***	−0.049*	0.011	0.033	0.019	−0.006	0.047	0.044	0.094***	0.108***
Age	0.247***	0.191***	0.078*	0.005	−0.183***	−0.141***	−0.258***	−0.161*	−0.156***	−0.096**
Income	0.161***	0.048	0.229***	0.108***	−0.078*	−0.008	−0.079*	0.000	0.077*	0.020
Education	0.065	0.014	0.074*	0.016	0.040	0.059	−0.107**	−0.039	0.058	0.088**
Change in r-squared	0.127		0.100		0.065		0.172		0.156
Health status
BMI	−0.022	0.016	−0.017	0.006	0.025	0.011	−0.040	−0.063*	−0.032	−0.028
Self-rated health	0.335***	0.276***	0.418***	0.374***	−0.312***	−0.271***	−0.208***	−0.130***	0.165***	0.110***
Health index	0.250***	0.240***	0.191***	0.189***	−0.051	−0.043	−0.179***	−0.158***	−0.049	−0.034
Complications	−0.034	−0.026	−0.024	−0.030	0.018	0.022	0.125***	0.079**	0.035	0.011
Frequency of hypoglycemia	−0.052	−0.037	−0.017	0.005	0.069*	0.068*	0.106***	0.025	0.108**	0.062*
Frequency of severe  hypoglycemia	−0.070*	−0.035	−0.123***	−0.095***	−0.036	−0.040	0.079**	0.008	0.131***	0.102***
Change in r-squared	0.271		0.309		0.117		0.182		0.053
Healthcare access/utilization
See physician	0.056	0.056*	0.098***	0.098***	−0.081*	−0.079*	0.026	0.007	0.043	0.024
See other healthcare  professional	−0.056	−0.070*	−0.002	0.002	−0.029	−0.013	0.054	0.031	0.019	−0.004
Diabetes education	0.087***	0.080***	0.006	−0.002	0.006	0.010	0.020	0.022	0.099***	0.090***
Access barriers	0.013	0.029	−0.032	−0.022	0.018	0.001	0.140***	0.089***	0.072**	0.046
Patient-centered care	0.061*	−0.017	0.048	0.010	−0.063*	−0.010	0.045	0.063*	0.335***	0.248***
Change in r-squared	0.015		0.013		0.014		0.026		0.133
Subjective burden of diabetes
Hypoglycemia worry	−0.068*	−0.089**	−0.082**	−0.080**	−0.037	−0.015	0.157***	0.150***	0.086**	0.061*
Weight worry	−0.111***	−0.108***	−0.060*	−0.052	0.020	0.021	0.122***	0.113***	0.061*	0.063*
Dietary restrictions	−0.080**	−0.082**	−0.011	−0.002	0.092**	0.101**	0.175***	0.158***	0.043	0.036
Diabetes-related discrimination	0.053*	0.047	0.015	0.027	0.024	0.037	0.067**	0.045	0.074**	0.063*
Change in r-squared	0.028		0.010		0.009		0.098		0.023
Social support/burden
Diabetes support network	0.076**	0.076**	0.078**	0.078**	−0.040	−0.040	0.017	0.017	0.093***	0.093***
Diabetes-related family arguments	0.074**	0.074**	−0.046	−0.046	−0.102**	−0.102**	0.118***	0.118***	0.103***	0.103***
Diabetes support gap	−0.147***	−0.147***	−0.058*	−0.058*	0.108**	0.108**	0.086**	0.086**	−0.139***	−0.139***
Change in r-squared	0.031		0.011		0.020		0.014		0.037
Final r-squared	0.472		0.443		0.225		0.492		0.402

Cell entries are standardized regression coefficients (except as noted). ‘At entry’ column reports coefficients for all variables in a block after all variables from that block and preceding blocks were entered into the model (i.e., the total unconfounded association). ‘Final’ column reports coefficients for all variables in the model with all variables entered into the model (i.e., the unconfounded and unmediated association). R-squared is that for the model when preceding variables have been entered. Coefficients not shown for ethnic groups; ethnic group disparities from these models are reported in Table 4.

¶Reference group is type 2 diabetes, no diabetes medication.

*P < 0.05; **P < 0.01; ***P < 0.001.

Szende A, Oppe M, Devlin N, eds. EQ-5D value sets: Inventory, Comparative Review and User Guide. EuroQol Group Monographs. Dordrecht, the Netherlands: Springer, 2007

 Google Scholar

Nicolucci A, Kovacs Burns K, Holt RI, et al.; DAWN2 Study Group. Diabetes Attitudes, Wishes and Needs second study (DAWN2™): cross-national benchmarking of diabetes-related psychosocial outcomes for people with diabetes. Diabet Med 2013;30:767-77

 PubMed Web of Science ®Google Scholar

Peyrot M, Skovlund S, Landgraf R. Epidemiology and correlates of weight worry in the multi-national Diabetes Attitudes, Wishes and Needs (DAWN) study. Curr Med Res Opin 2009;25:1985-93

 PubMed Web of Science ®Google Scholar

Bech P, Gudex C, Johansen, KS. The WHO (Ten) Well-Being Index: validation in diabetes. Psychother Psychosom 1996;65:183-90

 PubMed Web of Science ®Google Scholar

The WHOQOL Group. Development of the World Health Organization WHOQOL-BREF quality of life assessment. Psychol Med 1998;28:551-8

 PubMed Web of Science ®Google Scholar

McGuire BE, Morrison TG, Hermanns N, et al. Short-form measures of diabetes-related emotional distress: the Problem Areas in Diabetes Scale (PAID)-5 and PAID-1. Diabetologia 2010;53:66-9

 PubMed Web of Science ®Google Scholar