Abstract
Objective:
This study explored nurse suicide in the United States.
Methods:
Characteristics were compared between occupations using 2014 National Violent Death Reporting System data.
Results:
Female nurse suicides were significantly higher (11.97/100,000) than in the female population (7.58/100,000) (p<0.001); similarly male nurses (39.8/100,000) compared to the male population (28.2/100,000) (p<0.001). Benzodiazepines and opioids were the most commonly used substances used in clinician suicide.
Conclusion:
These results suggest a public health imperative for future research and development of effective preventative strategies for nurses; a largely understudied population.
Keywords: suicide, occupational health, overdose, firearms, nurses
Suicide rates are increasing globally ("Increase in suicide rates, 1999-2014," 2016; Lieberman, 2016) with about 800,000 people dying of suicide each year, making it the 17th leading cause of death worldwide in 2016. ("Community Engagement Kit," 2017; "Suicide," 2018; World Health Organization, 2014) Suicide is the 10th leading cause of death in the United States (U.S.) with almost 45,000 deaths related to suicide yearly. The annual U.S. age-adjusted rate of suicide is 13.42 per 100,000 individuals. (Last, Tufts, & Auger, 2017; Organizations & Hospital, 2007) Approximately 400 physicians complete suicide annually. (Andrew, 2015) However, the incidence of nurse suicide in the United States is largely unchartered (; Davidson, Stuck, Zisook, & Proudfoot, 2018). Outdated U.S. (Alston, 1986; Feskanich et al., 2002; K. Hawton & Vislisel, 1999; Katz, 1983; Stack, 1982, 2001) and more recent international literature (Alderson, Parent-Rocheleau, & Mishara, 2015; Andersen, Hawgood, Klieve, Kolves, & De Leo, 2010; Cheung & Yip, 2015, 2016; K. Hawton et al., 2002; Kolves & De Leo, 2013; A. Milner, Spittal, Pirkis, & LaMontagne, 2013; A. J. Milner, Maheen, Bismark, & Spittal, 2016; Tramutola, 2015) suggests that nurses are at higher risk of suicide than the general public.
As a pilot to refine the methodology for this study, we conducted a longitudinal retrospective review of nurse suicides in San Diego County between 2005-2014. Overall RN (18.51) and physician (40.72) incidence of suicide were higher than the general population excluding nurses (15.81) normalized to 100,000 person years. Nurse gender was available for only 2014. In that year female nurses were at a higher risk of suicide than females in the general population (10.41 vs. 7.41 per 100,000 person years). Although nurses completed suicide at a numerically higher rate (18.51) than the general public (15.81) normalized to 100,000 person years, we did not find a statistically significant difference in the two rates. (Davidson, Stuck, et al., 2018)
The purpose of this study was to quantify the incidence, methods, and predictive factors of nurse suicide in the United States through retrospective correlational analysis of the National Violent Death Reporting System (NVDRS) database for the year 2014.
Methods
Ethics
This project was excused from Investigational Review Board oversight as research on deceased individuals using a de-identified data set (#170165). The Centers for Disease Control and Prevention (CDC) department for the National Violent Death Reporting System (NVDRS) provided ethical oversight for use of this data.
Denominators
Denominators for nurses were obtained from the National Council of State Boards of Nursing website (https://www.ncsbn.org/6161.htm) access date 7/16/2017 for all states excluding Oklahoma, which does not report to the national database. Oklahoma nursing denominators were obtained from the Oklahoma Board of Nursing website (http://nursing.ok.gov/pubs.html) access date 7/16/2017. Denominators represent active licensed practitioners (not necessarily employed). Gender data were obtained (Budden, Moulton, Harper, Brunell, & Smiley, 2016) Of note, nurse gender data are generated from surveys conducted periodically over time. There were published data for 2013 (Budden, Zhong, Moulton, & Cimiotti, 2013) and 2015 (Budden et al., 2016) but not 2014. We retrieved the unpublished estimated survey data directly from the NCSBRN for the 18 states represented in the NVDRS dataset to generate an estimated average of 7% male. The denominator was age matched for suicides from the general population of people aged 20 and older. We confirmed that age of entry of nurses into the profession with the National Councils of State Boards of Registered Nursing.
NVDRS/Sample
The NVDRS is a restricted dataset provided by the Center for Disease Control and Prevention. Unlike the general mortality statistics, it is coded by occupation. The 2014 dataset contained all suicides reported by medical examiners from 18 states. (Electronic supplement 1)
Occupation Codes
Though the NVDRS is constructed to be coded by occupation using occupation codes from death certificates, this field was found to be universally left blank. However, the occupation had been entered free text into the database. Using iterative manual natural language processing, two investigators analyzed the free text fields. First, the occupation column was sorted alphabetically, then the search function was with ‘nur’, ‘RN’. Next, the column was scanned row by row to detect other derivatives of the terms. The final list was critiqued by the other members of the research team (Electronic supplement 2).
Data Analysis
Victim characteristics were compared between each occupation group (nurse vs. other with Fisher’s exact test for categorical variables and a two-sample t-test for continuous variables (i.e. number of crises). Conditional maximum likelihood estimates of odds ratios and their 95% confidence intervals (CI) were calculated for binary categorical characteristics. Incidence rates for each occupation of interest are reported in units of 100000 person-years. Incidence rate ratios (IRR) and associated 95% Wald confidence intervals were computed. A chi-squared test was used to test for differences in exposed and non-exposed proportions for each strata, with exposure defined by medical profession and/or gender. Homogeneity of IRR’s across strata was tested using the Breslow-Day test.
Results
Demographics
Characteristics of subjects including sex, age, body mass index (BMI), race, and ethnicity are found in Table 1. Denominator counts of the general population are found in Electronic Supplement 3. In 2014 in the 18 states included in the dataset there were 14,774 suicides, of which there were 205 nurses.
Table 1:
Characteristics of subjects from the NVDRS database for 2014.
| Occupation | ||||
|---|---|---|---|---|
| Nurse | Other | Overall | p-value | |
| Sex | n = 205 | n = 14504 | n = 14709 | |
| Female | 164 (80.0%) | 3210 (22.1%) | 3374 (22.9%) | < 0.001*** |
| Male | 41 (20.0%) | 11294 (77.9%) | 11335 (77.1%) | |
| Age | n = 205 | n = 14503 | n = 14708 | |
| 51.13 (12.64) | 46.67 (18.33) | 46.73 (18.27) | < 0.001*** | |
| BMI | n = 139 | n = 9485 | n = 9624 | |
| 27.27 (7.19) | 26.79 (6.14) | 26.79 (6.16) | 0.431 | |
| Race | n = 205 | n = 14368 | n = 14573 | |
| American Indian | 1 (0.5%) | 194 (1.4%) | 195 (1.3%) | 0.430 |
| Asian / PI | 4 (2.0%) | 256 (1.8%) | 260 (1.8%) | |
| Black | 9 (4.4%) | 933 (6.5%) | 942 (6.5%) | |
| Other | 0 (0.0%) | 169 (1.2%) | 169 (1.2%) | |
| Two or more | 1 (0.5%) | 112 (0.8%) | 113 (0.8%) | |
| White | 190 (92.7%) | 12704 (88.4%) | 12894 (88.5%) | |
| Ethnicity | n = 205 | n = 14413 | n = 14618 | |
| Hispanic | 3 (1.5%) | 692 (4.8%) | 695 (4.8%) | 0.020* |
| Non-Hispanic | 202 (98.5%) | 13721 (95.2%) | 13923 (95.2%) | |
Suicide Method
The method of suicide varied between nurses and others. The most common method in nurses was poisoning by pharmaceuticals. In contrast, in the general population, the most common method was firearms. (Table 2) Substances listed as cause of death were diverse in nature, and did not cluster to a specific classification. Further, opioids, benzodiazepenes and antihistamines were the most common classes of medications attributed to cause of death found in the toxicology screens of nurses who completed suicide. (Electronic Supplement 4)
Table 2:
Suicide method by occupation.
| Occupation | |||||
|---|---|---|---|---|---|
| Nurse (n = 205) |
Physician (n = 50) |
Pharmacist (n = 15) |
Other (n = 14504) |
Overall (n = 14774) |
|
| Method | |||||
| Firearm | 69 (33.7%) | 29 (58.0%) | 5 (33.3%) | 7464 (51.5%) | 7567 (51.2%) |
| Hanging, strangulation, suffocation | 28 (13.7%) | 8 (16.0%) | 4 (26.7%) | 3954 (27.3%) | 3994 (27.0%) |
| Poisoning (Pharm.) | 72 (35.1%) | 5 (10.0%) | 2 (13.3%) | 1319 (9.1%) | 1398 (9.5%) |
| Poisoning (Other) | 26 (12.7%) | 4 (8.0%) | 3 (20.0%) | 818 (5.6%) | 851 (5.8%) |
| Sharp instrument | 5 (2.4%) | 2 (4.0%) | 0 (0.0%) | 258 (1.8%) | 265 (1.8%) |
| Fall | 4 (2.0%) | 1 (2.0%) | 1 (6.7%) | 253 (1.7%) | 259 (1.8%) |
| Drowning | 1 (0.5%) | 1 (2.0%) | 0 (0.0%) | 123 (0.8%) | 125 (0.8%) |
| Motor vehicle | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 106 (0.7%) | 106 (0.7%) |
| Other transport vehicle | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 82 (0.6%) | 82 (0.6%) |
| Fire or burns | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 56 (0.4%) | 56 (0.4%) |
| Unknown | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 22 (0.2%) | 22 (0.1%) |
| Other | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 21 (0.1%) | 21 (0.1%) |
| Blunt instrument | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 20 (0.1%) | 20 (0.1%) |
| Non-powder gun | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 5 (0.0%) | 5 (0.0%) |
| Explosives | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 2 (0.0%) | 2 (0.0%) |
| Intentional neglect | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 1 (0.0%) | 1 (0.0%) |
Incidence
The incidence rate of suicides in nurses (male and female) was 13.9 per 100,000 person-years compared to 17.7 per 100,000 person-years in the general population (ages 20+). The IRR was 0.79 with 95% CI 0.68 to 0.90 (p < 0.001). The incidence rate of suicides in female nurses was 11.97 per 100,000 person-years, and 7.58 per 100,000 person-years in the female general population (ages 20+), for an IRR between female nurses and the female general population (age 20+) of 1.58 (95% CI: 1.34 to 1.85, p < 0.001). The incidence rate of suicides in male nurses was 39.8 per 100,000 person-years compared to 28.2 per 100,000 person-years in the male general population (ages 20+). The IRR between male nurses and the male general population (age 20+) is 1.41 (95% CI: 1.01 to 1.91, p = 0.028). The incidence rate of suicides in female nurses was significantly lower than in male nurses (11.97 per 100,000 person-years in female nurses vs. 39.8 per 100,000-person years in male nurses (IRR = 0.30 ((95% CI: 0.21 to 0.43, p < 0.001)). The incidence rate of suicides in females in the general population was 7.58 per 100,000 person-years vs. 28.23 per 100,000 person-years in males in the general population (IRR = 0.27 ((95% CI: 0.26 to 0.28, p <0.001)). The male:female suicide ratios in nurses was 3.32:1 compared to3.72:1 in the general population. There was no evidence of non-homogeneity between the female and male IRR in the nurse population and non-nurse population (p = 0.515).
Mental health and social risk factors for nurses vs. the general population are provided (Table 3). Nurses were statistically significantly more likely to have reported mental health problems, history of a treatment of mental illness, history of previous suicide attempt, leaving a suicide note and physical health problems than the general population. Nurses were significantly less likely to have a recent criminal problem or alcohol problem than the general population.
Table 3:
Mental health and other characteristics for nurses and other occupations (not including nurses, physicians, or pharmacists) from the NVDRS database for 2014 (only including subjects who are at least 20 years old).
| Occupation | ||||
|---|---|---|---|---|
| Nurse (n = 205) |
Other (n = 13645 |
p-value | Odds Ratio | |
| Mental Health Problems | ||||
| No | 93 (45.4%) | 7749 (56.8%) | 0.001** | 0.632 (0.474, 0.841) |
| Yes | 112 (54.6%) | 5896 (43.2%) | ||
| Depressed Mood | ||||
| No | 129 (62.9%) | 9097 (66.7%) | 0.264 | 0.849 (0.633, 1.145) |
| Yes | 76 (37.1%) | 4548 (33.3%) | ||
| Mental Illness Treatment Current | ||||
| No | 124 (60.5%) | 10186 (74.7%) | < 0.001*** | 0.520 (0.389, 0.699) |
| Yes | 81 (39.5%) | 3459 (25.3%) | ||
| History Mental Illness Treatment | ||||
| No | 114 (55.6%) | 9230 (67.6%) | < 0.001*** | 0.599 (0.450, 0.800) |
| Yes | 91 (44.4%) | 4415 (32.4%) | ||
| Alcohol Problems | ||||
| No | 185 (90.2%) | 11323 (83.0%) | 0.005** | 1.897 (1.190, 3.185) |
| Yes | 20 (9.8%) | 2322 (17.0%) | ||
| Substance Abuse | ||||
| No | 169 (82.4%) | 11656 (85.4%) | 0.232 | 0.801 (0.554, 1.186) |
| Yes | 36 (17.6%) | 1989 (14.6%) | ||
| Intimate Partner Problem | ||||
| No | 154 (75.1%) | 10127 (74.2%) | 0.810 | 1.049 (0.758, 1.473) |
| Yes | 51 (24.9%) | 3518 (25.8%) | ||
| Relationship Problem | ||||
| No | 203 (99.0%) | 13410 (98.3%) | 0.589 | 1.779 (0.481, 14.875) |
| Yes | 2 (1.0%) | 235 (1.7%) | ||
| Interpersonal Violence Perp. | ||||
| No | 203 (99.0%) | 13316 (97.6%) | 0.247 | 2.508 (0.680, 20.934) |
| Yes | 2 (1.0%) | 329 (2.4%) | ||
| Interpersonal Violence Victim | ||||
| No | 205 (100.0%) | 13598 (99.7%) | 1.000 | Inf (0.183, Inf) |
| Yes | 0 (0.0%) | 47 (0.3%) | ||
| Argument | ||||
| No | 176 (85.9%) | 11806 (86.5%) | 0.757 | 0.945 (0.633, 1.457) |
| Yes | 29 (14.1%) | 1839 (13.5%) | ||
| Circumstances Known | ||||
| No | 13 (6.3%) | 1307 (9.6%) | 0.149 | 0.639 (0.333, 1.123) |
| Yes | 192 (93.7%) | 12338 (90.4%) | ||
| Precipitated By Other Crime | ||||
| No | 203 (99.0%) | 13224 (96.9%) | 0.098 | 3.231 (0.878, 26.951) |
| Yes | 2 (1.0%) | 421 (3.1%) | ||
| Other Crime In Progress | ||||
| No | 204 (99.5%) | 13543 (99.3%) | 1.000 | 1.536 (0.267, 61.570) |
| Yes | 1 (0.5%) | 102 (0.7%) | ||
| Terrorist Attack | ||||
| No | 205 (100.0%) | 13645 (100.0%) | - | |
| Suicide Attempt History | ||||
| No | 149 (72.7%) | 11229 (82.3%) | < 0.001*** | 0.572 (0.417, 0.796) |
| Yes | 56 (27.3%) | 2416 (17.7%) | ||
| Suicide Intent Disclosed | ||||
| No | 157 (76.6%) | 10644 (78.0%) | 0.611 | 0.922 (0.662, 1.306) |
| Yes | 48 (23.4%) | 3001 (22.0%) | ||
| Suicide Note | ||||
| No | 114 (55.6%) | 9525 (69.8%) | < 0.001*** | 0.542 (0.407, 0.724) |
| Yes | 91 (44.4%) | 4120 (30.2%) | ||
| Recent Criminal Legal Problem | ||||
| No | 199 (97.1%) | 12494 (91.6%) | 0.002** | 3.055 (1.375, 8.444) |
| Yes | 6 (2.9%) | 1151 (8.4%) | ||
| Legal Problem (Other) | ||||
| No | 197 (96.1%) | 13261 (97.2%) | 0.390 | 0.713 (0.351, 1.687) |
| Yes | 8 (3.9%) | 384 (2.8%) | ||
| Physical Health Problem | ||||
| No | 149 (72.7%) | 10808 (79.2%) | 0.030* | 0.698 (0.509, 0.970) |
| Yes | 56 (27.3%) | 2837 (20.8%) | ||
| Job Problem | ||||
| No | 177 (86.3%) | 12231 (89.6%) | 0.133 | 0.731 (0.487, 1.136) |
| Yes | 28 (13.7%) | 1414 (10.4%) | ||
| Finance Problem | ||||
| No | 191 (93.2%) | 12362 (90.6%) | 0.275 | 1.416 (0.820, 2.647) |
| Yes | 14 (6.8%) | 1283 (9.4%) | ||
| School Problem | ||||
| No | 205 (100.0%) | 13592 (99.6%) | 1.000 | Inf (0.207, Inf) |
| Yes | 0 (0.0%) | 53 (0.4%) | ||
| Recent Suicide (Friend / Family) | ||||
| No | 199 (97.1%) | 13402 (98.2%) | 0.188 | 0.601 (0.268, 1.675) |
| Yes | 6 (2.9%) | 243 (1.8%) | ||
| Death (Friend / Family / Other) | ||||
| No | 189 (92.2%) | 12894 (94.5%) | 0.163 | 0.688 (0.410, 1.235) |
| Yes | 16 (7.8%) | 751 (5.5%) | ||
| Crisis Recently | ||||
| No | 198 (96.6%) | 13249 (97.1%) | 0.672 | 0.845 (0.399, 2.144) |
| Yes | 7 (3.4%) | 396 (2.9%) | ||
| Other Addiction | ||||
| No | 205 (100.0%) | 13566 (99.4%) | 0.634 | Inf (0.313, Inf) |
| Yes | 0 (0.0%) | 79 (0.6%) | ||
| Family Relationship | ||||
| No | 186 (90.7%) | 12561 (92.1%) | 0.438 | 0.845 (0.523, 1.441) |
| Yes | 19 (9.3%) | 1084 (7.9%) | ||
| Death Abuse | ||||
| No | 205 (100.0%) | 13638 (99.9%) | 1.000 | Inf (0.022, Inf) |
| Yes | 0 (0.0%) | 7 (0.1%) | ||
| Abused as Child | ||||
| No | 203 (99.0%) | 13509 (99.0%) | 1.000 | 1.022 (0.274, 8.581) |
| Yes | 2 (1.0%) | 136 (1.0%) | ||
| Fight Between Two People | ||||
| No | 205 (100.0%) | 13574 (99.5%) | 0.629 | Inf (0.281, Inf) |
| Yes | 0 (0.0%) | 71 (0.5%) | ||
| Stalking | ||||
| No | 204 (99.5%) | 13639 (100.0%) | 0.099 | 0.090 (0.011, 4.148) |
| Yes | 1 (0.5%) | 6 (0.0%) | ||
| Prostitution | ||||
| No | 205 (100.0%) | 13643 (100.0%) | 1.000 | Inf (0.003, Inf) |
| Yes | 0 (0.0%) | 2 (0.0%) | ||
| Walk By Assault | ||||
| No | 205 (100.0%) | 13645 (100.0%) | - | |
| Suicide Thought History | ||||
| No | 130 (63.4%) | 9768 (71.6%) | 0.012* | 0.688 (0.513, 0.929) |
| Yes | 75 (36.6%) | 3877 (28.4%) | ||
| Eviction or Loss Of Home | ||||
| No | 198 (96.6%) | 13200 (96.7%) | 0.842 | 0.954 (0.451, 2.417) |
| Yes | 7 (3.4%) | 445 (3.3%) | ||
| Traumatic Anniversary | ||||
| No | 203 (99.0%) | 13568 (99.4%) | 0.327 | 0.576 (0.152, 4.875) |
| Yes | 2 (1.0%) | 77 (0.6%) | ||
| Disaster Exposure | ||||
| No | 205 (100.0%) | 13633 (99.9%) | 1.000 | Inf (0.041, Inf) |
| Yes | 0 (0.0%) | 12 (0.1%) | ||
| Number of Crises | 0.35 (0.67) | 0.39 (0.67) | 0.330 | |
| Any Crisis | ||||
| No | 152 (74.1%) | 9444 (69.2%) | 0.147 | 1.276 (0.925, 1.783) |
| Yes | 53 (25.9%) | 4201 (30.8%) | ||
Discussion
We failed to detect significance in nurses vs. the overall general population, explainable due to the higher proportion of females amongst nurses. (Norton & Divine, 2015) More importantly, when taking gender into account, both male and female nurses were significantly greater risk of suicide than the gender and aged matched general population. The limitations on the data imposed by gender could be overcome by a larger sample size. The 2015 NVDRS dataset increases from 18 states to 40 states. We plan to replicate this study with that larger sample.
The findings amongst nurses are congruent with our previous pilot of nurse suicides in San Diego in that female nurses were higher than the female general population to complete suicide in this same year, (Davidson, Stuck, et al., 2018) and given the larger sample size the difference was statistically significant in this study.
Perhaps even more clinically meaningful than the findings related to the relative incidence of suicide among nurses, this investigation begins to inform us of methods of suicide and risks associated with suicide that are specific to the profession of nursing. The findings are congruent with our previous San Diego pilot reporting nurses most commonly use pharmaceutical poisoning while physicians and the general population most commonly complete suicide by firearms. (Davidson, Stuck, et al., 2018) We hope in the future to be able to use qualitative research methods coupled with natural language processing to analyze the law enforcement and medical examiner free text fields for a more complete analysis of suicide risks. That significantly more nurses had reported previous mental health treatment and previous attempts and were still successful at completing suicide suggests that mental health issues may have been undertreated. This is congruent with our findings from the pilot of a nurse suicide prevention program where high risk nurses, including those with previous attempts, reported continued depression and suicidal ideation despite past treatment. (Davidson, Zisook, Kirby, DeMichele, & Norcross, 2018)
Manually coding the occupation, though at first seen as a limitation, provided information about the types of nurses who completed suicide that would not otherwise have been available. If the dataset had been used as designed we would only have known that the suicide had been completed by a nurse. From the free text descriptions it is clear that in 2014 in these states nurse anesthetists and retired nurses were at higher risk than other nurses (Electronic Supplement 2). These potentially important yet unanticipated findings bear further exploration with larger, hypothesis -driven studies.
Nurses completed suicide more commonly by poisoning using pharmaceuticals and other substances, followed closely by firearms, where all others completed suicide most commonly by firearms. This is similar to our San Diego findings. (Davidson, Stuck, et al., 2018) There has been debate whether the use of medications in healthcare clinician suicide represents risks associated with access in the workplace vs. knowledge of lethality of medications. The substances identified as cause of death in nurse suicide contain a combination of medications used only in the hospital setting (e.g: succinylcholine, phenobarbital) as well as those commonly found in the home (e.g: oxycodone, ibuprofen). This differs from the San Diego pilot, where the drugs implicated in cause of death were universally medications available in the home. (Davidson, Stuck, et al., 2018) Benzodiazepines, antihistamines and opioids were the most common agents implicated in cause of nurse death, validating the need to address substance use, pain, and mental health issues that may lead to medication misuse within the workforce. The dangers of concomitant use of benzodiazepines and opioids have been previously published ("Benzodiazepines and Opioids. National Institute on Drug Abuse," 2018; Dasgupta et al., 2016; Gomes, Mamdani, Dhalla, Paterson, & Juurlink, 2011) and this analysis suggests that healthcare clinicians are not immune to toxicity from these agents. Interestingly, common substances of abuse such as alcohol, marijuana or illicit substances were not found in this study of nurses This quantitative analysis taken without context cannot answer the question of access vs. knowledge at this time. We plan a more in-depth analysis of this data in a future study pairing the qualitative data provided by law enforcement and medical examiners with the coded demographic variables to explore substance use in healthcare clinician suicide in greater detail.
Limitations
Before fully embracing this study’s findings, certain caveats must be considered. First, this is a one year cross-section of data. Suicide rates fluctuate from year to year. Due to the fluctuation in incidence of suicide from year to year a longitudinal analysis is planned and warranted. Second, the dataset contained suicides from only 18 states, and may reflect geographic variation. As data become available the study will be replicated with more states represented. Third, the gender demographic data for nursing was derived from the best available sampling methodology but could be improved with actual workforce data. Finally, annual detection of occupation codes was a limitation. We retained ‘retired’ clinicians in the counts because nurses are known to keep their licenses into retirement, and the denominator of nurses was ‘actively licensed’. It is possible that this could have biased the results in either direction. It would not have been possible to cleanly remove retired nurses from either the NVDRS or the demographic dataset. We determined that the risk of bias was less by leaving them included than the risk of inaccurately removing them.
Conclusion
In conclusion, of the professions studied, nurses were at higher risk than the age and gender matched controls to complete suicide. Nurses were more likely to complete suicide by pharmaceutical poisoning, the general population most commonly used firearms. Confidence in results would be improved if workforce gender data were available in nursing to conduct between gender-specific analyses. Results suggest that both female and male nurses may be at risk of suicide as has been found internationally. Future research is planned with a larger sample and longitudinal study design.
Supplementary Material
Highlights.
Female nurse suicides were significantly higher than in the female population
Male nurse suicides were significantly higher than the male population
Benzodiazepines and opioids were the most common substances used in nurse suicide
Future research and suicide preventative strategies for nurses are needed
Acknowledgements:
We would like to thank Mr. Soo Yong Lee, Luann Eberhardt RN, and Nadia Bukar RN for their support identifying denominator data.
The project described was partially supported by the National Institutes of Health, Grant UL RR031980 The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.” Partial funding was also provided by the Gamma Gamma chapter of Sigma Theta Tau.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributor Information
Judy E. Davidson, University of California San Diego Health, Research Scientist University of California San Diego School of Medicine, Department of Psychiatry.
James Proudfoot, Clinical Translational Research Institute, University of California San Diego.
Kelly Lee, University of California San Diego, Skaggs School of Pharmacy and Pharmaceutical Sciences.
Sidney Zisook, University of California San Diego School of Medicine, Department of Psychiatry.
References
- The 10 Professions With the Highest Suicide Rates. (2016, October/24/2014). The Richest. Retrieved from http://www.therichest.com/rich-list/the-biggest/the-10-professions-with-the-highest-suicide-rates/?view=all
- Aasland OG, Ekeberg Ø, & Schweder T (2001). Suicide rates from 1960 to 1989 in Norwegian physicians compared with other educational groups. Social science & medicine, 52(2), 259–265. [DOI] [PubMed] [Google Scholar]
- Alderson M, Parent-Rocheleau X, & Mishara B (2015). Critical Review on Suicide Among Nurses. Crisis, 1–11. doi: 10.1027/0227-5910/a000305 [DOI] [PubMed] [Google Scholar]
- Alston MH (1986). Occupation and suicide among women. Issues Ment Health Nurs, 8(2), 109–119. [DOI] [PubMed] [Google Scholar]
- Andersen K, Hawgood J, Klieve H, Kolves K, & De Leo D (2010). Suicide in selected occupations in Queensland: evidence from the State suicide register. Australian and New Zealand Journal of Psychiatry, 44(3), 243–249. doi: 10.3109/00048670903487142 [DOI] [PubMed] [Google Scholar]
- Andrew LB (2015). Physician Suicide. Retrieved from http://emedicine.medscape.com/article/806779-overview
- Benzodiazepines and Opioids. National Institute on Drug Abuse. (2018). Retrieved from https://www.drugabuse.gov/drugs-abuse/opioids/benzodiazepines-opioids
- Budden JS, Moulton P, Harper KJ, Brunell M, & Smiley R (2016). The 2015 national nursing workforce survey. Journal of Nursing Regulation, 7(1), S1–S90. [Google Scholar]
- Budden JS, Zhong EH, Moulton P, & Cimiotti JP (2013). Supplement: The National Council of State Boards of Nursing and the Forum of State Nursing Workforce Centers 2013 National Workforce Survey of Registered Nurses. J Nurs Regul, 4(2). [Google Scholar]
- Charlton J, Kelly S, Dunnell K, Evans B, & Jenkins R (1993). Suicide deaths in England and Wales: trends in factors associated with suicide deaths. POPULATION TRENDS-LONDON-, 34–34. [Google Scholar]
- Cheung T, & Yip PS (2015). Depression, Anxiety and Symptoms of Stress among Hong Kong Nurses: A Cross-sectional Study. Int J Environ Res Public Health, 12(9), 11072–11100. doi: 10.3390/ijerph120911072 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cheung T, & Yip PS (2016). Self-harm in nurses: prevalence and correlates. Journal of Advanced Nursing. doi: 10.1111/jan.12987 [DOI] [PubMed] [Google Scholar]
- Community Engagement Kit. (2017). Mental Health. Retrieved from http://www.who.int/mentalhealth/suicide-prevention/en/
- Dasgupta N, Funk MJ, Proescholdbell S, Hirsch A, Ribisl KM, & Marshall S (2016). Cohort study of the impact of high-dose opioid analgesics on overdose mortality. Pain medicine, 17(1), 85–98. [DOI] [PubMed] [Google Scholar]
- Davidson JE, Stuck AR, Zisook S, & Proudfoot J (2018). Testing a Strategy to Identify Incidence of Nurse Suicide in the United States. Journal of Nursing Administration, 48(5), 259–265. doi: 10.1097/nna.0000000000000610 [DOI] [PubMed] [Google Scholar]
- Davidson JE, Zisook S, Kirby B, DeMichele G, & Norcross W (2018). Suicide Prevention: A Healer Education and Referral Program for Nurses. Journal of Nursing Administration. doi: 10.1097/nna.0000000000000582 [DOI] [PubMed] [Google Scholar]
- Davis M, Detre T, Ford DE, Hansbrough W, Hendin H, Laszlo J, … Michels R (2003). Confronting depression and suicide in physicians: a consensus statement. Jama, 289(23), 3161–3166. [DOI] [PubMed] [Google Scholar]
- Feskanich D, Hastrup JL, Marshall J, Colditz G, Stampfer M, Willett WC, & Kawachi I (2002). Stress and suicide in the Nurses' Health Study. Journal of Epidemiology and Community Health, 56(2), 95–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Frank E, Biola H, & Burnett CA (2000). Mortality rates and causes among US physicians1. American journal of preventive medicine, 19(3), 155–159. [DOI] [PubMed] [Google Scholar]
- Gaither C, Schommer J, Doucette W, Kreling D, & Mott D (2015). 2014 National Pharmacist Workforce Survey. [DOI] [PubMed]
- Gomes T, Mamdani MM, Dhalla IA, Paterson JM, & Juurlink DN (2011). Opioid dose and drug-related mortality in patients with nonmalignant pain. Archives of Internal Medicine, 171(7), 686–691. [DOI] [PubMed] [Google Scholar]
- Hawton K, Agerbo E, Simkin S, Platt B, & Mellanby RJ (2011). Risk of suicide in medical and related occupational groups: A national study based on Danish case population-based registers. J Affect Disord, 134(1-3), 320–326. doi: 10.1016/j.jad.2011.05.044 [DOI] [PubMed] [Google Scholar]
- Hawton K, Clements A, Simkin S, & Malmberg A (2000). Doctors who kill themselves: a study of the methods used for suicide. Qjm, 93(6), 351–357. [DOI] [PubMed] [Google Scholar]
- Hawton K, Simkin S, Rue J, Haw C, Barbour F, Clements A, … Deeks J (2002). Suicide in female nurses in England and Wales. Psychol Med, 32(2), 239–250. [DOI] [PubMed] [Google Scholar]
- Hawton K, & Vislisel L (1999). Suicide in nurses. Suicide and Life-Threatening Behavior, 29(1), 86–95. [PubMed] [Google Scholar]
- Increase in suicide rates, 1999–2014. (2016, April 22, 2016). NCHS Data Brief No. 241, April 2016.
- Katz RM (1983). Causes of death among registered nurses. Journal of Occupational Medicine, 25(10), 760–762 763p. [DOI] [PubMed] [Google Scholar]
- Kelly S, & Bunting J (1998). Trends in suicide in England and Wales, 1982–96. Population trends, 92(Summer), 29–41. [PubMed] [Google Scholar]
- Kolves K, & De Leo D (2013). Suicide in medical doctors and nurses: an analysis of the Queensland Suicide Register. Journal of Nervous & Mental Disease, 201(11), 987–990. doi: 10.1097/nmd.0000000000000047 [DOI] [PubMed] [Google Scholar]
- Last N, Tufts E, & Auger LE (2017). The Effects of Meditation on Grey Matter Atrophy and Neurodegeneration: A Systematic Review. J Alzheimers Dis, 56(1), 275–286. doi: 10.3233/jad-160899 [DOI] [PubMed] [Google Scholar]
- Lieberman J (2016, May 13, 2016). When Rome burns: Addressing rising suicide rates. Medscape Psychiatry. Retrieved from http://www.medscape.com/viewarticle/862900?nlid=105028_424&src=WNL_mdplsfeat_160517_mscpedit_psyc&uac=46298AT&spon=12&impID=1099957&faf=1
- Milner A, Spittal MJ, Pirkis J, & LaMontagne AD (2013). Suicide by occupation: systematic review and meta-analysis. The British Journal of Psychiatry, 203(6), 409–416. [DOI] [PubMed] [Google Scholar]
- Milner AJ, Maheen H, Bismark MM, & Spittal MJ (2016). Suicide by health professionals: a retrospective mortality study in Australia, 2001–2012. Medical Journal of Australia, 205(6), 260–265. [DOI] [PubMed] [Google Scholar]
- Norton HJ, & Divine G (2015). Simpson's paradox… and how to avoid it. Significance, 12(4), 40–43. [Google Scholar]
- Organizations, J. C. o. A. o. H., & Hospital, J. C. A. (2007). Comprehensive Accreditation Manual for Hospitals: The Official Handbook: Camh: Joint Commission Resources. [Google Scholar]
- Pharmacy is Right for Me. (2017). Retrieved from https://pharmacyforme.org/why-pharmacy-may-be-right-for-you/career-pathways/
- Roberts S, Jaremin B, & Lloyd K (2013). High-risk occupations for suicide. Psychol Med, 43(6), 1231–1240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schernhammer E (2005). Taking their own lives—the high rate of physician suicide. N Engl J Med, 352(24), 2473–2476. [DOI] [PubMed] [Google Scholar]
- Shanafelt TD, Hasan O, Dyrbye LN, Sinsky C, Satele D, Sloan J, & West CP (2015). Changes in burnout and satisfaction with work-life balance in physicians and the general US working population between 2011 and 2014. Paper presented at the Mayo Clin Proc. [DOI] [PubMed] [Google Scholar]
- Stack S (1982). Suicide: A decade review of the sociological literature. Deviant Behavior, 4(1), 41–66. [Google Scholar]
- Stack S (2001). Occupation and suicide. Social Science Quarterly, 82(2), 384–396. [Google Scholar]
- Suicide. (2018). Mental Health. Retrieved from http://www.who.int/mentalhealth/suicide-prevention/en/
- Tramutola K (2015). Suicide Among Nurses (Vol. 45, pp. 1–1 2/3p). Trenton, New Jersey: New Jersey State Nurses Association. [Google Scholar]
- World Health Organization. (2014). Preventing Suicide: A global imperative. Retrieved from http://www.who.int/mental_health/suicide-prevention/world_report_2014/en/
- Yaghmour NA, Brigham TP, Richter T, Miller RS, Philibert I, Baldwin DC Jr, & Nasca TJ (2017). Causes of death of residents in ACGME-accredited programs 2000 through 2014: implications for the learning environment. Academic Medicine, 92(7), 976. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Young A, Chaudhry HJ, Pei X, Halbesleben K, Polk D, & M, D. (2015). A census of actively licensed physicians in the United States, 2014. Journal of Regulatory Medicine, 101(2), 8–23. [Google Scholar]
- Young A, Chaudhry HJ, Pei X, Halbesleben K, Polk DH, & Dugan M (2015). A census of actively licensed physicians in the United States, 2014. Journal of Medical Regulation, 101(2), 7–22. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.