Skip to main content

Table 1 Characteristics of included studies

From: Effectiveness of non-specialist delivered psychological interventions on glycemic control and mental health problems in individuals with type 2 diabetes: a systematic review and meta-analysis

Study, design and country

Mean age of participants in years (SD), % of males and females

Sample size, N of conditions, control group

Duration of diabetes [mean years (SD)]

Cadre of non-specialist (mode of delivery)

Intervention description (follow-up)

Outcomes measures of relevance [mean baseline scores (SD)]

Results

Chiu et al. [35] RCT, Taiwan

64.6 (8.9) 51.7% males and 48.3% females

N = 174. Two conditions, Usual care

10.5 (8.3)

Research assistants (Telephone)

Focused on T2DM individuals aged 50 years and above with occasional distress or minor depressive symptoms. Individuals received 3–4 CBT sessions consisting of reattribution technique over the phone lasting 30–60 min for 6 weeks. (Follow up at 1-month for all outcome measures plus 3- and 8-month follow up for HbA1c)

Diabetes distress Intervention: 5.6 (7.6) Control: 5.4 Depression Intervention: 3.2 Control: 3.7 (4.8) HbA1c (%) Intervention: 7.6 Control: 7.7 (1.3)

Diabetes distress: No significant difference between intervention and control group at the end of intervention and 1 month postintervention. Depression: No significant difference between intervention and control group at the end of intervention and 1 month postintervention. HbA1c (%): No significant difference in reduction of HbA1c levels between intervention and control group at the end of the intervention and at follow-up periods. Participants in intervention with HbA1c < 8% significantly reduced HbA1c levels compared to control group 3 months post intervention. Reduction sustained at 8-months follow-up

Dale et al. [36] RCT, UK

55.4% males and 44.6% females

N = 231. Three conditions, Usual care

–

Peers and Diabetes nurses (Telephone)

Targeted at individuals with raised HbA1c levels. It consisted of 6 individual motivational interviewing sessions for 6 months. (No follow-up)

Diabetes distress Intervention 1: 14.6 (12.7) Intervention 2: 22.7 (18.8) Control: 19.8 (15.5) HbA1c Intervention 1: 8.4 (1.1) Intervention 2: 8.9 (1.5) Control: 8.7 (1.3)

Diabetes distress: No statistically significant difference between the three groups. HbA1c (%): No statistically siginificant difference in HbA1c between the three groups

Dobler et al. [37] RCT, Germany

52 (5.5) 70% males and 30% females

N = 199. Two conditions, Usual care

Intervention: 8.7 (6.6) Control: 9.6 (5.9)

Non-medical dietitians (Telephone)

Targeted at individuals who are German speakers. It consisted of 12 individual motivational interview sessions and problem-solving therapy for 12 months. (No follow-up)

Diabetes distress Intervention: 33.2 (19.5) Control: 38.0 (17.5) Depression Intervention: 9.2 (5.2) Control: 10.1 (6.3) HbA1c (%) Intervention: 7.8 (1.7) Control: 7.6 (1.4)

Diabetes distress: Decreases in PAID scores for diabetes distress post intervention in intervention group by 4.77% and control group by 1.4% were not significant. Depression: Significant decreases in PHQ scores in intervention group compared to control group postintervention. HbA1c (%): HbA1c scores decreased significantly by 0.68 in intervention group and increased by 0.12 in control group

Fisher et al. [38] RCT, USA

56.11 (9.55) 46.2% males and 53.8% females

N = 392. Three conditions, Diabetes education

6.90 (5.93)

Non-professional college graduates (Web-based, Telephone, In person)

Aimed at distressed, non-clinically depressed T2DM individuals and lasted for 12 months. CASM was a web-based self management intervention for 40 min. Participants in CASM received telephone calls at 8 different time intervals for 15 min. CAPS was an in-person PST combined with CASM for 60 min. Participants in CAPS received a review of PST steps (booster session) at month 5. (Follow-up at 4-months and 12-months)

Diabetes distress CASM: 2.37 (0.86) CAPS: 2.38 (0.89) LA: 2.48 (0.95) HbA1c (%) CASM: 7.45 (1.5) CAPS: 7.34 (1.6) LA: 7.45 (1.7)

Diabetes distress: Significant decrease observed in the 3 groups (CASM, CAPS and LA) and was maintained at follow-up, as participants in CAPS reported significantly greater reduction in diabetes distress. HbA1c (%): For the 3 groups (CASM, CAPS and LA), there was no significant reduction at posttreatment and at follow-up

Gabbay et al. [39] RCT, USA

Intervention: 58 (11.41) Control: 58 (11.34) 42% males and 58% females

N = 545. Two conditions, Usual care

–

Nurses (In person, Telephone, E-mail)

Targeted at T2DM individuals with high risk of developing cardiovascular complications. It consisted of 2–9 Individual motivational interview sessions for 1 h for 24 months. Telephone and e-mail was used in between face to face visits. (No follow-up)

Diabetes distress Intervention: 29 (22.64) Control: 29 (24.32) Depression Intervention: 14 (14.76) Control: 15 (15.24) HbA1c (%) Intervention: 8.82 (2.38) Control: 9.05 (2.27)

Diabetes distress: No significant difference between participants in the intervention group and control group at the end of the intervention. Depression: Significant reduction in CES-D scores for depression in participants in the intervention group compared to the control group at the end of the intervention. HbA1c (%): No significant difference in HbA1c decline between both groups as HbA1c reduced in intervention group and in control group

Heinrich et al. [40] RCT, Netherlands

59 (5.27) 55.1% males and 44.9% females

N = 584. Two conditions, Usual care

26.4% had diabetes for less than 1 year 47% had diabetes between 2 and 3 years 26.4% had diabetes between 4 and 5 years

Nurses (In person, web-based)

Aimed at T2DM individuals between 40 and 70 years with diabetes duration of less than 5 years. Individuals received 3 face to face MI sessions for 20 min in conjuction with a web-based educational programme within 12 months. (Follow-up at 1-year and 2 years)

Diabetes distress Intervention: 16.83 (13.32) Control: 16.98 (13.92) HbA1c (%) Intervention: 6.49 (0.85) Control: 6.51 (0.74)

Diabetes distress: There was no significant difference between the intervention and control group at the end of the intervention and at 1 and 2-year follow up. HbA1c (%): There was no significant difference between the intervention and control group

Inouye et al. [41] RCT, USA

57.3 (10.9) 45.4% males and 54.6% females

N = 207. Two conditions, Diabetes education

–

Research assistants (In Person)

Aimed at T2DM individuals. The group CBT sessions consisted of biofeedback assisted relaxation, cognitive restructuring, problem solving, contracting, behavioral rehearsal and reinforcements. The group met for six successive weeks for 1–2 h and group size ranged between 2 to 6 individuals. (Follow-up at 12 months)

Depressionb Intervention: 10.83 ± 0.83 Control: 9.68 ± 0.83 HbA1c (%)b Intervention: 8.1 ± 0.2 Control: 7.8 ± 0.2

Depression: Intervention group had improved in CES-D depressive symptom scores compared to the control group at the end of the study. The improvements were not sustained at follow-up HbA1c (%): Differences observed in the intervention group and control group at post-intervention were not statistically significant. However, the scores increased at follow up for intervention and control group

Kim et al. [42] RCT, USA

56.9% males and 43.1% females

N = 250. Two conditions, Waitlist

8.5

Nurses and Community health workers (In Person, Telephone)

Targeted at Korean Americans with poor glycemic control. It consisted of group and individual sessions. The group sessions involved 2-h sessions of CBT techniques (problem solving and cognitive reframing) and diabetes education over the course of 6 weeks. The individual sessions involved 11 sessions of motivational interviewing ranging from 15–45 min. (No follow-up)

Depressionb Intervention: 5.3 (0.5) Control: 5.4 (0.5) HbA1cb Intervention: 8.9 (0.2) Control: 8.8 (0.2)

Depression: Control group experienced significant improvement in depressive symptom scores compared to the intervention group at the end of the study. HbA1c (%): Statistically signifiacnt difference in HbA1c levels between both groups, with clinically significant reductions observed in the intervention group than in the control group

Lamers et al. [43] RCT, Netherlands

Intervention: 70.7 (6.6) Control: 69.7 (6.6) 49% males and 51% females

N = 208. Two conditions, Usual care

Intervention: 8.2 (8.8) Control: 9.8 (9.1)

Nurses (In Person)

It was aimed at T2DM individuals aged 60 years and above with minor to moderate depression and consisted of individual cognitive behaviour therapy elements combined with self-management techniques for 3 months. Participants received 2–10 visits lasting 60 min–90 min over a period of 3 months. (Follow-up at 1-week, 3-months and 9-months)

Diabetes distress Intervention: 22.6 (20.5) Control: 23.4 (19.5) HbA1c (%) Intervention: 7.5 (1.2) Control: 7.2 (1.4)

Diabetes distress: Non-significant difference observed in PAID scores for both groups postintervention at 1 week, 3 months and 9 months. HbA1c (%): Non-significant differences observed between both groups postintervention at 1 week and 3 months. Significant improvement between the two groups at 9 months as participants in the intervention group had reduced HbA1c level (7.3%) when compared with increased HbA1c level in the control group (7.8%)

Sacco et al. [44] RCT, USA

52 (8.6) 42% males and 58% females

N = 62. Two conditions, Usual care

9.5 (7.2)

Undergraduate students (Telephone)

Focused on T2DM individuals with poor glycemic control. The individual CBT sessions consisted of activity rescheduling and behavioral experiments for 6 months. Participants received one phone call per week for the first 3 months and one biweekly call for the remaining 3 months. Phone calls lasted 15–20 min. (No follow-up)

Depression Intervention: 16.32 (6.60) Control: 16.45 (6.77) HbA1c (%) Intervention: 8.4 (1.37) Control: 8.5 (2.01)

Depression: Significant reduction in PHQ depressive symptom scores in participants in the intervention group compared with those in the control. HbA1c (%): No significant difference between intervention and control group as both groups reported reduced HbA1c levels

Simmons et al. [45] cRCT, UK

Group support: 65.2 (10.2) Individual support: 65.2 (8.9) Combined group and individual support: 65.3 (9.3) Control: 64.6 (10.3), 59.3% 60.4% males and 39.6% females

N = 1299. Four conditions, Usual care

Group support: 7.0 (3.0–12.0)a Individual support: 7.0 (3.0–12.0)a Combined group and individual support: 6.0 (3.0–11.0)a Control group: 6.5 (3.0–12.0)a

Peers (In person, Telephone)

Focused on participants with T2DM for at least 12 months. It consisted of motivational interviewing techniques for 8- 12 months received in group, individual or combined group and individual sessions. The duration of the individual and group sessions were 60 min and 90 min respectively. (No follow-up)

Diabetes distress Group support: 6.27 (4.22) Individual support: 6.53 (4.12) Combined group and individual support: 6.71 (4.27) Control: 6.61 (4.05) Depression Group support: 4.49 (4.92) Individual support: 4.39 (5.13) Combined group and individual support: 4.59 (4.60) Control: 4.49 (5.01) HbA1c (%) Group support: 7.5 (1.3) Individual support: 7.4 (1.3) Combined group and individual support: 7.3 (1.3) Control: 7.3 (1.3)

Diabetes distress: Participants in the individual support group had a significant reduction in diabetes distress compared to participants in the group interventions Depression: No significant difference between participants in the individual support intervention and group interventions with respect to reducing depressive symptoms. HbA1c (%): No significant diffeence between participants in the individual support intervention and group support intervention. In participants with HbA1c above 8%, no significant difference between the individual and group support intervention with respect to reducing HbA1c levels

Spencer et al. [46] RCT, USA

Intervention: 50 (47, 52)c Delayed group: 55 (53, 57)c 29% males and 71% females

N = 164. Two conditions, Waitlist

Intervention: 8 (6, 9)c Delayed group: 9 (7, 11)c

Community Health Workers (In person, Telephone)

Aimed at African American and Latino individuals with T2DM. It consisted of combination of diabetes education classes and motivational interviewing sessions. 11 group sessions of motivational interviewing and diabetes education lasting for 2 h were delivered every 2 weeks. The duration of the intervention was 6 months. (Follow-up at 6-months for intervention group)

Diabetes distress Intervention: 23.8 (18.7, 29.0)c Delayed group: 25.9 (21.2, 30.6)c Depression Intervention: 5.2 (3.9. 6.5)c Delayed group: 5.0 (4.0, 5.9)c HbA1c (%) Intervention: 8.6 (8.0, 9.1)c Delayed group: 8.5 (8.0, 8.9)c

Diabetes distress: Significant greater reduction in PAID scores in participants in the intervention group compared to the control control at end of the intervention. Depression: No significant changes in PHQ depressive symptom scores between both groups. HbA1c (%): Participants in the intervention group improved significantly in mean change in HbA1c values compared with the delayed group

Wagner et al. [47] RCT, USA

Intervention: 60.0 (11.2) Control: 60.8 (12.1) 27.1% males and 72.9% females

N = 107. Two conditions, Diabetes education

–

Community Health Worker (In person)

Targeted at individuals with T2DM diagnosis of 6 months or more than and glycemic level above 7.0% in the past one year. It consisted of combination of techniques of CBT and mindfulness therapy in addition to diabetes education. The intervention comprised of 8 groups sessions for 2 h over 8–10 weeks. Participants in the group ranged from 9 to 16. (Follow-up at 3-months)

Diabetes distress Intervention: 7.9 (6.7) Control: 8.1 (6.3) Depression Intervention: 6.7 (5.9) Control: 5.3 (4.4) HbA1c (%) Intervention: 8.5 (1.4) Control: 8.6 (1.9)

Diabetes distress: No significant difference between intervention and control group with diabetes distress decreasing in both groups. Depression: Significant reduction in depressive symptoms in participants in the intervention group compared with the increase in depressive symptoms observed in the control group. HbA1c (%): No significant difference between intervention and control group postintervention and 3 months follow-up

Welch et al. [48] RCT, USA

55.7 (10.2) 59.1% and 40.9%

N = 234. Four conditions, Diabetes education

8.2 (7.0)

Diabetes educators

Aimed at diabetic individuals between the ages of 30 to 70 years with poorly controlled diabetes above 7.5%. Individuals received four sessions of MI plus diabetes education for 6 months. The first session was one hour and the remaining sessions were for 30 min. (No follow-up)

Diabetes distress Intervention 1: 40.5 (23.3) Intervention 2: 41.9 (22.4) Control 1: 43.4 (25.0) Control 2: 42.5 (23.6)

Depression

Intervention 1: 19.1 (9.0) Intervention 2: 18.9 (8.7) Control 1: 19.9 (9.3) Control 2: 18.6 (10.9) HbA1c (%) Intervention 1: 9.1 (1.5) Intervention 2: 8.8 (1.0) Control 1: 8.8 (1.3) Control 2: 8.9 (1.62)

Diabetes distress: No significant difference between intervention and control groups

Depression: No significant difference between intervention and control groups. HbA1c (%): Participants in intervention had a mean change in HbA1c that was significantly lower than control group

Welschen et al. [49] RCT, Netherlands

Intervention: 60.5 (9.4) Control: 61.2 (8.8) 61.7% males and 38.3% females

N = 154. Two conditions, Usual care

Intervention: 7.6 (5.0) Control: 7.8 (6.1)

Dietitians and Diabetes nurses (In person)

Targeted at T2DM individuals with glycemic level higher than 7%. It consisted of 3–6 individual CBT sessions of 30 min. (Follow-up at 6-months and 12-months)

Depression Intervention: 11.1 (8.1) Control: 9.6 (8.2) HbA1c (%) Intervention: 6.8(1.0) Control: 6.7 (1.0)

Depression: Participants in the intervention group reported significant decrease in depressive symptoms compared with the control group. This was not sustained 6 months later. HbA1c (%): No significant difference between intervention and control group at 6 months follow-up and 12 months follow-up

Whittemore et al. [50] RCT, USA

57.6 (10.9) Participants were females

N = 53. Two conditions, Usual care

2.7 (3.0)

Nurse (In person, telephone)

Aimed at T2DM women with glycemic level higher than 7%. It consisted of 6 individual sessions of MI in addition to diabetes education for 1 h lasting for 6 months. Two telephone calls were provided between session 5 and 6. (No follow-up)

Diabetes distress Intervention: 59.9 (22) Control: 42.3 (14) HbA1c (%) Intervention: 7.7 (1.0) Control: 7.6 (1.0)

Diabetes distress: Participants in the intervention group significant reported decrease in diabetes distress compared with the reported increase in the control group. HbA1c (%): HbA1c levels decreased in both groups postintervention. The differences between the groups were not significant

  1. aDuration of diabetes (IQR)
  2. bMean baseline scores ± standard error
  3. cmean (95% CI)
  4. CASM, Computer-Assisted Self-Management; CAPS, Computer-Assisted self-management with Problem Solving therapy; CES-D, Center for Epidemiological Studies-Depression scale; CBT, Cognitive Behavior Therapy; cRCT, Cluster Randomized Controlled Trial; HbA1c, Glycated Hemoglobin; LA, Leap Ahead; MI, Motivational Interviewing; PAID, Problem Areas in Diabetes; PST, Problem Solving Therapy; RCT, Randomized Controlled Trial; SD, Standard Deviation; T2DM, Type 2 Diabetes Mellitus