Background
Methods
Setting
Study sample and procedures
Depression instruments
Statistical analysis
Results
Participant flow
Baseline characteristics of participants
Characteristic | |
---|---|
Age, years, mean ± SD* | 78 ± 7 |
Female, % | 62 |
Non-white, % | 32 |
High school graduate, % | 82 |
Chronic medical conditions | |
Mean ± SD | 2.3 ± 1.5 |
Median (interquartile range) | 2 (2) |
Three or more chronic medical conditions, % | 42 |
Patient Health Questionnaire-9 (PHQ-9) | |
Score, mean ± SD | 5.9 ± 6.1 |
Score ≥10, %†
| 23 (16/71) |
Patient Health Questionnaire-2 (PHQ-2) | |
Score, mean ± SD | 1.2 ± 1.6 |
Score ≥3, %‡
| 20 (14/71) |
15-Item Geriatric Depression Scale (GDS) | |
Score, mean ± SD | 5.8 ± 1.8 |
Score >5, %§
| 48 (33/69) |
Structured Clinical Interview for Depression (SCID) | |
Major depression, % | 12 (8/69) |
Minor depression, % | 13 (9/69) |
Relative performance of depression screening instruments for major depression
Instrument and Cutpoint | Sensitivity (%) (95% CI) | Specificity (%) (95% CI) | + Likelihood Ratio | - Likelihood Ratio |
---|---|---|---|---|
PHQ-9 Cutpoint | ||||
≥8 | 88 | 75 | 3.6 | 0.16 |
(56-98) | (71-77) | |||
≥9
|
88
|
80
| 4.4 | 0.16 |
(56-98) | (76-82) | |||
≥10
|
63
|
82
| 3.5 | 0.46 |
(33-86) | (78-85) | |||
≥11 | 63 | 84 | 3.8 | 0.45 |
(33-85) | (80-87) | |||
≥12 | 63 | 84 | 3.8 | 0.45 |
(33-85) | (80-87) | |||
PHQ-2 Cutpoint | ||||
≥1 | 88 | 61 | 2.2 | 0.21 |
(55-98) | (56-62) | |||
≥2
|
75
|
67
| 2.3 | 0.37 |
(43-93) | (63-70) | |||
≥3
|
63
|
85
| 4.2 | 0.44 |
(33-85) | (81-88) | |||
≥4 | 38 | 93 | 5.7 | 0.67 |
(15-62) | (91-97) | |||
≥5 | 38 | 98 | 22.9 | 0.64 |
(16-48) | (96-100) | |||
15-item GDS Cutpoint | ||||
≥5 | 100 | 15 | 1.2 | 0.00 |
(73-100) | (11-15) | |||
≥6
|
100
|
58
| 2.4 | 0.00 |
(70-100) | (54-59) | |||
≥7
|
75
|
77
| 3.2 | 0.33 |
(43-93) | (72-79) | |||
≥8 | 25 | 87 | 1.9 | 0.86 |
(7-54) | (84-91) | |||
≥9 | 13 | 93 | 1.9 | 0.94 |
(2-37) | (92-97) |
N* | PHQ-9 AUC†
(95% CI)‡
| PHQ-2 AUC (95% CI) | GDS AUC (95% CI) | |
---|---|---|---|---|
Overall | 69 | 0.87 (0.74-1.00) | 0.81 (0.65-0.98) | 0.81 (0.70-0.91) |
Gender | ||||
Male | 26 | 0.88 (0.72-1.00) | 0.91 (0.78-1.00) | 0.76 (0.58-0.93) |
Female | 43 | 0.85 (0.61-1.00) | 0.70 (0.37-1.00) | 0.84 (0.70-0.98) |
Race | ||||
White | 47 | 0.88 (0.73-1.00) | 0.87 (0.75-0.99) | 0.85 (0.74-0.95) |
Non-white‡
| 22 | 0.93 | 0.31 | 0.90 |
Age | ||||
< 80 years | 38 | 0.92 (0.81-1.00) | 0.81 (0.54-1.00) | 0.82 (0.68-0.96) |
≥80 years | 31 | 0.80 (0.52-1.00) | 0.83 (0.64-1.00) | 0.80 (0.62-0.98) |
Comorbidities | ||||
< 3 | 41 | 0.93 (0.82-1.00) | 0.92 (0.85-1.00) | 0.79 (0.55-1.00) |
≥3 | 28 | 0.80 (0.57-1.00) | 0.71 (0.43-0.99) | 0.82 (0.67-0.97) |
Relative performance of depression screening instruments using broadened definition of depression
Instrument and Cutpoint | Sensitivity (%) (95% CI) | Specificity (%) (95% CI) | + Likelihood Ratio | - Likelihood Ratio |
---|---|---|---|---|
PHQ-9 Cutpoint | ||||
≥6 | 77 | 69 | 2.5 | 0.34 |
(56-90) | (63-74) | |||
≥7 | 77 | 77 | 3.3 | 0.31 |
(56-90) | (70-81) | |||
≥8
|
77
|
83
| 4.4 | 0.28 |
(57-89) | (76-87) | |||
≥9 | 71 | 87 | 5.2 | 0.34 |
(51-85) | (80-91) | |||
≥10 | 59 | 89 | 5.1 | 0.47 |
(40-74) | (82-93) | |||
PHQ-2 Cutpoint | ||||
≥1
|
82
|
67
| 2.5 | 0.26 |
(62-94) | (61-71) | |||
≥2 | 71 | 73 | 2.6 | 0.40 |
(50-86) | (66-78) | |||
≥3 | 53 | 90 | 5.5 | 0.52 |
(35-67) | (85-95) | |||
≥4 | 35 | 98 | 18.4 | 0.66 |
(21-40) | (94-100) | |||
≥5 | 18 | 98 | 9.2 | 0.84 |
(7-22) | (95-100) | |||
15-item GDS Cutpoint | ||||
≥4 | 94 | 4 | 0.98 | 1.63 |
(85-99) | (1-5) | |||
≥5 | 88 | 14 | 1.0 | 0.93 |
(71-96) | (8-16) | |||
≥6
|
81
|
62
| 2.1 | 0.30 |
(60-93) | (55-65) | |||
≥7 | 56 | 79 | 2.7 | 0.55 |
(36-74) | (73-84) | |||
≥8 | 31 | 90 | 3.3 | 0.76 |
(16-46) | (86-95) |