(1) Patient characteristics
Table 1 and eTable 3 present the characteristics of the patients in the exposure cohort (those prescribed ketamine) and the comparison cohort (those prescribed other antidepressants) before and after propensity-score matching. Before matching, the exposure cohort was older (average 49.4 vs. 43.2 years) and had significantly higher prevalence of comorbidities and adverse socioeconomic determinants of health. The two cohorts were balanced after propensity-score matching, yielding 12,662 patients each in the exposure and comparison cohorts (Table 1).
Table 1. The characteristics of patients prescribed ketamine (“Ketamine cohort”) and other antidepressants (“Comparison cohort”) before and after propensity-score matching (SMD: standardized mean differences, *SMD greater than 0.1, a threshold being recommended for declaring imbalance.)
Characteristics
|
Before matching
|
After matching
|
Cohort, No. (%)
|
Cohort, No. (%)
|
Ketamine cohort
|
Comparison cohort
|
SMD
|
Ketamine cohort
|
Comparison cohort
|
SMD
|
Cohort size
|
12,708
|
308,659
|
|
12,662
|
12,662
|
|
Age at Index
|
49.4 ± 16.7
|
43.2 ± 20.1
|
0.33*
|
49.4 ± 16.7
|
50.2 ± 19.1
|
0.03
|
Female
|
65.8
|
69.0
|
0.07
|
65.8
|
65.8
|
<.001
|
Race and ethnicity
|
White
|
74.5
|
72.8
|
0.04
|
74.4
|
74.4
|
<.001
|
Unknown Race
|
12.2
|
13.7
|
0.05
|
12.2
|
12.5
|
0.01
|
Black or African American
|
11.8
|
11.2
|
0.02
|
11.8
|
11.4
|
0.01
|
Hispanic or Latino
|
6.2
|
7.5
|
0.05
|
6.2
|
6.1
|
0.01
|
Asian
|
0.8
|
1.6
|
0.08
|
0.8
|
0.8
|
0.01
|
Risk factors and complications (based on encounter diagnosis International Classification of Diseases (ICD) codes)
|
Potential health hazards related to family and personal history
|
85.6
|
57.3
|
0.66*
|
85.5
|
86.7
|
0.03
|
Anxiety disorder, unspecified
|
59.9
|
44.9
|
0.31*
|
59.8
|
60.7
|
0.02
|
Hypertensive diseases
|
56.7
|
33.3
|
0.48*
|
56.6
|
57.1
|
0.01
|
Sleep disorders
|
49.9
|
28.3
|
0.45*
|
49.8
|
50.1
|
0.01
|
Chronic pain, not elsewhere classified
|
47.3
|
19.9
|
0.61*
|
47.2
|
47.6
|
0.01
|
Overweight and obesity
|
46.1
|
23.2
|
0.49*
|
46.0
|
46.5
|
0.01
|
Mental and behavioral disorders due to psychoactive substance use
|
46.0
|
29.0
|
0.36*
|
45.9
|
46.4
|
0.01
|
Diabetes mellitus
|
27.7
|
14.6
|
0.32*
|
27.7
|
28.1
|
0.01
|
Chronic ischemic heart disease
|
17.7
|
7.5
|
0.31*
|
17.6
|
18.3
|
0.02
|
Post-traumatic stress disorder (PTSD)
|
15.3
|
11.6
|
0.11*
|
15.2
|
15.1
|
<.001
|
Heart failure
|
14.5
|
4.8
|
0.33*
|
14.5
|
14.5
|
<.001
|
Persons with potential health hazards related to socioeconomic and psychosocial circumstances
|
14.3
|
11.3
|
0.09
|
14.3
|
14.2
|
<.001
|
Pre-existing suicidal ideations
|
13.9
|
14.8
|
0.03
|
13.8
|
13.8
|
<.001
|
Epilepsy and recurrent seizures
|
6.4
|
3.2
|
0.15*
|
6.3
|
6.7
|
0.02
|
Schizophrenia, schizotypal, delusional, and other non-mood psychotic disorders
|
5.8
|
4.0
|
0.08
|
5.7
|
5.6
|
0.01
|
Cerebral infarction
|
5.7
|
2.8
|
0.14*
|
5.7
|
6.0
|
0.01
|
Personal history of self-harm
|
5.7
|
6.0
|
0.01
|
5.6
|
5.4
|
0.01
|
Intentional self-harm
|
2.4
|
1.3
|
0.08
|
2.3
|
2.5
|
0.01
|
Pre-existing suicide attempt
|
2.1
|
1.5
|
0.04
|
2.0
|
1.9
|
0.01
|
(2) Ketamine prescription is associated with decreased suicidal ideation compared to other antidepressant prescription.
As shown in Figure 2, the prescription of ketamine was associated with significant decrease in suicidal ideation: HR = 0.65 (95% CI: 0.53 – 0.81) at 1 day – 7 days, 0.78 (95% CI: 0.66 – 0.92) at 1 day – 30 days, 0.81 (95% CI: 0.70 – 0.92) at 1 day – 90 days, 0.82 (95% CI: 0.72 – 0.92) at 1 day – 180 days, 0.83 (95% CI: 0.74 – 0.93) at 1 day – 270 days. No significant difference was observed for suicide attempt.
(3) The association between ketamine prescription and suicidal ideation or suicide attempt varies by age, gender, and race.
In patients ages 24 and older, ketamine prescription is associated with significant decrease in suicidal ideation compared to prescription of other antidepressants at 1 day – 7 days, 1 day -30 days, 1 day – 90 days, 1 day – 180 days, 1 day – 270 days after initial prescription: HR = 0.69 (95% CI: 0.54 – 0.88), 0.81 (95% CI: 0.67 – 0.98), 0.81 (95% CI: 0.70 – 0.95), 0.84 (95% CI: 0.73 - 0.96), 0.86 (95% CI: 0.76-0.98) (Figure 3). In male patients, ketamine prescription is associated with significant decrease in suicidal ideation compared to prescription of other antidepressants at 1 day – 7 days, 1 day -30 days, 1 day – 90 days, 1 day – 180 days, 1 day – 270 days after initial prescription: HR = 0.57 (95% CI: 0.42 – 0.78), 0.71 (95% CI: 0.56 – 0.88), 0.76 (95% CI: 0.63 – 0.92), 0.78 (95% CI: 0.66 - 0.93), 0.79 (95% CI: 0.68-0.93) (Figure 3). In White patients, the prescription of ketamine is associated with significant decrease in suicidal ideation compared to prescription of other common antidepressants at 1 day – 7 days, 1 day -30 days, 1 day – 90 days, 1 day – 180 days, 1 day – 270 days after initial prescription: HR = 0.63 (95% CI: 0.50 – 0.81), 0.75 (95% CI: 0.63 – 0.91), 0.74 (95% CI: 0.64 – 0.87), 0.75 (95% CI: 0.65 - 0.86), 0.79 (95% CI: 0.69-0.90) (Figure 3). In female patients, the prescription of ketamine is associated with significant decrease in suicidal ideation compared to prescription of other common antidepressants at 1 day – 90 days, 1 day – 180 days, 1 day – 270 days after initial prescription: HR = 0.74 (95% CI: 0.60 – 0.90), 0.78 (95% CI: 0.66 – 0.93), 0.85 (95% CI: 0.72 – 1.00). However, no significant differences were observed at 1 day – 7 days or 1 day – 30 days (Figure 3). No significant differences were observed for risk of suicidal ideation between the exposure and comparison cohorts in adolescents (aged 10-24 years) or Black patients.
In patients aged 10-24, ketamine prescription is associated with increased risk of suicide attempt at 1 day – 30 days: HR = 2.22 (95% CI: 1.01-4.87). No other significant differences were observed for suicide attempt among the demographic-stratified subgroups (Figure 4), and the results for Black patients are not presented due to insufficient sample size.