Types and clinical outcomes of chemical ingestion in emergency departments in South Korea (2011-2016).

OBJECTIVE: This study aims to provide basic data on the types and frequency of chemical ingestions and the clinical outcomes of chemical ingestion injury.
METHODS: This study retrospectively analyzed the data obtained from the Emergency Department-Based Injury In-depth Surveillance of the Korea Centers for Disease Control and Prevention (South Korea) from 2011 to 2016. Patients ingesting chemicals aged ≥ 18 years were included, but those ingesting unknown chemical substances or with unknown clinical outcomes were excluded.
RESULTS: This study included 2,712 (47.2% were men and 52.8% were women, mean age, 47.05 years) patients ingesting chemicals. Unintentional and intentional ingestions were reported in 1,673 (61.7%) and 1,039 (38.3%), respectively. The most commonly ingested chemical substances were hypochlorites, detergents, ethanol, and acetic acid. In the unintentional ingestion group, the most common chemicals upon admission were hypochlorites (74), glacial acetic acid (60), and detergent (33). The admission rates were 60% for glacial acetic acid, 58.3% ethylene glycol, and 30.4% other alkali agents. In the intentional ingestion group, the most common chemicals upon admission were hypochlorites (242), glacial acetic acid (79), ethylene glycol (42), and detergent (41). The admission rates were 91.9% for glacial acetic acid, 87.5% ethylene glycol, 85.7% potassium cyanide, and 81.4% hydrochloric acid. In total, 79 deaths (10 unintentional ingestions, 69 intentional ingestion) were reported, and glacial acetic acid had an odds ratio of 9.299 for mortality.
CONCLUSION: We compared the intentional and unintentional ingestion groups, and analyzed the factors affecting hospital admission and mortality in each group. The types and clinical outcomes of chemical ingestion varied depending on the purpose of chemical ingestion. The findings are considered beneficial in establishing treatment policies for patients ingesting chemicals.

Introduction

With the advancement of manufacturing and industry, approximately 120,000 chemicals are being distributed worldwide, with approximately 40,000 of them being distributed in South Korea [1]]. Moreover, as the distribution of chemicals increases, exposure to these chemicals also increases. In the United States, an estimated 10,000 chemical exposures are reported annually. occur each year [[2,3]. During chemical exposure, both the exposure and the victims should be properly monitored. Exposure events are primarily monitored by the departments responsible for managing chemicals, and data collection from the victims is hospital- or poison center-based [4-7]]. In South Korea, data regarding chemical exposures are limited to the exposure events, such as chemical accidents or chemical exposures at the workplace [1,8-10] In this study, we aim to provide basic data on chemical ingestion, including the types of chemicals and frequency of ingestion and clinical outcomes of chemical ingestion. Therefore, we compared the characteristics of chemical ingestion according to their intentionality and analyzed the factors that affect the poor prognosis in each group.

Materials and methods

Setting and data collection

This retrospective observational study was conducted with the approval of the institutional review board of Ewha Womans’ University Mok-dong hospital (IRB No.2019-08-006). The informed consent for gathering the data related to the emergency department (ED) visit was obtained from all participating patients or patients’ guardians by the Korea Centers for Disease Control and Prevention (KCDC) according to national research committee. This study analyzed the data from the Emergency Department-Based Injury In-depth Surveillance under the supervision of the KCDC from 2011 to 2016. The Emergency Department-Based In-depth Surveillance is a data collection system that proactively collects data on basic epidemiology, treatment, and outcomes of injured patients presenting to the ED. Implemented in 2006, the collection system involved 20 agencies from 2011 to 2014 and 23 agencies since 2015. Each hospital employs personnel who are responsible for data entry and quality control and who are regularly trained and supervised by the KCDC.

In this study, the definition of chemical ingestion was as follows: consumption of an artificial toxic substance (according to the toxic substance classification) that caused poisoning and injury in patients who visited the ED. All study subjects were aged older than 18 years. Excluded cases were subjects aged younger than 17 years and subjects with missing information on intentionality or ED clinical outcomes.

Outcome measures

According to the injury surveillance system, in cases of poisoning, the trained personnel must classify the substance according to code and enter the proper name (general name or product name). In this study, two emergency physicians and one chemist identified and categorized the main components of the ingested chemicals. All patients ingesting chemicals were classified into either the intentional ingestion group or the unintentional ingestion group, and comparison and subgroup analysis were performed between the two groups. Sex, age, mode of ED arrival, and insurance type were compared as the general features. The analyzed characteristics related to injury were as follows: period from the time of injury to ED presentation, place where the injury occurred, activity at the time of injury, alcohol ingestion before injury, date of the visit, time of the visit, and duration of ED stay. The results of ED treatment (discharge from ED, admission to general ward (GW) or intensive care unit [ICU]) and mortality were used to compare the severity of injuries.

Statistical analysis

We compared and analyzed the general and injury-related characteristics between the intentional and unintentional ingestion groups. The mean with standard deviation and median with 25th and 75th percentiles were expressed as the continuous variables. The number of patients and percentage were expressed as the noncontinuous variables. For items requiring statistical verification, an independent T-test and Mann-Whitney U test were used for the continuous variables. A chi-squared test was used for the noncontinuous variables. Using cross-analysis and average comparison, we selected the variables that showed significant differences between the two groups, and based on the researchers’ discretion, we selected the variables considered as factors for hospital admission and mortality. Multiple logistic regression using backward stepwise selection (likelihood ratio) was performed, and the remaining variables with p-value less than 0.05 were listed in the table. The statistics program Statistical Package for the Social Sciences Statistics for Windows version 21 (International Business Machines Corporation, Armonk, NY, USA) was used, and 95% confidence intervals were considered statistically significant.

Results

From 2011 to 2016, a total of 4,741 cases of chemical ingestion were reported. After applying the exclusion criteria, this study included 2,712 patients ingesting chemicals (1,247 patients aged younger than 18 years, 644 patients with missing data for ED treatment results, 138 patients with missing data for intentionality). A total of 52.8% of the patients were women. The mean age was 47.05 years. The unintentional ingestion and intentional ingestion groups were composed of 1,673 (61.7%) and 1,039 patients (38.3%), respectively.

The comparison results of the unintentional and intentional ingestion groups are presented in Table 1. The proportion of women in the intentional ingestion group (66.3%) was higher than that in the unintentional ingestion group (44.5%). Chemical ingestion was reported in homes in 88.8% of patients in the intentional ingestion group and in 53% of patients in the unintentional ingestion group (p<0.0001). The activity at the time of injury was described as “other(unspecified)” in 95.7% of patients in the intentional ingestion group, while daily living activity was noted in 48.2% of patients in the unintentional ingestion group. Alcohol was ingested before the injury in 43.1% of patients in the intentional ingestion group, a percentage significantly higher than the 12.6% in the unintentional ingestion group. ED treatment results showed that 79.3% of patients in the unintentional ingestion group were discharged from the ED. Patients in the intentional ingestion group had significantly larger proportions of GW admission, ICU care, and mortality in the ED than the unintentional ingestion group. The total number of deaths from chemical ingestion was 79, comprising 10 from the unintentional ingestion group and 69 from the intentional ingestion group.

Table 1

General characteristics of chemical ingestion between unintentional and intentional ingestion groups.

	Unintentional		Intentional		Total		P value
	N	%	N	%	N	%
No. of patients	1673	61.7	1039	38.3	2712	100
Sex
Male	929	55.5	650	33.7	1279	47.2	<0.0001
Female	744	44.5	689	66.3	1433	52.8
Age (yrs, mean±SD)	47.33±16.91		46.51±17.48		47.05±17.24		0.226
Age distribution in 10 years							<0.0001
20–30	332	19.8	210	20.2	542	20.0
31–40	315	18.8	234	22.5	549	20.2
41–50	314	18.8	202	19.4	516	19.0
51–60	350	20.9	185	17.8	535	19.7
61–70	179	10.7	64	6.2	243	9.0
71–80	121	7.2	99	9.5	220	8.1
≥81	62	3.7	45	4.3	107	3.9
Mode of arrival							0.420
Walk-in (include car, foot, etc.)	1266	75.7	817	78.6	2083	76.8
119	238	14.2	132	12.7	370	13.6
Private ambulance	162	9.7	86	8.3	248	9.1
Police	0	.0	1	.1	1	.0
Air	4	.2	1	.1	5	.2
Others	1	.1	1	.1	2	.1
Unknown	2	.1	1	.1	3	.1
Time interval from injury to ED visit (hrs, median, 25%, 75%)	1 (0, 5)		1 (0, 4)		1 (0, 5)		0.002a
Insurance							0.812
National health insurance	1410	84.3	887	85.4	2297	84.7
Vehicle	190	11.4	112	10.8	302	11.1
Medicaid beneficiary	30	1.8	20	1.9	50	1.8
Self-pay (uninsured)	35	2.1	17	1.6	52	1.9
Others	8	.5	3	.3	11	.4
Where the poisoning occurred							<0.0001
House	887	53.0	914	88.0	1801	66.4
Residential facility	19	1.1	17	1.6	36	1.3
Medical facility	28	1.7	21	2.0	49	1.8
School, education facility	57	3.4	3	.3	60	2.2
Sport facility	9	.5	1	.1	10	.4
Road	36	2.2	9	.9	45	1.7
Transportation area except road	3	.2	6	.6	9	.3
Factory, industrial facility	291	17.4	4	.4	295	10.9
Farm	32	1.9	2	.2	34	1.3
Amusement, cultural public facility	26	1.6	7	.7	33	1.2
Commercial facility	238	14.2	42	4.0	280	10.3
Outdoor, river, sea	24	1.4	9	.9	33	1.2
Other	1	.1	1	.1	2	.1
Unknown	22	1.3	3	.3	25	.9
Activity							<0.0001
Work	434	25.9	0	.0	434	16.0
Unpaid labor	214	12.8	3	.3	217	8.0
Education	34	2.0	0	.0	34	1.3
Exercise	6	.4	0	.0	6	.2
Leisure	123	7.4	1	.1	124	4.6
Daily living activity	806	48.2	41	3.9	847	31.2
Hospital treatment	7	.4	0	.0	7	.3
Travel	1	.1	0	.0	1	.0
Other	33	2.0	994	95.7	1027	37.9
Unknown	15	.9	0	.0	15	.6
Alcohol ingestion before injury							<0.0001
No	1313	78.5	487	46.9	1800	66.4
Yes	210	12.6	448	43.1	658	24.3
Unknown	150	9.0	104	10.0	254	9.4
Day of presentation							0.095
Weekday (Mon-Thu)	913	54.6	601	57.8	1514	55.8
Weekend (Fri-Sun)	760	45.4	438	42.2	1198	44.2
Time of presentation							<0.0001
Day	527	31.5	280	26.9	807	29.8
Evening	810	48.4	390	37.5	1200	44.2
Night	336	20.1	369	35.5	705	26.0
ED stay (hrs, mean±SD)	40.50 ±108.17		24.21±63.34		33.79±92.91		<0.0001
ED treatment result							<0.0001
Discharge	1327	79.3	482	46.4	1809	66.7
General ward	261	15.6	328	31.6	589	21.7
Intensive care unit	83	5.0	194	18.7	277	10.2
Death	2	.1	35	3.4	37	1.4
Hospital mortalityb	10	0.60	69	6.60	79	2.90	<0.0001

aMann-Whitney test

bHospital mortality included ED death.

The subgroup analysis of patients ingesting chemicals was performed according to intentionality. Regarding the agents ingested in 1,673 patients in the unintentional ingestion group, hypochlorite-related agents were most common (410), followed by detergents, ethanol, toluene, sodium hydroxide, and glacial acetic acid. The most common chemicals upon admission in the unintentional ingestion group were hypochlorites in 74 patients, glacial acetic acid in 60, and detergent in 33. The admission rates were as follows: 60% for glacial acetic acid, 58.3% ethylene glycol, and 30.4% other alkali agents. Ten deaths were noted in the unintentional ingestion group due to glacial acetic acid (3), other hydrocarbons (2), hypochlorite-related agent (1), and detergent (1), toluene (1), methanol (1), and hydrochloric acid (1) (Table 2). The different characteristics between the admitted and discharged patients in the unintentional ingestion group were compared and are listed in Table 3. The proportion of men in the admission group was significantly higher than that in the discharged group, and 23.04% of men and 17.74% of women were admitted. The mean age of the admission group was higher (54.01 years) than that of the discharge group (45.59 years). Regression analysis of risk factors for admission in the unintentional ingestion group revealed that male sex and increased age were significant factors. Among the ingested chemicals, ethylene glycol, glacial acetic acid, sulfuric acid, methanol, and other acidic substances significantly increased the risk for admission. Only consumption of hydrofluoric acid was associated with decreased risk of admission (Table 4).

Table 2

Common agents of unintentional and intentional chemical ingestion in adult at ED, South Korea, 2011–2016.

Unintentional				Intentional
Chemical	Total	Admission	Death	Chemical	Total	Admission	Death
	(n = 1673)	(n = 346)	(n = 10)		(n = 1039)	(n = 557)	(n = 69)
Hypochlorites related agents	410	74	1	Hypochlorites related agents	514	242	9
		18.0%	0.2%			47.1%	1.8%
Detergents (soaps-anionic and nonionic)	209	33	1	Detergents (soaps-anionic and nonionic)	121	41	6
		15.8%	0.5%			33.9%	5.0%
Ethanol	154	21	(-)	Glacial acetic acid	86	79	33
		13.6%	0.0%			91.9%	38.4%
Toluene	119	19	1	Ethylene glycol	48	42	1
		16.0%	0.8%			87.5%	2.1%
Sodium hydroxide	100	23	(-)	Hydrochloric acid	43	35	6
		23.0%	0.0%			81.4%	14.0%
Glacial acetic acid	100	60	3	Ethanol	41	10	1
		60.0%	3.0%			24.4%	2.4%
Other hydrocarbon	76	18	2	Sodium hydroxide	39	26	3
		23.7%	2.6%			66.7%	7.7%
Hydrofluoric acid	66	3	(-)	Potassium Cyanide	21	18	3
		4.5%	0.0%			85.7%	14.3%
Hydrogen peroxide	37	4	(-)	Methanol	15	10	(-)
		10.8%	0.0%			66.7%	.0%
Sulfuric acid	35	2	(-)	Toluene	15	9	1
		5.7%	0.0%			60.0%	6.7%
Methanol	29	11	1	Sodium percarbonate	13	3	(-)
		37.9%	3.4%			23.1%	0.0%
Hydrochloric acid	28	6	1	Other hydrocarbon	10	6	2
		21.4%	3.6%			60.0%	20.0%
Ethylene glycol	24	14	(-)	Acetone	9	1	(-)
		58.3%	0.0%			11.1%	0.0%
Other alkali agent	23	7	(-)	Hydrogen peroxide	8	3	(-)
		30.4%	0.0%			37.5%	0.0%

Table 3

Comparison of general characteristics between discharge and admission groups in unintentional and intentional chemical ingestion.

	Unintentional							Intentional
	Discharge		Admission		Total		p-value	Discharge		Admission		Total		p-value
	N	%	N	%	N	%		N	%	N	%	N	%
No. of patients	1327	79.3	346	20.7	1673	100		482	46.4	557	53.6	1039	100
Sex
Male	715	53.9	214	61.8	929	55.5	0.008	150	31.1	200	35.9	350	33.7	0.104
Female	612	46.1	132	38.2	744	44.5		332	68.9	357	64.1	689	66.3
Age (yrs, mean±SD)	45.59±16.14		54.01±18.11		47.33±16.91		<0.0001	42.56±15.16		49.94±18.61		46.51±17.48		<0.0001
Time interval from injury to ED visit(hrs, median, 25%,75%)	1 (0, 5)		2 (0, 7)		1 (0, 5)		0.008*	1 (0, 4)		1 (0, 5)		1 (0, 4)		<0.0001a
Alcohol ingestion before injury							0.180							<0.0001
No	1047	78.9	266	76.9	1313	78.5		187	38.8	300	53.9	487	46.9
Yes	157	11.8	53	15.3	210	12.6		254	52.7	194	34.8	448	43.1
Unknown	123	9.3	27	7.8	150	9.0		41	8.5	63	11.3	104	10.0
Time of presentation							0.005							<0.0001
Day	393	29.6	134	38.7	527	31.5		108	22.4	172	30.9	280	26.9
Evening	661	49.8	149	43.1	810	48.4		173	35.9	217	39.0	390	37.5
Night	273	20.6	63	18.2	336	20.1		201	41.7	168	30.2	369	35.5
ED stay (hrs)	48.18±120.29		11.02±0.77		40.50±108.17		<0.0001	39.61±90.63		10.89±0.86		24.21±63.34		<0.0001

aMann-Whitney test

Table 4

Multivariate analysis of unintentional and intentional chemical ingestion for admission.

	Multivariate				Univariate
	OR	95% C.I. for EXP(B)		Sig.	OR	95% C.I. for EXP(B)		Sig
		Lower	Upper			Lower	Upper
Unintentional
Sex: male	1.370	1.057	1.776	0.017	1.388	1.089	1.768	0.008
Age	1.025	1.017	1.032	<0.0001	1.030	1.022	1.037	<0.0001
Ethylene glycol	5.842	2.522	13.530	<0.0001	5.554	2.445	12.614	<0.0001
Glacial acetic acid	5.568	3.591	8.632	<0.0001	6.750	4.435	10.274	<0.0001
Hydrofluoric acid	0.281	0.087	0.908	0.034	0.175	0.055	0.562	0.003
Sulfuric acid	26.132	3.105	219.890	0.003	23.400	2.808	195.017	0.004
Methanol	3.219	1.464	7.077	0.004	2.388	1.117	5.104	0.025
Other acid agent	11.817	2.062	67.733	0.006	7.749	1.413	42.480	0.018
Intentional
Age	1.019	1.011	1.028	<0.0001	1.026	1.018	1.033	<0.0001
Alcohol ingestion before injury: Yes	0.469	0.355	0.621	<0.0001	0.480	0.374	0.616	<0.0001
Sodium hydroxide	3.159	1.515	6.590	0.002	1.766	0.897	3.477	0.100
Ethylene glycol	17.008	6.840	42.287	<0.0001	6.470	2.726	15.357	<0.0001
Glacial acetic acid	16.064	6.986	36.941	<0.0001	11.215	5.124	24.545	<0.0001
Hydrochloric acid	7.189	3.126	16.532	<0.0001	3.973	1.825	8.650	0.001
Hydrochlorites related agent	1.623	1.170	2.252	0.004	0.593	0.464	0.759	<0.0001
Methanol	3.626	1.180	11.148	0.025	1.744	0.592	5.138	0.313
Potassium cyanide	11.270	3.165	40.130	<0.0001	5.332	1.561	18.214	0.008

The most commonly ingested chemicals of the 1,039 patients in the intentional group were hypochlorite-related agents (514), followed by detergent, glacial acetic acid, ethylene glycol, and hydrochloric acid. A total of 242 patients were admitted following the ingestion of hypochlorites, 79 for glacial acetic acid, 42 for ethylene glycol, and 41 for detergent. The admission rates were as follows: 91.9% for glacial acetic acid, 87.5% ethylene glycol, 85.7% potassium cyanide, and 81.4% hydrochloric acid. Among the 69 deaths in the intentional ingestion group, 33 patients ingested glacial acetic acid, 9 ingested hypochlorites, 6 detergents, and 6 hydrochloric acids (Table 2). The different characteristics of patients admitted and those discharged in the intentional chemical ingestion group were compared (Table 3). The mean age of patients in the admission group was higher (49.94 years) than that of the discharge group (42.56 years). Regression analysis of risk factors for admission in the intentional ingestion group demonstrated significantly increased risk of admission with older age and consumption of sodium hydroxide, ethylene glycol, glacial acetic acid, hydrochloric acid, hypochlorites, methanol, or potassium cyanide (Table 4).

Sixty-nine patients in the intentional ingestion group died, and a comparison of characteristics between patients who survived and those who died revealed an older mean age in the mortality group of 65.86 years compared with the 45.14 years of the survival group (Table 5). Regression analysis was performed to identify risk factors for mortality in the intentional ingestion group, age and glacial acetic acid ingestion significantly increased the risk of death. Alcohol and hypochlorite ingestion were shown to reduce the risk of death (Table 6). To determine the association between glacial acetic acid ingestion and age, the interaction was analyzed, and a significant result (p = 0.07) was observed. Regression analysis was performed by dividing patients into four groups according to age and glacial acetic acid ingestion. The risk of mortality increased to 25.477 times when glacial acetic acid was ingested by a patient younger than 65 years and 80.605 times when ingested by a patient 65 years or older (Supplement 2)

Table 5

General characteristics of intentional chemical ingestion between survival and mortality groups.

	Survival		Mortality		Total		P value
	N	%	N	%	N	%
No. of patients	970	93.4	69	6.6	1039	100
Sex
Male	325	33.5	25	36.2	350	33.7	0.643
Female	645	66.5	44	63.8	689	66.3
Age (yrs, mean±SD)	45.14±16.79		65.86±15.49		46.51±17.48		<0.0001
Time interval from injury to ED visit (hrs, median, 25%, 75%)	1 (0, 5)		1 (0, 2)		1 (0, 4)		0.724
Alcohol ingestion before injury							<0.0001
No	443	45.7	44	63.8	487	46.9
Yes	434	44.7	14	20.3	448	43.1
Unknown	93	9.6%	11	15.9	104	10.0
Time of presentation							<0.0001
Day	247	25.5	33	47.8	280	26.9
Evening	366	37.7	24	34.8	390	37.5
Night	357	36.8	12	17.4	369	35.5
ED stay (hrs, mean±SD)	25.22±65.44		10.00±0.00		24.21±63.34		<0.0001

Table 6

Multivariate analysis of intentional chemical ingestion for mortality.

	Multivariate				Univariate
	OR	95% C.I. for EXP(B)		Sig.	OR	95% C.I. for EXP(B)		Sig.
		Lower	Upper			Lower	Upper
Age	1.055	1.036	1.073	<0.0001	1.067	1.054	1.083	<0.0001
Glacial acetic acid	9.299	4.446	19.448	<0.0001	15.860	9.173	27.422	<0.0001
Hydrochlorites related agent	0.401	0.168	0.960	0.040	0.138	0.068	0.281	<0.0001

Discussion

This study is the first study to analyze of data from patients ingesting chemicals who visited the ED. To identify the clinical features and risk factors of patients ingesting chemicals who visited the ED, we not only conducted a comparison between the intentional and unintentional ingestion groups but also analyzed the factors affecting hospital admission and mortality in each group. Each analysis was necessary because the characteristics of the intentional and unintentional ingestion groups were different. According to this study, hypochlorites and detergent were the most commonly ingested chemicals in both the intentional and unintentional ingestion groups. The next most common substances were ethanol, toluene, sodium hydroxide, and glacial acetic acid in the unintentional ingestion group and glacial acetic acid, ethylene glycol, hydrochloric acid, and ethanol in the intentional ingestion group (Supplement). The 2017 American Association of Poison Control Centers (AAPCC) annual report showed that the most common poisoning categories in the United States, excluding therapeutic drugs, are cleaning substances, cosmetics/personal care products, foreign bodies/toys, plants, bites, chemicals, and fumes/gases/vapors, in that order. More than 10,000 individual substances were reported, including bleach, hypochlorites, other foreign bodies/toys, cream/lotion/makeup, desiccants, miscellaneous essential oils, glow products, toothpaste, deodorants, soaps, and laundry detergent/liquid [4].

In the present study, hypochlorite was the most commonly ingested chemical and had the highest number of hospital admissions. The United States reported a total of 40,302 cases of hypochlorite ingestion in 2017, the second highest frequency of exposure to a single substance [[4]]. Hypochlorite in combination with sodium or calcium is widely used for bleaching or discoloration and rust removal. For home use, the concentration is low (3%-5%), and it can irritate the skin without causing serious burns. On the contrary, the concentration of industrial hypochlorite is approximately 20%, which can cause severe corrosive induction and may also result in pulmonary irritation [[11]]. Hypochlorite was an admission risk factor in the intentional ingestion group in this study, but it was also associated with decreased risk of mortality considering that fatal side effects are influenced by the substance’s concentration.

Among the admitted patients in this study, the second most commonly ingested substance was glacial acetic acid, widely used for industrial purposes. At concentrations of 10–25%, it acts as an irritant, but at concentrations of 25% or more, it can cause corrosive injury. Skin contact can cause burns or blisters, and mucosal exposure can cause tissue damage [12]]. Hence, glacial acetic acid in concentrations greater than 20% is classified and managed as a toxic substance in several countries. However, in South Korea, after the removal of heavy metals from industrial glacial acetic acid, it is allowed for use in food products as vinegar. Thus, glacial acetic acid in concentrations greater than 99% can be easily purchased at large supermarkets and grocery stores [13,14]]. Notably, glacial acetic acid was the fourth most commonly ingested substance in 2,712 patients ingesting chemicals in the present study. South Korea has a higher incidence of glacial acetic acid ingestion compared to other countries. For example, in the United States, a total of 1,068,976 cases of nontherapeutic drug exposures were reported for the whole of 2017, and only 7347 of these were due to acid exposure. Among them, the most commonly ingested acid was hydrochloric acid (1,839 cases) followed by hydrofluoric acid (676 cases) [[4]]. Considering that Koreans can easily access glacial acetic acid, accidental exposure is common, regardless if precautionary measures are provided by various media outlets. However, for individuals who are planning to intentionally ingest highly toxic materials, provision of this information is dangerous. Among patients ingesting chemicals who present to an ED in South Korea, glacial acetic acid ingestion is a risk factor of admission and mortality when a chemical is ingested intentionally. In the future, glacial acetic acid must be carefully managed, or its 100% concentrated formulation must be prohibited.

Additionally, ethylene glycol and methanol were identified as the risk factors for admission in both the unintentional and intentional ingestion groups in the present study. Ingestion of ethylene glycol and methanol has a serious prognosis because their metabolites, produced by alcohol dehydrogenase, have toxic effects, including metabolic acidosis, kidney injury (ethylene glycol), and visual loss (methanol). The 2017 AAPCC reported one case of methanol exposure resulting in death [[4]]. The administration of ethanol as an antidote should be performed for patients who ingest ethylene glycol or methanol [[15]].

This study has several limitations. First, detailed types of chemicals were not classified during the data collection phase. If more detailed information was obtained during the collection stage, the data would be more reliable. Second, insufficient information on the amounts and concentrations of the chemicals is a limitation of this study. In the future, the amount of ingestion should be included to the injury survey registry of the KCDC. Third, information on underlying diseases that could affect patient prognosis was insufficient. Fourth, the data colleting system, based on the ED, does not include all patients ingesting chemicals. Patients who did not receive hospital treatment or who used an outpatient clinic were excluded in the study.

Among patients aged older than 18 years who visited the ED due to chemical ingestion from 2011 to 2016, the most commonly ingested substances were hypochlorites, detergents, ethanol, and glacial acetic acid, and the type of substances ingested differed depending on presence or absence of intentionality. High-concentration glacial acetic acid, a chemical substance that is commercially available in South Korea, is used in cooking, resulting in a significant number of glacial acetic acid ingestions, and is a risk factor for admission and mortality in the intentional ingestion group. The result of this study are considered beneficial in determining treatment policies for patients ingesting chemicals who presented to the ED and in managing and preventing chemical exposures.

Total incidence, admission and mortality of chemical ingestion in adult (>20 years) at ED, South Korea, 2011–2016.

(DOCX)

Click here for additional data file.

Univariate analysis of glacial acetic acid and age for mortality in intentional chemical ingestion.

(DOCX)

Click here for additional data file.

13 Jan 2020

PONE-D-19-33190

Types and Clinical Outcomes of Chemical Ingestion in the Emergency Department: South Korea, 2011-2016

PLOS ONE

Dear Hee Lee,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

ACADEMIC EDITOR:

This manuscript is not presented in a standard English, I suggest an English native speaker revision.

Statistical analysis was not well performed.

The main study aim were not clear.  I couldn't understand what’s the definition of intentional vs. unintentional groups.

There are many comments from reviewers, and conflict between them.

I ask you to produce a very well performed revision in order to publish this paper.

Especially check the statistical corrections suggested, and try to reduce the tables, focusing on most important data or transferring them into supplementary materials.

We would appreciate receiving your revised manuscript by February 3rd. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Martina Crivellari

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have reported significance probabilities of 0 in places. Since p=0 is not strictly possible, please correct this to a more appropriate limit, eg 'p<0.0001'.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

4. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The retrospective study reported the major types of chemical poisoning agents in patients who visited ED in South Korea from 2011 to 2016 and accessed the associations between hospital admission/mortality and potential risk factors including intentions of chemical ingestion, patients’ demographics characteristics, time from injury to ED visits, ED LOS, date, time, places of injury occurrence, and so on. My review/comments are from statistical perspective.

Study design:

Lines 74 – 75: the authors mentioned “all patients were classified into either intentional group or unintentional group.” What’s the definition of intentional vs. unintentional groups? Did patients in intentional group had psychiatric problems? Were those patients susceptible to suicide? The sample of intentional group maybe bias by only including patients who presented at ED.

The main study aim were not clearly mentioned. In the materials and methods section, it looks like the authors want to access the association between admitted/discharged and intentions. But, in the results section the authors reported results of parallel analyses on intentional group and unintentional group separately. I recommend the authors setup the primary aims and the corresponding hypotheses before analyzing the data.

Statistical Analysis:

The author didn’t clearly illustrate covariates used in multiple logistic regression. Did the authors perform any variable selection algorithms?

Period from time of injury to ED presentation may have crucial clinical meaning with respect to the primary endpoint (admitted/discharged). The measurement accuracy for intentional group maybe bias because of the potential bias in sampling.

In table 4, the author listed ingested chemicals with significant p-values in model. Did the author include single covariate (e.g. chemical agents) with all types of chemicals as different categories? If so, which chemical was defined as the reference level? Or did the author include all types of chemicals as dummy binary covariates? Is it possible that one patient ingested more than one chemicals?

In table 4, the authors reported “sulfuric acid” was statistically significant in the multivariable logistic regression and had a very large odds ratio and a huge upper limit (26.132 [3.105, 219.890]). It is confusing me that, in table 2, the admission rates of ingested “sulfuric acid” and “hydrofluoric acid” were 5.7% (n = 2) and 4.5% (n = 3) respectively. But “sulfuric acid” has such a huge odds ratio (OR=26.132) and “hydrofluoric acid” has odds ratio (OR=0.281) less than 1.

In addition, among 1673 unintentional ingestion cases only 35 (2.09%) ingested “sulfuric acid”. A quasi complete separation may occur and vague/suspected findings can be drawn.

Same issues in intentional group (both discharged vs. admitted and survival vs. mortality) as unintentional group.

I recommend authors pay more attention to interpret results of multivariable logistic regressions and move characteristics comparisons from result section to the discussion or supplementary section. For categorical characteristics with more than two categories, it is suspected and tedious to report p-values of Chi-square tests of each category without multiple adjustment. I recommend to perform one single Chi-square test for categorical characteristics and report a single p-value.

Additional comments:

Line 100 – 101: the authors reported three age groups with p-values < 0.05. Those p-values were separately calculated for each sub-group without showing literature evidence of associations between age groups and intentions of chemical ingestion. I recommend to exclude this sentence or do a multiple comparison adjustment in order to avoid erroneous inferences.

Table 1: how to explain different ED LOS?

Lines 121 – 122: the sentence is confused. The authors would like to report the top three unintentional chemical ingestions for admitted patients. In order to claim potential associations between admission and type of chemicals, admission rate would be better. Admission rates of ingesting hypochlorites (18.0%) and detergent (15.8%) were less than 20%. However, admission rates of ingesting ethylene glycol was 58.3%.

Lines 122 – 124: Why the authors ignored three death cases by ingesting toluene, methanol, and hydrochloric acid?

Lines 126 – 127: I recommend to report: the admission rate for male were statistically significantly higher than it for female. Report both admission rates.

Lines 127: Report both mean ages if it has clinical meanings.

Line 128: collapse insignificant levels and run a single Chi-square test to test association between age groups and admitted/discharged.

Table 4: it should be multivariable analysis.

Reviewer #2: General

Detailed and easy to understand article.

However, the main claims of the paper are of interest for South-Korea, mainly.

The data presented are less numerous than those already reported in the literature.

There are no data about the severity of outcome; the exposure concentrations / exposure doses.

There is no real novelty in comparison with previous literature.

The English style and language should be checked by an English native speaker.

Too many tables. A choice should be made to focus on the most important results. The remaining data could be available as Supplementary Material.

Details

Line 16: why were only ingestions and not inhalation or other routes of exposure also included?

Line 29: “OD” abbreviation: the first time the term is used, write it in full, please.

Line 32: “Differences in clinical outcomes by kinds of chemical, and intentionality” : these conclusions have already been reported in the literature

Line 33: “… study as a basis for treatment policies etc…”: policies already exist; e.g. Slaughter R’s publication in 2019 about the toxicology of sodium hypochlorite.

Lines 47-52: the aim of the article is relevant mainly for South-Korea.

Line 61: how many agencies are there in total in South-Korea? Are 20 or 23 agencies representative of the whole country? What is the served population in South-Korea?

Line 67: how many cases were excluded due to missing information on intentionality? And to missing on ED outcome?

Line 73: “categorized”: according to which classification?

Lines 76-77: “insurance type were compared etc…” why? What is the aim?

Line 79: “alcohol ingestion before injury…”: any data about the dose?

Line 89-90: “a logistic regression analysis was performed to determine the risks of hospital admission and mortality”: separate outcomes or composite outcome?

Line 90: “mortality”: in-hospital mortality?

Line 91: “p values less than 0.05”: use 95% Confidence Intervals instead of p values (more informative).

Line 94: (4741-2712)/4741 = 43%. Please, indicate the reason for exclusion of these 43% of cases.

Lines 109-111: “patients in the intentional group had significantly larger proportions of general ward admission, ICU care, and mortality /…/ than the unintentional ingestion group”: is this result really new ? Several articles have already reported that most fatalities after chemical ingestion in adults are intentional (e.g. CDC data from North America).

Table 1: “Age distribution in 10 years”, line “81”: add “≥” so as to get “≥ 81”

Table 1: “Mode of arrival”: what is the interest of this information? What is the aim?

Table 1: “Time interval from injury to ED visit (hrs)”: values between 12 and 13 hrs = very long ! and huge variation; what is the reason?

Table 1: “Insurance”: what is the interest of this information? Probably country dependent and difficult to generalize.

Table 1: “Place”: place of what? (where the poisoning occurred, I suppose?)

Table 4: “Age”: was age entered as a continuous variable?

Table 9: 4 rows i.e. “Non-Elderly, glacial acetic acid (-)” up to “Elderly, glacial acetic acid (+)”: consider including an interaction term “age * glacial acetic acid” in the regression model.

Line180-182: “The purpose of this study… affected the clinical results”: already said.

Lines 187-190: “the 2017 AAPCC annual report… in that order”: this report included all routes of exposures (not only ingestions); so, is it comparable?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

6 Feb 2020

ACADEMIC EDITOR:

This manuscript is not presented in a standard English, I suggest an English native speaker revision.

Statistical analysis was not well performed.

The main study aim were not clear. I couldn't understand what’s the definition of intentional vs. unintentional groups.

There are many comments from reviewers, and conflict between them.

I ask you to produce a very well performed revision in order to publish this paper.

Especially check the statistical corrections suggested, and try to reduce the tables, focusing on most important data or transferring them into supplementary materials.

� This manuscript underwent English editing before submission, but grammatical errors were still observed after editing.

This revised manuscript edited by Editage English editing service.

I attach the certification of editing. Thank you very much.

� The term “unintentional” in this study refers to an accident or assault by others, while “intentional” refers to a patient who attempted self-harm or suicide.

� This study aimed to provide basic data on the types and frequency of chemical ingestions and the clinical outcomes of chemical ingestion injury. Based on several literature reports, intentional poisoning has a poor prognosis on admission and mortality. We compared the intentional and unintentional groups and analyze the factors affecting admission and mortality in each group. Each analysis is considered necessary considering the patients’ different characteristics. I will add these descriptions in the aim and hypothesis of this study.

� Statistical analysis was reflected in the text and tables in consultation with the professor of statistics.

� I have simplified all the tables according based on your suggestion. Thank you for your advice.

� Thank you for your kind review. We have thoroughly checked the manuscript and paid special consideration in addressing your suggestions. I appreciate you. Thank you very much.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

� The text has been revised to fit the template of PLOSOne.

2. We note that you have reported significance probabilities of 0 in places. Since p=0 is not strictly possible, please correct this to a more appropriate limit, eg 'p<0.0001'.

� I modified P=0 to P<0.0001. Thank you very much.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

4. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

� I linked my ORCID ID to my account. Thank you very much.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

� After reviewing the study with a statistician, I amended the statistics or statistical descriptions according to your advices. Thank you very much.

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

� This manuscript underwent English editing before submission, but grammatical errors were still observed after editing.

This Revised manuscript edited by Editage English editing service.

I attach the certification of editing. Thank you very much.

________________________________________

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The retrospective study reported the major types of chemical poisoning agents in patients who visited ED in South Korea from 2011 to 2016 and accessed the associations between hospital admission/mortality and potential risk factors including intentions of chemical ingestion, patients’ demographics characteristics, time from injury to ED visits, ED LOS, date, time, places of injury occurrence, and so on. My review/comments are from statistical perspective.

� Thank you for your kind review. We have thoroughly checked the manuscript and have paid special consideration in addressing your suggestions.

Study design:

Lines 74 – 75: the authors mentioned “all patients were classified into either intentional group or unintentional group.” What’s the definition of intentional vs. unintentional groups? Did patients in intentional group had psychiatric problems? Were those patients susceptible to suicide? The sample of intentional group maybe bias by only including patients who presented at ED.

� The term “unintentional” in this study refers to an accident or assault by others, while “intentional” refers to a patient who attempted self-harm or suicide.

The Korea Centers for Disease Control and Prevention (KCDC) has been prospectively collecting injury-related information nationwide. The KCDC has been receiving real-time Emergency Department-Based Injury In-depth Surveillance information from 23 academic tertiary hospital emergency departments (EDs) nationwide. The KCDC has begun collecting the data of injury and poisoning in 2008 and has also been investigating chemical exposure cases. The KCDC data were examined for intentionality, but the causes of suicide attempts and psychiatric history were not investigated. South Korea had a suicide rate of 24.6 per 100,000 population, the highest among the Organisation for Economic Co-operation and Development (OECD) countries. The Korean Statistical Information Service dataset showed that suicide rates increased as people aged. I believe that it is necessary to investigate the patients’ psychiatric history of suicide/self-harm. The KCDC should investigate the patients’ psychiatric history and causes of self-harm because this information is considered potentially significant for public health.

Based on your suggestion, I also agree that there may be biases when only ED patients are included. However, a specialized center that treats poisoned patients in South Korea has not been established yet, and most of the poisoned patients are frequently encountered in the ED because the primary clinic does not treat these types of patients. Even in general hospitals, poisoned patients are referred to the ED. Most toxicologists in South Korea are emergency physicians. I am also an emergency physician, and I treat a poisoned patient. Nevertheless, depending on the severity of poisoning, there may be cases in which individuals are treated as outpatients. For patients who visit the ED, an association between the injury and chemical ingestion can be easily determined considering that patients are in the acute phase of the injury. Additionally, the severity of the patient in the ED is higher than that of an outpatient, so it is possible to investigate the injury that affects the patient’s poor prognosis. I will include the following as one of the limitations in this study: this study involved patients who visited the ED; therefore, these ED patients may have different characteristics to the generally poisoned patients.

The main study aim were not clearly mentioned. In the materials and methods section, it looks like the authors want to access the association between admitted/discharged and intentions. But, in the results section the authors reported results of parallel analyses on intentional group and unintentional group separately. I recommend the authors setup the primary aims and the corresponding hypotheses before analyzing the data.

� This study aimed not only to compare the intentional and unintentional groups but also to analyze the factors affecting admission and mortality in each group. Based on several literature reports, intentional poisoning has a poor prognosis on admission and mortality. We analyze the factors that affect admission and mortality in each group and compare the intentional and unintentional ingestion groups. Each analysis is considered necessary considering the patients’ different characteristics. I will add these descriptions in the aim and hypothesis of this study.

Statistical Analysis:

The author didn’t clearly illustrate covariates used in multiple logistic regression. Did the authors perform any variable selection algorithms?

� In this study, three multiple logistic regressions were performed. Using cross-analysis and average comparison, we selected the variables that showed significant differences between the two groups, and based on the researchers’ discretion, we selected the variables considered as factors for hospital admission and mortality. Multiple logistic regression using backward stepwise selection (likelihood ratio) was performed, and the remaining variables with p-value less than 0.05 were listed in the table.

Period from time of injury to ED presentation may have crucial clinical meaning with respect to the primary endpoint (admitted/discharged). The measurement accuracy for intentional group maybe bias because of the potential bias in sampling.

� The intentionality of ingestion is determined by a well-trained emergency physician, and the data of this surveillance system are entered based on the physician’s medical records. The study is retrospective, and researchers are also medical staff who are not involved in the study. There is no investigator’s bias for each group.

We recheck the data. We found out that the average value is significantly high, and the standard deviation increases due to the presence of extreme values. The frequency distribution of the injury visit intervals is as follows. Since the data are obtained from 23 hospitals, the value should not be excluded. According to a statistician, it is more appropriate to present the median value due to the characteristics of the data. Therefore, the values in the table were modified based on the median values, and the Mann-Whitney U test was performed. The period from the time of injury to ED presentation was also entered when performing multivariate regression analysis for admission/discharge and mortality and was excluded as a significant factor affecting prognosis.

In table 4, the author listed ingested chemicals with significant p-values in model. Did the author include single covariate (e.g. chemical agents) with all types of chemicals as different categories? If so, which chemical was defined as the reference level? Or did the author include all types of chemicals as dummy binary covariates? Is it possible that one patient ingested more than one chemicals?

�In this study, we reviewed the survey data of all chemical exposure cases registered in the Emergency Department-Based Injury In-depth Surveillance system operated by the KCDC. Information collected through the surveillance system contains the component or product name of the chemical that the patient ingested. Based on this, the ingredients with the largest proportion are classified as the main ingredients. This process was performed by one chemist and two emergency physicians. We used the dummy binary covariates to calculate the odds ratio (OR) for each chemical exposure.

In table 4, the authors reported “sulfuric acid” was statistically significant in the multivariable logistic regression and had a very large odds ratio and a huge upper limit (26.132 [3.105, 219.890]). It is confusing me that, in table 2, the admission rates of ingested “sulfuric acid” and “hydrofluoric acid” were 5.7% (n = 2) and 4.5% (n = 3) respectively. But “sulfuric acid” has such a huge odds ratio (OR=26.132) and “hydrofluoric acid” has odds ratio (OR=0.281) less than 1.

� It is not sulfuric acid but hydrogen sulfide. My sincere apology as there was a mistake in the English translation, and it was not comprehensively reviewed. Thank you for your careful review. A total of seven cases of hydrogen sulfide ingestion in the unintentional group were reported, of which six were hospitalized, showing an 85.7% admission rate. This can be checked in Supplement 1.

In addition, among 1673 unintentional ingestion cases only 35 (2.09%) ingested “sulfuric acid”. A quasi complete separation may occur and vague/suspected findings can be drawn.

Same issues in intentional group (both discharged vs. admitted and survival vs. mortality) as unintentional group.

� Thank you for pointing out the typographical error regarding hydrogen sulfate as previously stated. We wanted to present significantly comprehensive details regarding the type of chemicals ingested and to provide accurate data for patients’ clinical care. In intentional poisoning, sulfuric acid is not commonly observed. It is believed that sulfuric acid is a substance that cannot be easily accessed by common individuals. However, sulfuric acid can be possibly purchased in Korea. Moreover, glacial acetic acid, which is a significantly strong acid, can also be found in convenience stores in Korea. Most unintentional sulfuric acid poisoning is observed in factories.

I recommend authors pay more attention to interpret results of multivariable logistic regressions and move characteristics comparisons from result section to the discussion or supplementary section. For categorical characteristics with more than two categories, it is suspected and tedious to report p-values of Chi-square tests of each category without multiple adjustment. I recommend to perform one single Chi-square test for categorical characteristics and report a single p-value.

� For variables that do not have significant clinical significance, we presented a single p-value.

Additional comments:

Line 100 – 101: the authors reported three age groups with p-values < 0.05. Those p-values were separately calculated for each sub-group without showing literature evidence of associations between age groups and intentions of chemical ingestion. I recommend to exclude this sentence or do a multiple comparison adjustment in order to avoid erroneous inferences.

� A multiple comparison adjustment was not performed in this study. Hence, the sentence was already deleted. Thank you for your suggestion.

Table 1: how to explain different ED LOS?

� According to Table 1, the proportions of general ward admission, intensive care unit care, and death are significantly high in the intentional group. As shown in Tables 3, 6, and 8, the ED length of stay (LOS) is significantly longer in the admission group than in the discharge group, and the ED LOS is significantly longer in the mortality group than in the survivor group. Significantly high rates of admission and death in the intentional group are possibly attributed to longer ED LOS.

Lines 121 – 122: the sentence is confused. The authors would like to report the top three unintentional chemical ingestions for admitted patients. In order to claim potential associations between admission and type of chemicals, admission rate would be better. Admission rates of ingesting hypochlorites (18.0%) and detergent (15.8%) were less than 20%. However, admission rates of ingesting ethylene glycol was 58.3%.

� I agree with your comment. I modified the manuscript showing the high ingestion incidence and high admission rates. The high incidence of admission is associated with the incidence of ingestion, and the high admission rate is possibly associated with chemical toxicity. The results were presented because identifying the frequency of ingestion and managing the chemical and reducing the incidence of chemical ingestion in patients are also significant.

I added the following sentence: “The admission rates were as follows: 60% for glacial acetic acid, 58.3% ethylene glycol, and 30.4% other alkali agents.”

Lines 122 – 124: Why the authors ignored three death cases by ingesting toluene, methanol, and hydrochloric acid?

�We did not describe the information presented in the table because we believed that repeatedly describing this information in the main text would result in redundancy. Nevertheless, we will add a description regarding these substances to avoid misunderstanding. Hence, toluene (1), methanol (1), and hydrochloric acid (1) were added. Thank you for your suggestion.

Lines 126 – 127: I recommend to report: the admission rate for male were statistically significantly higher than it for female. Report both admission rates.

� I added the following sentence in the main text: “23.04% of men and 17.74% of women were admitted.”

Lines 127: Report both mean ages if it has clinical meanings.

�I added the following in the main text: “The mean age of the admission group was higher (54.01 years) than that of the discharge group (45.59 years).”

Line 128: collapse insignificant levels and run a single Chi-square test to test association between age groups and admitted/discharged.

� We believed that presenting the distribution by age group would help in identifying the frequency of poisoning and in establishing preventive measures. A single chi-squared test was performed based on your suggestion.

Table 4: it should be multivariable analysis.

�Multiple logistic regression using backward stepwise selection (likelihood ratio) was performed by selecting variables that showed a significant difference between the two groups using cross-analysis and average comparison. Subsequently, the remaining variables with p-value less than 0.05 were listed in the table. We added the result of the univariate analysis in Table 4.

Reviewer #2:

.

I would like to thank you for your careful review and suggestions on my paper. I have closely reviewed and answered the points you pointed out.

General

Detailed and easy to understand article.

However, the main claims of the paper are of interest for South-Korea, mainly.

� The KCDC has been prospectively collecting injury-related information nationwide. The KCDC has been receiving real-time Emergency Department-Based Injury In-depth Surveillance information from 23 academic tertiary hospital EDs nationwide. The KCDC has begun collecting the data of injury and poisoning in 2008 and has also been investigating chemical exposure cases. There are 36 regional emergency medical centers and 116 local emergency medical centers in Korea. A total of 36 regional emergency medical centers can treat poisoned patients. The KCDC has provided financial support to 23 regional emergency medical centers and maintained a high-quality patient information by sending one coordinator to each hospital. My hospital is one of the 23 hospitals, and we requested and received high-quality information from the KCDC. We studied and analyzed this information retrospectively. Hence, this study aimed to determine the types and clinical outcomes of chemical ingestion in South Korea.

The types and frequencies of chemical exposures may vary from country to country, which is dependent on the country’s development. South Korea’s nationwide information regarding chemical ingestion is considered significant.

I believe it would be more significant, but practically difficult, if some developing countries conduct research together.

Developed countries’ emergency centers, including the Poison Control Center (PCC) in the United States and the Canadian Transport Emergency Center in Canada, operate an addiction management system.

According to the American Chemical Society, approximately 246,000 chemicals are commercially distributed worldwide, with 40,731 chemicals being distributed in Korea and approximately 400 new chemicals entering the Korean domestic market annually.

As the use of chemicals increases, exposure to chemicals or chemical products also increases. Moreover, the types of chemicals are evolving. Accordingly, it is necessary to establish an information system for public and healthcare providers. However, Korea does not have an emergency poison response or control center.

The KCDC collects data regarding the chemical types and exposure routes. Ingestion was the most frequent exposure route, with intentional exposure being the most common.

The authors work in the emergency medical center, teach medical students, and admit toxically poisoned patients. In Korea, these data are not published. Recently, we believe that a study should be conducted on chemical ingestion considering the insufficient information on the frequencies, types, and clinical outcomes of chemical ingestion.

One of the limitations in this study was as follows: a specific amount of chemical ingestion was not assessed in this study.

Hence, it is believed that the patients should be classified by intentionality, the unintentionally poisoned and intentionally poisoned patients.

It is already known that intentional poisoning is more fatal than unintentional poisoning. Korea has the highest suicide rate among the OECD countries, and the number of suicide attempts was 28,278 in 2017, according to the Korea Suicide Prevention Center. It was significant to divide the patients into the unintentional and intentional ingestion groups to assess their clinical results.

Data regarding the types, frequencies, and clinical outcomes of chemical ingestion are considered beneficial in establishing health policies, determining the patient’s prognosis, and educating patients regarding chemical ingestion.

The data presented are less numerous than those already reported in the literature.

� The data analyzed a total of 4,741 chemical ingestions based on the 1,537,617 injured and addicted patients from 2011 to 2016. Although the amount of chemical ingestion may be less than that of the United States or Europe, the data from the 23 nationwide emergency medical centers in a country comprising 50 million people are considered significant.

There are no data about the severity of outcome; the exposure concentrations / exposure doses. There is no real novelty in comparison with previous literature.

� Mortality and length of hospital stay are considered the risk factors for chemical ingestion. Chemical exposure in everyday life is increasing, and the types of chemicals used in industrial products are increasing in developing countries such as Korea. The results regarding chemical ingestion in the past 5 years are considered significant. As previously stated, the absence of a specific amount of chemical ingestion was considered a limitation in this study.

The English style and language should be checked by an English native speaker.

� This manuscript underwent English editing before submission, but grammatical errors were still observed after editing. This revised document was edited by Editage Editing service. (I attach the certification of editing)

Too many tables. A choice should be made to focus on the most important results. The remaining data could be available as Supplementary Material.

� I have simplified all the tables according based on your suggestion. Thank you for your advice.

Details

Line 16: why were only ingestions and not inhalation or other routes of exposure also included?

� Oral ingestion is the most frequent routes of poisoning in the clinical field. In chemical poisoning, ingestion is considered to have a more serious prognosis than other routes of exposure. Ingestion of various chemicals is quickly absorbed by the body, resulting in the uncertainty of prognosis. According to KCDC’s input classification, chemical exposure is divided into solid, liquid, gas, and other unknown classifications. Moreover, exposure routes are divided into ingestion and other exposure routes.

Gas accounted for 20% of all chemical poisonings. Regarding gaseous chemicals, most of the exposure route was inhalation. A total of 90.4% of gas exposure were from carbon monoxide (CO). If gas was reported, it is likely to be in the form of a single substance. Hence, a study assessing chemical ingestion through the oral route was conducted. In addition, CO poisoning has a well-established literature, treatment, and prognosis prediction.

Line 29: “OD” abbreviation: the first time the term is used, write it in full, please.

�I modified the full term and abbreviation. I changed OD to OR.

Line 32: “Differences in clinical outcomes by kinds of chemical, and intentionality” : these conclusions have already been reported in the literature

� We are aware that there are several papers and case reports regarding chemical exposure. Our data are considered beneficial in determining the types of chemicals that are currently poisoning the industrialized countries with a population of 50 million and in assessing the mortality rate. Certainly, according to the reviewer, assessing the poisonous doses and chemical concentrations in the blood is considered significant.

Line 33: “… study as a basis for treatment policies etc…”: policies already exist; e.g. Slaughter R’s publication in 2019 about the toxicology of sodium hypochlorite.

� There are several existing studies and case reports regarding chemical ingestion.

Slaughter R’s publication has conducted a review and case reports regarding sodium hypochlorite for 68 years.

This study also assesses the frequency and mortality of the frequently ingested chemicals.

We believe that this study can provide sufficient information on the recent toxic substances and patients’ clinical prognosis.

Lines 47-52: the aim of the article is relevant mainly for South-Korea.

� Korea is a developing country that has progressed significantly over the past 50 years, recently ranking as top 10 in terms of gross domestic product. The type and frequency of chemical exposures may vary from country to country, which is dependent on the country’s development. South Korea’s nationwide information regarding chemical ingestion is considered significant.

I think it would be more significant, but practically difficult, if some developing countries conduct research together.

Line 61: how many agencies are there in total in South-Korea? Are 20 or 23 agencies representative of the whole country? What is the served population in South-Korea?

� Korea has a total of 51,780,000 individuals living in a small land area.

In Korea, there are 36 regional emergency medical centers (Level 1), 116 local emergency medical centers (Level 2), and approximately 150 local emergency medical rooms. Regional emergency medical centers and some local emergency medical centers can treat poisoned patients.

The KCDC has been receiving real-time Emergency Department-Based Injury In-depth Surveillance information from the 23 academic tertiary hospital EDs (regional emergency medical centers) nationwide. It is not possible to represent all injuries in Korea, but Korea has detailed information regarding injuries. Hence, an in-depth analysis of injured patients is considered significant.

Line 67: how many cases were excluded due to missing information on intentionality? And to missing on ED outcome?

� Children (1247) and individuals with unknown ED results (644) and unknown intentionality (138) were excluded according to the exclusion criteria.

Line 73: “categorized”: according to which classification?

�We used the classification of the 2017 Annual Report of the American Association of PCCs’ National Poison Data System: 35th Annual Report.

We reviewed the data from all chemical exposure cases of the KCDC. Two emergency physicians and one chemist identified and categorized the main components of the ingested chemicals.

Lines 76-77: “insurance type were compared etc…” why? What is the aim?

� South Korea had a suicide rate of 24.6 per 100,000 population, the highest among the OECD countries.

South Korea is a country where national health insurance is available to all individuals. Among them, those who are experiencing financial difficulties receive Medicaid care (the country pays all the healthcare costs).

The analysis was performed because the type of insurance, that is, the economic situation of the patient, is possibly associated with intentional chemical ingestion. However, a statistically significant difference regarding the type of insurance was not observed.

Line 79: “alcohol ingestion before injury…”: any data about the dose?

�Whether the patient has consumed alcohol or not is investigated. Data regarding the amount of alcohol are not provided in this study.

My hospital is one of the 23 hospitals that send information to the KCDC. For patients in the hospital, they do not undergo screening for alcohol use using a complete Alcohol Use Disorders Identification Test, or their blood alcohol concentration (BAC) is not assessed. The KCDC has no available information regarding BAC. Although ideally the BAC levels should be measured, difficulties with privacy, cost, and ethical issues usually arise when measuring the BAC. I instructed the KCDC to measure the patients’ self-reported amount of alcohol (in addition to the BAC level). According to the KCDC, government funding is required to accurately measure the patients’ alcohol consumption.

Line 89-90: “a logistic regression analysis was performed to determine the risks of hospital admission and mortality”: separate outcomes or composite outcome?

�These were considered as separate outcomes for each hospital admission and mortality.

Line 90: “mortality”: in-hospital mortality?

� It was in-hospital mortality.

The table shows that mortality in the ED is defined as mortality in the emergency room during treatment after arriving at the ED. Hospital mortality is defined as any death in the hospital, including ED death.

An explanation regarding this will be provided below.

Line 91: “p values less than 0.05”: use 95% Confidence Intervals instead of p values (more informative).

� I modified this sentence in the Statistics section based on your suggestion. Thank you for your advice.

Line 94: (4741-2712)/4741 = 43%. Please, indicate the reason for exclusion of these 43% of cases.

� Children (1247) and individuals with unknown ED results (644) and unknown intentionality (138) were excluded according to the exclusion criteria.

Lines 109-111: “patients in the intentional group had significantly larger proportions of general ward admission, ICU care, and mortality /…/ than the unintentional ingestion group”: is this result really new? Several articles have already reported that most fatalities after chemical ingestion in adults are intentional (e.g. CDC data from North America).

� Intentional ingestion has a high rate of admission and mortality, even if the patient is addicted to the same substance. It is considered that patients who intentionally ingest chemicals ingest a large amount of chemicals. Although measuring blood levels would be accurate, each chemical does not have an accurate test. According to other reports, intentional poisoning has a high mortality rate without measuring the patients’ blood levels.

As reported in several literatures, intentional addiction has a poor prognosis in admission and mortality. The results of our study were expectedly significant in the intentional ingestion group. We analyze the factors that affect admission and mortality in each group and compare the intentional/unintentional ingestion groups. Each analysis is considered necessary considering the patients’ different characteristics.

I believe that the classifications and clinical outcomes of chemical ingestion in this study are more specific than in other studies.

Table 1: “Age distribution in 10 years”, line “81”: add “≥” so as to get “≥ 81”

� I added “≥.” Thank you for your suggestion.

Table 1: “Mode of arrival”: what is the interest of this information? What is the aim?

�“Mode of arrival” is a variable of the KCDC. 119 (which is equivalent to 911 in North America) is free of charge. The use of 119 is the public cost of the country. The intentional ingestion group was expected to use more 119 due to severe clinical symptoms than the unintentional ingestion group. However, there was no statistical difference between the two groups.

Table 1: “Time interval from injury to ED visit (hrs)”: values between 12 and 13 hrs = very long ! and huge variation; what is the reason?

�We rechecked the data. We found out that the average value is significantly high, and the standard deviation increases due to the presence of extreme values. The frequency distribution of the injury visit intervals was as follows. Since the data are obtained from 23 hospitals, the value should not be excluded. According to a statistician, it is more appropriate to present the median value due to the characteristics of the data. Therefore, the values in the table were corrected based on the median values, and Mann-Whitney U test was performed. The period from the time of injury to ED presentation was also entered when performing multivariate regression analysis for admission/discharge and mortality and were excluded as a significant factor affecting prognosis. I modified the table.

Table 1: “Insurance”: what is the interest of this information? Probably country dependent and difficult to generalize.

� I agree that it is difficult to generalize the insurance types because insurance systems vary from country to country.

South Korea is a country where national health insurance is available to all individuals. Among them, those who are experiencing financial difficulties receive Medicaid care (the country pays all the healthcare costs).

The analysis was performed because the type of insurance, that is, the economic situation of the patient, is possibly associated with intentional chemical ingestion. However, a statistically significant difference regarding the type of insurance was not observed.

Table 1: “Place”: place of what? (where the poisoning occurred, I suppose?)

� I agree with your suggestion. Hence, the phrase was changed to “where the poisoning occurred.”

Table 4: “Age”: was age entered as a continuous variable?

� Age was entered as a continuous variable.

Table 9: 4 rows i.e. “Non-Elderly, glacial acetic acid (-)” up to “Elderly, glacial acetic acid (+)”: consider including an interaction term “age * glacial acetic acid” in the regression model.

� I performed an analysis using age * glacial acetic acid and mentioned the p-value in the text. Moreover, the table containing “Elderly, glacial acetic acid” was moved to the supplementary table. Thank you for your suggestion.

Line180-182: “The purpose of this study… affected the clinical results”: already said.

� This phrase was already mentioned in the previous text. Hence, I will delete this per your suggestion.

Lines 187-190: “the 2017 AAPCC annual report… in that order”: this report included all routes of exposures (not only ingestions); so, is it comparable?

� The AAPCC reports all poisoning substances and all routes. It is not only about chemical ingestion. However, the poisoning of all substances was considered comparable with the most extensively reported systematic report. This study presented more information regarding the types and mortality rates of chemical ingestion than our study.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

19 Feb 2020

Types and clinical outcomes of chemical ingestion in emergency departments in South Korea (2011-2016)

PONE-D-19-33190R1

Dear Dr. Hee Lee,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Martina Crivellari

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors quickly responded all my questions/comments, adjust statistical methods, modified the conclusion, and further discussed limitations. A lot of work have been done. I have no further comments.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

26 Feb 2020

PONE-D-19-33190R1

Types and clinical outcomes of chemical ingestion in emergency departments in South Korea (2011-2016)

Dear Dr. Lee:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Martina Crivellari

Academic Editor

PLOS ONE