Suicide Among U.S. Adults With Hepatocellular Carcinoma

TO THE EDITOR: Depression, which is highly prevalent in cancer patients, can result in fatal outcomes from suicide if inadequately managed. In a population-based study by Misono et al¹ in 2008, the overall incidence of suicide in patients with cancer in the United States was approximately twice that of the general population. Recently, in a retrospective analysis of the SEER (Surveillance, Epidemiology, and End Results) dataset, Turaga et al² reported that male patients with pancreatic adenocarcinoma were 11 times more likely to commit suicide than those in the general U.S. population. Epidemiologic evidence suggests that the incidence rates of hepatocellular carcinoma (HCC) are increasing in the United States.³ Patients with HCC can be considered at increased risk of depression and suicide and should be evaluated for intervention, but data on which to base such initiatives are lacking. We report the suicide rate and determinants of suicide among patients diagnosed with HCC in the U.S. population.

The SEER database was queried to identify patients with HCC from 1990 to 2009, using the ICD-3 code: 8170/3. Patients were considered to have committed suicide if the cause-of-death variable was coded as “suicide and self-inflicted injury.” Patients with other causes of death were not classified as suicides. Standardized mortality was calculated via the ratio of observed mortality with expected mortality from a standardized U.S. population.⁴ Logistic regression models were used to perform multivariate modeling for factors associated with suicide, while Kaplan-Meier analyses were used to assess factors affecting survival.

Among 51,331 patients followed for 60,325 person-years, the suicide rate was 51.4/100,000 person-years. The corresponding rate in the overall U.S. population was 12.0/100,000 person-years, with a standardized mortality ratio (SMR) of 4.3 (95% confidence interval [CI], 3.3–5.5). Table 1 summarizes the suicide rates and SMR in our study. In the multivariate analysis, male gender (adjusted odds ratio [OR], 10.2; 95%, 1.4–75.8, P = .02) was independently associated with suicide. Median survival among patients who committed suicide was 3.5 months (95% CI, 0.2–6.8) compared with 5.5 months (95% CI, 5.4–5.6) for those who did not commit suicide (log rank test; P = .08).

Table 1.

Suicide rates and standardized mortality ratios

Variable	Suicides (n)	Person-years	Suicide rate per 100,000 person-years	SMR compared to U.S. population	95% CI
Age (y)
<55	5	18,542	27.0	2.3	1.5–3.3
55–64	11	17,660	62.3	3.7	2.9–4.7
65–74	8	14,743	54.3	3.9	3.0–4.9
>74	7	9,379	74.6	4.8	3.8–5.9
Gender
Female	1	16,112	6.2	1.2	0.7–2.3
Male	30	44,213	67.9	3.5	2.8–4.4
Race
White	22	38,186	57.6	4.3	3.3–5.4
Black	5	5,794	86.3	16.9	13.8–20.6
Asian/Pacific Islander	4	15,355	26.1	4.5	3.2–6.4
American Indian/Alaskan Native	0	703	0	0	0
Marital status
Married	20	36,873	54.2	4.5	3.5–5.8
Single	11	23,452	46.9	3.9	2.9–5.0
SEER Stage
Localized	17	37,510	45.3	3.7	2.9–4.9
Regional	7	12,600	55.6	4.6	3.6–5.9
Distant	3	4,194	71.53	5.9	4.8–7.4
Surgical intervention	8	29,855	26.8	2.2	1.6–3.1
(resection or transplant)
No surgical intervention	23	30,470	75.5	6.4	5.1–7.7
intervention
Period of diagnosis
1990–1999	9	13,237	68.0	6.4	5.0–8.0
2000–2009	22	47,088	46.7	3.9	2.9–5.0

Abbreviations: SMR = standardized mortality ratios; CI = confidence interval; SEER = Surveillance, Epidemiology, and End Results.

Overall SMR for the general U.S. population was 12.0/100,000 person-years in 2009; SMR for the male U.S. population was 19.2/100,000 person-years and for the female U.S. population was 5.0/100,000 person-years. Age group, race, and period of diagnosis were also compared with adjusted U.S. population suicide rates.⁴

In recent years, because of the improvement in survival of cancer patients, a paradigm shift in focus among both healthcare providers and patients has emerged toward improvement in quality of life for survivors. Misono et al¹ suggested that the suicide rate among cancer patients in the United States was approximately 4 times that of the general population. Male patients with HCC have a risk of suicide that is nearly 4 times that of the general population and about 10 times greater than their female counterparts. We realize that our study is subject to several limitations, including misclassification bias, related primarily to the nature of the data collected. However, one study has shown that data were entered accurately in the SEER database and that misclassification bias is minimal.⁵ Despite these limitations, our findings suggest the need for early social support and prevention efforts in patients (especially males) diagnosed with HCC. Further research into the suicide risk of HCC patients should extend the range of concerns from clinical to socioeconomic factors.