Weighing the Association Between BMI Change and Suicide Mortality

Abstract

Objective:

Suicide rates continue to rise, necessitating identification of novel risk factors to prevent suicide. Obesity and suicide mortality have been examined, but no studies have examined associations between weight change, suicide mortality, and depression among adults in the U.S.

Methods:

People who died by suicide from 2000-2015 with a recorded height and weight in the first and second 6 months preceding their death (“index date”; n = 387) were extracted from a larger sample from the Mental Health Research Network; each was matched with five people in the control group by age, sex, index year, and healthcare site (n = 1,935). Data were analyzed in 2020.

Results:

People who died by suicide were predominantly male (71%), Caucasian (69%), depressed (55%), middle aged (M = 57 years), with chronic health issues (57%), [Control: 66% Caucasian; M age = 56 years; 43% chronic health issues, 14% depressed]. T-test analyses showed BMI change within a year before the index date differed between those who died by suicide (M = −0.72±2.42 kg/m²) and the control group, (M = 0.06±4.99 kg/m²) p < 0.001, d = .17. A 1-unit decrease in BMI was associated with 9% increased risk of suicide after adjusting for demographics, mental disorders, and Charlson score (aOR = 0.91, 95% CI = 0.87-0.95, p < 0. 001). BMI change and suicide were associated for those without depression (p < 0.001).

Conclusions:

Weight loss in the year before suicide was associated with increased suicide mortality after accounting for physical and mental health indicators.

Introduction

Suicide ranks tenth among all-cause mortality in the United States, with suicide mortality on the rise within recent years.¹ Mental health conditions, including depression, are associated with suicide mortality, and severe depression nearly doubles the odds of suicide death.² As weight loss or weight gain comprise part of the diagnostic criteria for major depressive disorder, there is the potential for weight change to be associated with suicide mortality.³ Indeed, odds of weight or appetite loss among individuals with major depressive disorder who died by suicide was nearly 2.6 times that of depressed controls⁴ (i.e., those without weight/appetite loss). Other research, however, has not found an association between weight loss and suicide mortality during a 5-year follow up among depressed individuals.⁵ Of note, these studies only included individuals with major depression. The need for research examining weight change and suicide mortality among larger, more representative samples is supported by data that only 24% of those who died by suicide were diagnosed with a mental health condition in the month prior to their death.⁶ Therefore, there is a need to identify risk factors for suicide beyond psychiatric diagnoses; non-biased (e.g., objective) physiological indicators, including weight change, may present one such novel risk factor for suicide mortality.

Only one study to date has examined associations between weight change and suicide mortality among a large sample of men in London, United Kingdom.⁷ In this study of 18,784 men over a 38-year period, there were 61 reported suicides. Unexplained weight loss over the past year was associated with an elevated suicide risk. However, the study was limited in that unexplained weight loss was self-reported and therefore may be biased, only men were included in the study, diagnoses of depression were not reported, and the study was conducted outside the United States. Indeed, there is a marked difference in suicide means between individuals in the United Kingdom and United States, such that less lethal means are more commonly used in the United Kingdom.⁸ Therefore, results may not be generalizable to both men and women in the United States.

Further evidence for a potential link between weight change and suicide comes from the field of bariatric surgery. Among those who undergo bariatric surgery, which contributes to significant reductions in body weight, rates of suicide mortality are significantly higher than in the general population.⁹ Interestingly, however, obesity has been associated with decreased risk for suicide in several but not all studies.¹⁰ Thus, there may be an effect of weight loss following bariatric surgery that elevates risk for suicide. While existing evidence links specialized populations (e.g., depressed individuals, those who have undergone bariatric surgery) with suicide mortality, there is a need to further examine whether weight change is associated with variation in risk of suicide mortality among the general population, including those with and without depression and other potential risk factors for suicide mortality. This study contributes to the literature by examining these associations while also accounting for the influence of demographic variables, mental health or substance use diagnoses including and excluding depression, and medical conditions which may also be associated with the risk of suicide mortality.

Methods

Study Sample

The sample for this study was derived from a larger case-control project examining health care services utilization in the year prior to suicide.⁶ Patients were members of eight large Mental Health Research Network (MHRN)-affiliated healthcare systems located across the United States in the states of Colorado, Georgia, Hawaii, Massachusetts, Minnesota, Michigan, Oregon, and Washington. All patients had insurance coverage provided by the health systems’ affiliated health plans, and all patients had been health plan members for at least 10 months during the year prior to the index date. The combination of both health plan and health system membership data allowed comprehensive capture of both electronic health records and insurance claims data from within and outside of the healthcare system. Health plan membership included commercial, Medicaid, Medicare, and self-pay insurance plans, representing a diverse range of plans. Beginning with the full sample of individuals who died by suicide (n = 3,117), we included only those who had vital records data for height and weight in the 0-6 months and in the 7-12 months preceding their death (n = 387). Using date of suicide as the index date, the individuals who died by suicide were then matched with five controls each based on their age, sex, index year, and healthcare system site. A total sample of 2,322 patients were identified. Data were collected from January 2000- September 2015 and analyzed in 2020. Institutional review board approval was obtained at each health system site.

Suicide mortality was identified using International Statistical Classification of Diseases and Related Health Problems (ICD)-10 codes (X60-X84, Y87.0) from official government mortality records matched to health system records, using the definition and coding process developed by the Centers for Disease Control. ICD-9 diagnosis codes were extracted from the MHRN’s Virtual Data Warehouse (VDW), a federated data model in each health system that includes a combination of electronic health record and insurance claims organized in the same format using the identical variables and data definitions across sites. In addition, all data (e.g., encounters, pharmacy fills, diagnoses, demographics, vital signs) are organized using the same definitions across sites and are quality checked locally.¹¹

The heights and weights were measured during patients’ clinical visits. Patients might have had multiple measurements during the year before the index date. The change in BMI was calculated by subtracting the closest measure to the index date (0-6 months) from the furthest measure in the 7-12 months prior to the index date. Mental health and substance use diagnoses were captured using ICD-9 codes 290 to 319. Demographic information (e.g., age, sex, race/ethnicity) was extracted from the VDW. For each person, estimated neighborhood income and neighborhood education were derived from geocoded addresses and census block data. Finally, the Charlson comorbidity index provides a measure of health burden by weighting key indicators of conditions associated with post-hospitalization one-year mortality.¹² The Charlson scale is used to identify medical comorbidity such as cancer or other chronic diseases in this study. The Charlson scale was calculated by using ICD-9 codes from clinical diagnoses that occurred in the year prior to the index date. Patients were categorized as either having (1) or not having (0) a co-morbid medical illness.

Statistical Analyses

T-tests and chi-square analyses were conducted to examine descriptive differences between people who died by suicide and people in the control group on all demographic, psychiatric, and chronic illness characteristics, as well as differences in BMI across varying time points of the period preceding the index date. Conditional logistic regression was used to explore the relationship between the change in BMI over the earliest and latest recorded weights in the first and second 6 months preceding the index date and suicide death, including univariable model and multivariable model with adjustment for the variables of age, sex, race/ethnicity, neighborhood poverty level, neighborhood educational level, mental health disorders, and Charlson score. We also specified the linear combinations of BMI change and depression diagnosis as well as BMI change and Charlson score to test the significance of the interaction effects. Analyses were conducted using SAS Institute Inc., Version 9.4, Cary, NC.¹³ The mean change in BMI between people who died by suicide and people in the control group for those patients with depression and without depression were demonstrated separately in Figure 1 with the corresponding 95% Confidence Interval.

Figure 1.

Mean change in BMI between those who died by suicide and those in the control group by depression diagnosis

Results

Demographic characteristics of the sample are presented in Table 1. As age and sex were matched, the sample was predominantly male and approximately 56 to 57 years old. In general, people who died by suicide were Caucasian (68%), had depression (55%) or another mental or substance use disorder diagnosis (23%) and a Charlson score > 0 (57%), reflective of co-morbid medical illnesses. The people who died by suicide and people in the control group differed significantly by race, mental/substance use disorder diagnosis, and co-morbid medical illness. There were no significant differences between the groups in neighborhood poverty level or neighborhood education. In examining change in BMI during the previous year between the people who died by suicide and people in the control group, there was a significant difference between the groups, as those who died by suicide lost an average of 0.72 kg/m² whereas those in the control group, gained an average of 0.06 kg/m² comparing the furthest BMI measurement to the BMI measurement closest to the index date (p < 0.001); the Cohen’s d effect size for this association was d = 0.17. None of the individuals who died by suicide received bariatric surgery in the year prior to their death, and one individual in the control group had received bariatric surgery in the previous 1 year.

TABLE 1.

Demographic characteristics of study sample^a

Characteristic	Study Participants				P
	Individuals who died by suicide (N=387)		Individuals in the control group (N=1,935)
	N	%	N	%
Age (M±SD)	57.18±19.90		55.99±19.80		-
Gender					-
Female	111	28.7	555	28.7
Male	276	71.3	1380	71.3
Race					<0.001
White	265	68.5	1281	66.2
Non-White	46	11.9	410	21.2
Unknown	76	19.6	244	12.6
Neighborhood Poverty Level					0.862
Higher than Low Income	284	73.4	1444	74.6
Low Income	33	8.5	162	8.4
Missing	70	18.1	329	17.0
Neighborhood Education					0.684
≥25% college grad	236	61.0	1223	63.2
<25% college grad	135	34.9	631	32.6
Missing	16	4.1	81	4.2
Co-morbid medical illness b					<0.001
Charlson score = 0	165	42.6	980	50.7
Charlson score > 0	221	57.1	833	43.1
Mental/substance use disorderc					0.004
No	298	77.0	1608	83.1
Yes	89	23.0	327	16.9
Depression diagnosis					<0.001
No	174	45.0	1661	85.8
Yes	213	55.0	274	14.2
BMI Difference (M±SD) d	−0.72±2.42		0.06±4.99		<0.001

^aMeans were compared by t tests, and proportions were compared by chi-square tests.

^bThe Charlson Index was calculated using ICD-9 codes from clinical diagnoses. Total score is categorized as a binary variable

^cICD-9: 290-319 excluding depression diagnoses (ICD-9 codes: 2962, 2963, 296.82, 2980, 3004, 301.12, 3090, 3091, 309.28, 311)

^dSubtracting BMI measured on the earliest date from the BMI measured on the latest date prior to the index date

Note: Index date refers to date of suicide death

Appendix 1 includes a demographics table comparing individuals in our sample to those from which the study sample was drawn. Our sample was older and had a higher percentage of females and White individuals, as well as higher levels of medical morbidity, mental/substance use disorder diagnoses, and diagnoses of depression than those in the full sample. However, the individuals in our sample and in the full sample did not differ by neighborhood education level or neighborhood poverty level.

Univariable analyses revealed a significant association between BMI change, race, mental/substance use disorder diagnosis, and co-morbid medical illness and our main outcome, suicide mortality (Table 2). After accounting for demographic predictors, mental/substance use disorder diagnoses excluding depression, depression, and co-morbid medical illness, the association between BMI change from the latest to most recent BMI measurement prior to the index date and suicide mortality remained significant (aOR = 0.91, 95% CI = 0.87-0.95, p < 0.001), such that a 1-unit decrease in BMI change was associated with 9% increased risk of suicide mortality. The association between BMI change and risk of suicide mortality was different between people with and without depression. Among those with depression diagnosis, there was no significant association between BMI change and suicide mortality (p = 0.098). In contrast, among those without depression, there was a significant association between increased BMI change and decreased risk of suicide mortality (p < 0.001). This effect is displayed in Figure 1. In order to assess whether weight loss may be due to the presence of medical conditions, we tested the interaction between Charlson score and BMI change. Results showed there was a significant association between BMI change for both those with and without medical comorbidity (Table 2). This suggests that the association between BMI change and suicide is not restricted to individuals with a serious medical condition.

TABLE 2.

Conditional Logistic Regression for BMI difference on suicide death

	Univariate Model			Multivariate Model a
Variable	OR	95% CI	p	OR	95% CI	p
BMI Change	0.93	0.9–0.96	<0.001	0.91	0.87–0.95	<0.001
White (reference: Non-White)	1.81	1.29–2.52	0.001	1.79	1.2–2.65	0.004
Living in a poor neighborhood	1.04	0.7–1.54	0.861	1.08	0.67–1.75	0.741
Living in a neighborhood with lower college graduation	1.11	0.88–1.4	0.382	1.02	0.78–1.35	0.878
Co-morbid medical illness	1.66	1.31–2.1	<0.001	1.56	1.2–2.05	0.001
Mental/substance use disorder diagnosis excluding depression	1.48	1.13–1.93	0.004	3.86	2.76–5.41	<0.001
Depression diagnosis	7.69	6.03–9.79	<0.001	11.65	8.63–15.72	<0.001
Depression diagnosis * BMI Change
One unit increase in BMI change for patients with depression				0.92	0.82–1.03	0.163
One unit increase in BMI change for patients without depression				0.89	0.81–0.98	0.016
Co-morbid medical illness * BMI Change
One unit increase in BMI change for patients with Co-morbid medical illness				0.90	0.85–0.96	0.002
One unit increase in BMI change for patients without Co-morbid medical illness				0.89	0.81–0.98	0.016

^aAdjusting for all listed variables in the table, plus index year, site, and days difference between BMI measurement

Discussion

This is the largest study of people who died by suicide in the United States based on medical records among the general population. We found that documented weight loss in the year prior is linked with suicide mortality after accounting for demographic, medical, and psychiatric predictors. Consistent with statistics on suicide completion,^1,14 having a mental/substance use diagnosis and co-morbid medical illness independently increased odds of suicide mortality. In the current study, those who completed suicide lost approximately ¾ BMI point on average in the year prior to their death, which translates to roughly a four-pound weight loss. However, it should be considered that from 1992-2010, the BMI of U.S. adults who were initially normal weight increased approximately 0.7 to 1.1 pounds and among those who were initially overweight approximately 0.4 to 1.3 pounds.¹⁵ In addition, those in the control group gained weight, on average, during the same time period. Thus, the observed trend in weight loss among those who died by suicide contrasts with both those in the control group and population trends of increasing BMI in the United States. Additionally, the direction of the results such that losing weight increases suicide mortality is consistent with research that BMI and suicide risk are inversely associated; Kaplan, McFarland, Huguet ¹⁶ noted that each 5 kg/m² increase in BMI was associated with lower suicide risk by up to 24%.

The primary findings demonstrated that suicide mortality is associated with weight loss in the 1-year before suicide death, with a 7% increased risk of suicide mortality for each 1-unit decrease in BMI following adjustment for relevant covariates. Once adjusted for demographic, psychiatric, and medical conditions, there was a 9% increased risk of suicide mortality for each 1-unit decrease in BMI, although the magnitude of the effect can be considered small. Notably, this association does not appear to be accounted for by weight changes attributable to chronic medical conditions, as BMI change was associated with increased risk of suicide mortality regardless of Charlson score. Weight and height are routinely measured at medical visits, making this information highly visible to medical providers.¹⁷ Measured vitals from which BMI can be calculated provide a value which is not subject to bias with personal reporting of weight and height.¹⁸ Suicide screening has been shown to predict future suicide attempts and suicide death,¹⁹ illustrating the importance of routine screening, especially among individuals with weight loss.

Despite depression being most strongly associated with suicide mortality of all the predictors examined, only 55% of individuals who completed suicide were diagnosed with depression, and only 23% had a mental or substance use disorder other than depression. Interestingly, results showed a significant association between weight change and suicide mortality only among those without a diagnosis of depression. One rationale for this result is that depression can contribute to weight gain or weight loss, which, when considered together across individuals with depression, could obfuscate a potential association between suicide mortality and weight change. Additionally, consistent with previous research, antidepressant medications may attenuate weight change as they can contribute to both weight loss and weight gain.²⁰ However, the non-significant interaction between weight change and depression diagnosis should not be interpreted to mean that weight change is not pertinent among those with depression. Rather, it is possible that weight loss may confer risk for depression.

The present evidence suggests an association between weight loss and suicide mortality, although the causality cannot be ascertained. One potential is the relationship is direct, such that losing weight increases risk for suicide mortality. For example, hormones including ghrelin may be responsible for the association between weight loss and suicide mortality, as is posited among those who have undergone bariatric surgery.⁹ Alternately, it is possible that weight loss is a marker for the ‘true’ cause. As such, weight change may be an ‘externalized’ indicator of depressive symptoms or other mental health symptoms. Research has also found lower levels of triglycerides among those who attempted suicide,²¹ and triglyceride levels are shown to decrease with modest weight loss in overweight and obese individuals.²²

The strengths of this study include the large sample of individuals who died by suicide and vital measurements derived from reliable medical record data from health systems across the United States. The results inform clinical care by shedding light on the relevance of weight loss as a potential psychiatric marker for suicide. Additionally, findings can be used to inform future studies which utilize predictive modeling of risk for suicide mortality.

Limitations

In interpreting the study findings, it is important to recognize that people who died by suicide and people in the control group were included if they had at least two weight measurements in the year prior to the index date. However, the ability to determine the lapsed time between the measured weights could vary significantly across patients. Indeed, we found greater variability in change in BMI as the number of elapsed days between weight measurements increased, which is to be expected. Thus, future efforts to expand this study would benefit from using a prospective study design to capture weight measurements at clinically-relevant, potentially vulnerable time points for suicide mortality across the life span, such as within 3-months of a mental health emergency department visit or mental health inpatient stay.²³ It is important to also consider that because we only included those with 2 or more healthcare visits in the previous year, these individuals may be more engaged in their health which therefore may have biased the findings. Indeed, those included in our sample had higher levels of medical and psychological comorbidity. Additionally, this study did not examine suicidal ideation or suicide attempts, which are associated with suicide mortality² as well as with weight control behaviors²⁴ and perceived weight status.²⁵ Examining associations among objective weight change, suicidal ideation, and suicide attempts would be a valuable future research direction. Additionally, identifying whether weight loss was intentional or unintentional could have implications for risk and subsequent treatment, as demographic risk factors including older age, poorer health status, and lower BMI are mutually associated with both suicide mortality and unintentional weight loss.²⁶ Lastly, this study did not include antidepressant medication use, which could have contributed to weight changes, or consider particular medical conditions which could contribute to edema, which has a small positive correlation with weight.²⁷ Future studies should consider the potential degree of influence of particular medical conditions on weight, as well as measuring variables associated with antidepressant use, including medication type, dosage, and adherence, for their association with weight changes.

Conclusions

This study demonstrated that weight change is a physiological indictor which may be associated with increased risk of suicide mortality, and among those without a diagnosis of depression, there was an association between BMI change and suicide mortality. Unfortunately, existing research has demonstrated the ability to accurately capture an individual’s risk for suicide mortality above chance level is poor.²⁸ However, the findings of the current study suggest that providers, in addition to considering other clinically-relevant information, should consider that a patient losing weight may be a potential marker for suicide mortality, especially among high-risk patients. Although it is unknown whether rate of weight loss (e.g., abrupt versus gradual) is pertinent, intervening when weight loss is observed across two healthcare visits may facilitate a referral for behavioral health treatment if needed and potentially prevent future suicide.

Highlights:

1. In the last year, those who died by suicide lost weight and controls gained weight

2. Decrease in body mass index was linked with increased risk of suicide

3. Change in body mass index, suicide, and not having depression were associated