Skip Navigation U.S. Department of Health and Human Services www.hhs.gov
Agency for Healthcare Research Quality www.ahrq.gov
Archive print banner
Use of Behavioral Therapies for Treatment of Medical Disorders

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: https://info.ahrq.gov. Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to www.ahrq.gov for current information.

Part 1. Impact on Management of Patients with Diabetes Mellitus: Appendix D (continued)


Evidence Tables

Study Selected Inclusion/ Exclusion Criteria Study Design Patients Interventions Treatment Duration Outcomes/Results Comments

#3400

Campbell, Barth, Gosper, Jupp, Simons, & Chisolm, 1990.

Include: Randomized effects of an intensive educational approach to dietary change in NIDDM.

RCT with 2 treatment conditions:
1) Conventional Intervention (con).
2) Intensive intervention (int).

N=70
n con=29
n del=33

*8 subjects dropped out.

Age mean(SD):
con=59(9)
del=58(9)

% Female:
con=41.4
int=45.5

Race % not given

Baseline Fasting Blood Glucose (mM) means (SD):
con: 8.9 (2.1)
del: 9.7 (2.8)

1) Conventional Program—covered topics of explaining diabetes, diabetes complications, and diet, exercise, and food composition.

2) Intensive program— included longer, more in-depth sessions on diet, podiatry, cognitive-motivation components, unconscious and conscious mental processes affecting the desire to achieve goals or take action. Participants established adequate reasons for behavioral change. Subjects were asked to visualize the adverse effects of diabetic complications.

Convention—3 consecutive days.

Intensive—11 weeks (total 22 hrs)

Both had 1 month and 3 month followup

1) Metabolic control
 -Fasting Blood Glucose (mM) means (SD):
   con: 8.9 (2.1) base
     9.2 (3.4) 1 mo
     9.5 (3.4) 3 mo
     8.3 (2.7) 6 mo
   del: 9.7 (2.8) base
     9.4 (2.7) 1 mo
     9.1 (3.0) 3 mo
     9.6 (2.9) 6 mo

*RM-ANCOVA indicated no significant differences in fasting blood glucose between groups over time (=0.7).

2) Measures of risk:
 a) Body Mass Index—BMI means (SD):
   con: 32.0 (5.5) base
     31.5 (5.6) 1 mo
     31.2 (5.4) 3 mo
     31.1 (5.1) 6 mo
   del: 30.4 (4.8) base
     29.5 (4.7) 1 mo
     29.6 (4.5) 3 mo
     29.6 (4.6) 6 mo

*RM-ANCOVA indicated no significant differences between groups over time (p=0.28).

 b) Total Cholesterol means (SD):
   con: 6.5 (1.1) base
     6.5 (1.4) 1 mo
     6.3 (1.2) 3 mo
     6.5 (1.0) 6 mo
   del: 7.4 (1.2) base
     6.6 (1.1) 1 mo
     6.8 (1.1) 3 mo
     6.6 (1.0) 6 mo

*RM-ANCOVA indicated a significant difference between groups over time (p=0.007).

 c) HDL-Cholesterol means (SD):
   con: 1.2 (0.2) base
     1.1 (0.2) 1 mo
     1.2 (0.2) 3 mo
     1.1 (0.2) 6 mo
   del: 1.1 (0.2) base
     1.1 (0.2) 1 mo
     1.2 (0.2) 3 mo
     1.1 (0.3) 6 mo

*RM-ANCOVA indicated no significant differences between groups over time (p=0.27).

3) Events:
 a) Health care utilization: Not given
 b) Morbidity/mortality: Not given

Quality Assessment:

Internal Validity:

Described as randomized? Yes

Method of randomization clearly described? No

Concealment of allocation? No

Described as double-blind? No

Patient blinded? No

Investigators blinded? No

Outcome assessors blinded? No

No. of withdrawals in each group stated? Yes

External Validity:

Pop. Described? Yes

Intervention described well enough to reproduce? Yes

Intervention codified in manual? No

Provider training described? No

Patients assessed for DSM dx? No

Biases, etc: None noted

#6360

Cox, Gonder-Frederick, Julian, Cryer, Herrman, Richards & Clarke, 1991

Include: IDDM=2 years since diagnosis; insulin usage since diagnosis; using self-measurement of blood glucose

Exclude: history of heart disease, hypertension, seizure activity, or severe psychiatric disturbance; chronic medication other than insulin

RCT with 3 treatment conditions:
1) Control (con).
2) Standard BGAT (sta).
3) Intensive BGAT (int).

N=39
n con=14
n sta=13
n int=12

*withdrawals not stated

Age means: Intended to treat:
con=33.8
sta=33.7
int=31.1

% Female: Intended to treat:
con=57.1
sta=61.5
int=66.7

Race % not given

Baseline HbA1 means:
Intended to treat:
Con=11.4
Sta=10.4
Int=12.8

1) Standard BGAT—7 weekly classes with readings and BGAT with readings and homework exercises having to do with BGAT manual, BG symptoms, how insulin, food, and exercise effects BG. Daily systematic recordings of internal and external cues of BG.

2) Intensive BGAT—during hospitalization, subjects were provided with immediate BG feedback while hyper and hypoglycemic. At these times, subjects described their experiences on audio tape, rated perceived symptoms on a checklist, estimated BG level and then were told actual BG level. Patients were later given the audio tape and were allowed to recall how they felt when hyper- and hypoglycemic.

3) Placebo control group also attended group meetings and kept diaries recording daily stress factors and diabetic self-care behaviors.

7-week intervention following hospitalizations

Completer Results:

1) Metabolic control:
 - HbA1 % means (SD):
   con: 11.1 (2.2) base
     11.7 (2.6) post
     11.3 (2.6) f/u
   sta: 10.5 (2.4) base
     10.6 (2.6) post
     10.1 (2.4) f/u
   int: 12.8 (4.1) base
     12.1 (3.6) post
     10.3 (2.7) f/u

*ANOVA indicated int significantly different from con (p<0.02)

2) Measures of risk: Not given

3) Events:
 a) Health care utilization: Not given
 b) Morbidity/mortality: Not given

Quality Assessment:

Internal Validity:

Described as randomized? Yes

Method of randomization clearly described? No

Concealment of allocation? No

Described as double-blind? No

Patient blinded? No

Investigators blinded? No

Outcome assessors blinded? No

No. of withdrawals in each group stated? No

External Validity:

Pop. Described? Yes

Intervention described well enough to reproduce? Yes

Intervention codified in manual? Yes

Provider training described? No

Patients assessed for DSM dx? No

Biases, etc: None noted

#6250

D'Eramo-Melkus, Wylie-Rosett, Hagen, 1991

Include: NIDDM, 21-65 years old, 20-75% over desirable body weight.

Exclude: Insulin dependence, or serious illness.

RCT with 3 treatment conditions:
1) Single individual session (con).
2) 12-wk behavior oriented diabetes education and weight control group intervention (int.).
3) Group intervention plus six individual followup sessions (int + fu).

N=82
n con=28
n int=28
n int+fu=26

*33 drop-outs (13 In control, 13 in int, 7 in int+fu)

Age mean (SD): 55.6 (8.05)

% Female: 58.5

Race % not given

Baseline HbA1 % means (SD):
Completers:
con=10.91(2.6)
int=10.72(3.16)
int+fu=11.15(2.92)

1) All participants received minimal skills educational intervention, including food measurement, setting weight and calorie goals, self-monitoring blood glucose, and foot care.

2) Intervention—11 wk group intervention of 2 hr session consisting of lecture and slide presentation on general diabetes principals and skills and nutrition principals. Goals for changing eating behavior, increasing physical activity, and blood glucose control were set. Societal pressures, internal resistance to change and lack of self-reinforcement.

3) Intervention + Followup counseling—participants received intervention plus 2 followup sessions.

11 weeks, 12 and 18 week followup

Completer Results:

1) Metabolic control:
 a) HbA1 % means (SD):
   con: 10.91 (2.6) base
     10.54 (3.11) 3 mo
     10.5 (3.21) 6 mo
   int: 10.72 (3.16) base
     8.58 (2.55) 3 mo
     9.17 (3.3) 6 mo
   int+fu: 11.15 (2.9) base
     8.82 (2.8) 3 mo
     8.26 (2.7) 6 mo

*RM-ANOVA indicated a significant decrease in HbA1 for int (p<0.05) and int+fu (p<0.01) at 3 mo on HbA1c over time. Between groups not reported.

 b) Fasting Blood Glucose (mM) means (SD):
   con: 11.34 (3.29) base
     10.31 (4.05) 3 mo
     12.18 (5.46) 6 mo
   int: 11.59 (3.67) base
     8.83 (2.68) 3 mo
     9.45 (3.61) 6 mo
   int+fu: 12.21 (3.85) base
     10.08 (4.66) 3 mo
     9.03 (3.0) 6 mo

*RM-ANOVA indicated a significant decrease in fasting blood glucose for int and int+fu at 3 and 6 mo on HbA1c over time (p<.05 for all). Between groups not reported.

2) Measures of risk:
 a) Weight (lbs) means (SD):
   con: 215.25 (25.47) base
     209.46 (25.14) 3 mo
     205.14 (25.59) 6 mo
   int: 211.84 (27.78) base
     199.96 (30.13) 3 mo
     200.72 (30.44) 6 mo
   int+fu: 200.65 (30.7) base
     192.42 (32.09) 3 mo
     191.8 (31.73) 6 mo

*RM-ANOVA indicated a significant decrease in weight for all groups at 3 mo (p<0.05 for all)

 b) Cholesterol means (SD):
   con: 5.75 (1.19) base
     5.83 (1.23) 3 mo
     5.77 (1.61) 6 mo
   int: 6.19 (0.9) base
     5.58 (0.72) 3 mo
     5.71 (1.14) 6 mo
   int+fu: 6.08 (1.82) base
     5.48 (1.63) 3 mo
     5.57 (0.84) 6 mo

*RM-ANOVA indicated a significant decrease in weight for all groups at 3 mo (p<0.05 for all)

3) Events:
 a) Health care utilization: Not given
 b) Morbidity/mortality: Not given

Quality Assessment:

Internal Validity:

Described as randomized? Yes

Method of randomization clearly described? Yes

Concealment of allocation? No

Described as double-blind? No

Patient blinded? No

Investigators blinded? No

Outcome assessors blinded? No

No. of withdrawals in each group stated? Yes

External Validity:

Pop. Described? Yes

Intervention described well enough to reproduce? Yes

Intervention codified in manual? No

Provider training described? No

Patients assessed for DSM dx? No

Biases, etc: Large number of participants did not complete study

#20

Didjurgeit, Kruse, Schmitz, et al, 2002

Include: Type I diabetes, presence of self-reported persistent psychological problem, presence of at least one microvascular diabetic complications.

Exclude: not given

Randomized wait-list controlled trial for patients indicating psychological problems—single-center design

N=46
n con=21
n tx=23

*2 patients died during study-con: 1, tx: 1

Age means (SD):
con=41(10)
tx=36 (9)

61% Female

Race % not given

Baseline HbA1c means (SD):
Completers:
tx: 9.1(2.0)
con: 8.7 (1.7)

Psycho-therapeutic intervention:
1) Definition of the patient-therapist relationship.
2) Detailed description of a problematic situation of the patient.
3) Analysis of components of the problem.
4) Definition of the problem.
5) Handling the problem.
6) Conclusion of therapy.

Plus: Diverse psychotherapeutic interventions to foster awareness, modify thoughts, modify behavior, emotionality, awareness of body's ability to rely and support.

*all patients treated by one therapist

Weekly sessions—14 session maximum, 55-min sessions

Completer Results:

1) Metabolic control:
 -HbA1c mean (SD):
   total con: 8.7 (1.7) base
     8.8 (1.9) f/u
   tx: 9.1 (2) base
     8.5 (1.6) f/u

*ANOVA indicated a significant effect of group on f/u HbA1c (p=0.016)

2) Measures of risk: Not given

3) Events:
 a) Health care utilization: Not given
 b) Morbidity/mortality: Deaths=2; con: 1, tx: 1

4) Psychological Measures:
 a) Top 3 patient-indicated problems & severity (10- point scale) means (SD):
   con: #1: 8.3 (1.72) base
     6.8 (3) f/u
   #2: 7.61(1.79) base
     5.83(2.75) f/u
   #3: 7.36(2.65) base
     6.79(2.42) f/u
   tx: #1: 7.78(1.98) base
     4.3 (2.87) f/u
   #2: 7.67(2.31) base
     3.86(2.41) f/u
   #3: 7.71(2.33)
     4.71(2.43) f/u

*ANOVA indicated a significant effect of group on problem severity

 b) Severity of psychological distress symptoms related to disease (Symptoms Checklist 90 Revised-SCL-90Rc) means (SD):

   con: 0.99(0.47) base
     0.75(0.49) f/u
   tx: 1.1(0.71) base
     .93 (0.81) f/u

*ANOVA indicated no significant group by time interaction for SCL-90 (p=.49)

 c) Depression Score (ZERSSENc) means (SD):

   con: 13.8(8.9) base
     11.7(9.8) f/u
   tx: 16.3(9.6) base
     11.8(10.9) f/u

*ANOVA indicated no significant group by time interaction for ZERSSEN (p=.39).

 d) Quality of Life (IRESc) means (SD):

   con: 4.7(2) base
     4.3(1.6) f/u
   tx: 4(2.2) base
     4.4(1.7) f/u

*ANOVA indicated no significant group by time interaction for IRES (p=.21)

Quality Assessment:

Internal Validity:

Described as randomized: Yes

Method of randomization clearly described? No

Concealment of allocation? No

Described as double-blind? No

Patient blinded? No

Investigators blinded? No

Outcome assessors blinded? No

No. withdrawals stated? Yes

External Validity:

Pop. Described? Yes

Intervention described well enough to reproduce? Yes

Intervention codified in manual? Not sure

Provider training described? No

Biases, etc: 4 participants in intervention group did not complete therapeutic sessions, yet still completed f/u; Investigators note that therapy not easily replicated since not strictly structured; No objective measures of self-care used; Investigators note that no distinction made between how closely tied "problems" were to disease.

#5140

Dyson, Hammersley, Morris, Holman & Turner, 1997

Include: Patients with increased fasting glucose (5.5 to 7.7 mmol/L on 2 occasions.

Exclude: diabetes diagnosis

RCT with 2 x 2 factorial design.

Four conditions:
1) Sulfonylurea + reinforced healthy-living advice (S+RA).
2) Sulfonylurea + basic healthy-living advice (S+BA).
3) Control(placebo/no tablets) + reinforced healthy-living advice (con+RA).
4) Control(placebo/no tablets) + basic healthy-living advice (con+BA).

*groups 1 and 3 considered treatment (tx) and 2 and 4 considered control (con)

N=227
n S+RA=56
n S+BA=56
n con+RA=55
n con+BA=60

*26 drop-outs by 1-yr. f/u. (18 in RA, 8 in BA)

Age mean (SD): 50(9)

59% Female

Race % not given

Baseline HbA1c % mean:
Completers: 5.7

1) Sulfonylurea—an antihyperglycemic—helps body better respond to insulin and reduces the amount of sugar produced by liver.

2) Basic healthy-living advice— given written dietary information and seen by a physician who advised weight loss and increased physical activity. Patients seen every 3 months for assessment of glycemia, but basic advice was only given once at the initial visit.

3) Reinforced healthy-living advice—patients seen by dietitian and advised to change their diet, limit fat intake and increase consumption of unrefined carbs and dietary fiber. Individual energy requirements were calculated and caloric consumption. Saw a fitness Instructor every 3 months and were encouraged to increase physical activity gradually. Subjects filled out food and exercise diaries.

4) Placebo—half of the control group received a placebo tablet, the other half received no tablets.

3 months, 1 yr followup

Completer Results:

1) Metabolic control
 - HbA1c %
   RA: 5.7 base
     5.6 1 year
   BA: 5.7 base
     5.6 1 year

*Reported no significant effect of group on HbA1c. Statistical test not given. No change in findings when medicated Ss eliminated from analysis.

2) Measures of risk:
 a) Weight (kg)
   RA: 81.3 base
     80.8 1 year
   BA: 82.0 base
     81.8 1 year

*Reported no significant effect of group on weight loss. Statistical test not given. No change in findings when medicated Ss eliminated from analysis.

 b) Systolic blood pressure-SBP
   RA: 122 base
     120 1 year
   BA: 121 base
     121 1 year

*Reported no significant effect of group on SBP. Statistical test not given.

 c) Diastolic blood pressure-DBP
   RA: 78 base
     77 1 year
   BA: 76 base
     76 1 year

*Reported no significant effect of group on DBP. Statistical test not given.

 d) HDL Cholesterol
   RA: 1.1 base
     1.1 1 year
   BA: 1.1 base
     1.1 1 year

*Reported no significant effect of group on HDL-C. Statistical test not given.

 e) LDL-Cholesterol
   RA: 3.2 base
     3.1 1 year
   BA: 3.2 base
     3.01 year

*Reported no significant effect of group on LDL-C. Statistical test not given.

3) Events:

 a) Health care utilization: Not given
 b) Morbidity/mortality: Not given

Quality Assessment:

Internal Validity:

Described as randomized? Yes

Method of randomization clearly described? No

Concealment of allocation? No

Described as double-blind? No

Patient blinded? No

Investigators blinded? No

Outcome assessors blinded? No

No. of withdrawals in each group stated? Yes

External Validity:

Pop. Described? Yes

Intervention described well enough to reproduce? Yes

Intervention codified in manual? Yes

Provider training described? No

Patients assessed for DSM dx? No

Biases, etc: Statistical analyses not clearly explained; differential attrition—more in treatment group (RA)

#510

Fosbury, Bosley, Ryle, Sonksen, Judd, 1997

Include: type I diabetes, 18-55 years old, poor diabetes control, HbA1c >9%.

Exclude: pregnant, in other clinical trials, or those living a considerable distance from the hospital.

RCT with 2 treatment conditions:
1) CAT treatment— cognitive analytic therapy (cat).
2) DSNE Control— diabetes specialist nurse education (dsne)

N=32
n.cat=15
n.dsne=17

*6 drop-outs (5 from CAT, 1 from dsne)

Age means (SD):
cat=30.5(10.6)
dsne=32(9.2)

% Female:
cat=70
dsne=69

Race %:
88- Caucasian
8- African Amer.
4- Asian

Baseline HbA1 % means (SD):
Completers:
cat=12.12(1.37)
dsne=11.76(1.88)

1) CAT—a time limited (16-20 sessions) focused psychotherapy, using psychosomatic and CBT methods, where self-care and relationships with others are understood as sequences of mental and behavioral processes. CAT therapist makes links between the patients' past and present experiences and their use of procedures that are ineffective and harmful.

2) DSNE—involved teaching, counseling, and advice about diabetes management in relation to the personal needs and lifestyle of the patient.

16 (50 min) sessions, approx. once a week, 3 and 6 month followup

Completer Results:

1) Metabolic control
 - HbA1 % means (SD):
   cat: 12.1 (1.4) base
     11.0 (2.0) post
     10.6 (1.3) 3 mo
     10.1 (1.5) 9 mo
   dsne: 11.8 (1.9) base
     10.6 (2.0) post
     10.5 (2.2) 3 mo
     10.9 (1.5) 9 mo

*t-tests indicated no significant differences between groups. Both groups showed significant within group improvements at 3- and 6-months

2) Measures of risk: Not given

3) Events
 a) Health care utilization: Not given
 b) Morbidity/mortality: Not given

Quality Assessment:

Internal Validity:

Described as randomized? Yes

Method of randomization clearly described? Yes

Concealment of allocation? Yes

Described as double-blind? No

Patient blinded? No

Investigators blinded? No

Outcome assessors blinded? No

No. of withdrawals in each group stated? Yes

External Validity:

Pop. Described? Yes

Intervention described well enough to reproduce? Yes

Intervention codified in manual? Yes

Provider training described? No

Patients assessed for DSM dx? No

Biases, etc: No measures of risk assessed; disproportionate attrition in the intervention and control group.

#210

Gaede, Beck, Vedel & Pederson, 2001

Include: Type 2 diabetes mellitus, age 45-65

Exclude: not stated

RCT-single center with 2 groups:
1) Standard intervention (con).
2) Intensive multifactorial intervention (tx).

N=160
n con=76
n tx=73

*5 drop-outs- tx: 3, con: 2) and 6 died

Age mean (SD): 55.1 (7.2)

25% Female

Race % not given

Baseline HbA1c % means (SD):
Intended to treat:
con: 8.8 (1.7)
tx: 8.4 (1.5)

Both groups received information on diet, exercise, and smoking cessation. Tx group was taught to se individual goals for diet, smoking and exercise, received spouse-assisted training to help retain their goals, engaged in self-monitoring, and were both encouraged to exercise more and was offered smoking cessation programs

6 months.

Completer Results:

1) Metabolic control:
 -HbA1c % means (SD):
   con: 8.8 (1.7) base
     9.0 (1.8) post
   tx: 8.4 (1.5) base
     7.6 (1.0) post*

*Reported a significant decrease in HbA1c for tx group (p<0.01), and a significant difference between groups at post (p<0.000001). Statistical tests not given.

2) Measures of risk:
 a) Weight (kg) means (SD):
   con: 89.9 (17.3) base
     90.4 (16.4) post
   tx: 91.4 (13.6) base
     95.1 (13.2) post*

*Reported a significant increase in weight for tx group (p<0.001), and a significant difference between groups at post (p=0.001). Statistical test not given.

 b) Current Smokers:
con: 26 base
     21 post
   tx: 28 base
     22 post

*Reported a significant decrease in smokers for both con and tx groups (p<0.05), yet no significant difference between groups. Statistical test not given.

 c) Fasting Total Cholesterol means (SD):
   con: 5.8 (1.3) base
     5.5 (1.2) post
   tx: 5.4 (1) base
     4.8 (0.7) post

Reported a significant decrease in total cholesterol for tx group (p<0.001), and a significant difference between groups (p=0.00003). Statistical test not given.

 d) Fasting HDL Cholesterol means (SD):
   con: 1.01 (0.3) base
     1.04 (0.3) post
   tx: 1.03 (0.2) base
     1.05 (0.3) post

*Reported no significant differences between groups at post. Statistical tests not given.

3) Events:
 a) Health care utilization: Not given
 b) Morbidity/mortality: 6 patients died during f/u

Quality Assessment:

Internal Validity:

Described as randomized? Yes

Method of randomization clearly described? No

Concealment of allocation? No

Described as double-blind? No

Patient blinded? No

Investigators blinded? No

Outcome assessors blinded? No

No. of withdrawals in each group stated? Yes

External Validity:

Pop. Described? Yes

Intervention described well enough to reproduce? No

Intervention codified in manual? No

Provider training described? No

Patients assessed for DSM dx? No

Biases, etc: Behavioral intervention not explained clearly; statistical analyses not stated.

c.Higher scores on the SCL-90R, ZERSSEN, and IRES indicate more disease related distress and quality of life respectively.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.

 

AHRQ Advancing Excellence in Health Care