Evaluating the association between COVID-19 and psychiatric presentations, suicidal ideation in an emergency department

Michal J McDowell; Carrie E Fry; Mladen Nisavic; Mila Grossman; Charles Masaki; Emily Sorg; Suzanne Bird; Felicia Smith; Scott R Beach

doi:10.1371/journal.pone.0253805

. 2021 Jun 30;16(6):e0253805. doi: 10.1371/journal.pone.0253805

Evaluating the association between COVID-19 and psychiatric presentations, suicidal ideation in an emergency department

Michal J McDowell ^1,^2,^*, Carrie E Fry ³, Mladen Nisavic ¹, Mila Grossman ^1,², Charles Masaki ^1,², Emily Sorg ¹, Suzanne Bird ¹, Felicia Smith ¹, Scott R Beach ¹

Editor: Corstiaan den Uil⁴

PMCID: PMC8244888 PMID: 34191850

Abstract

Objective

To estimate the association between COVID-19 and Emergency Department (ED) psychiatric presentations, including suicidal ideation.

Methods

Using an interrupted time series design, we analyzed psychiatric presentations using electronic health record data in an academic medical center ED between 2018 and 2020. We used regression models to assess the association between the onset of the COVID-19 outbreak and certain psychiatric presentations. The period February 26–March 6, 2020 was used to define patterns in psychiatric presentations before and after the coronavirus outbreak.

Results

We found a 36.2% decrease (unadjusted) in ED psychiatric consults following the coronavirus outbreak, as compared to the previous year. After accounting for underlying trends, our results estimate significant differential change associated with suicidal ideation and substance use disorder (SUD) presentations following the outbreak. Specifically, we noted a significant differential increase in presentations with suicidal ideation six weeks after the outbreak (36.4 percentage points change; 95% CI: 5.3, 67.6). For presentations with SUD, we found a differential increase in the COVID-19 time series relative to the comparison time series at all post-outbreak time points and this differential increase was significant three weeks (32.8 percentage points; 95% CI: 4.0, 61.6) following the outbreak. Our results estimate no differential changes significant at the P value < 0.05 level associated with affective disorder or psychotic disorder presentations in the COVID-19 time series relative to the comparator time series.

Conclusions

The COVID-19 outbreak in Boston was associated with significant differential increases in ED presentations with suicidal ideation and SUD.

Introduction

The coronavirus disease 2019 (COVID-19) has emerged as a significant driver of physical morbidity and mortality worldwide, and it stands to exact sizable psychological toll. Psychological distress surrounding COVID-19 related illness and death may be exacerbated by fear, isolation, and physical distancing [1, 2]. Further compounding these factors are job loss and financial stress, which are well-recognized risk factors for suicide [1, 3]. Past global health crises offer cautionary lessons about potential mental health needs in the setting of a pandemic. During the 2003 severe acute respiratory syndrome (SARS) outbreak, stress was significantly higher in SARS-infected patients than in healthy controls [4]. Previous pandemics were also associated with escalations in suicidal ideation, suicide attempts, and completed suicides: rates of death by suicide in the United States increased following the 1918 influenza pandemic [5], and there was a 30% increase in suicide rates in older adults in Hong Kong during the 2003 SARS epidemic [6].

Particular to suicidal ideation, the coronavirus pandemic carries with it known risk factors for suicide at each level of the socio-ecological model [7–9]. The pandemic led to economic downturn, barriers to health care access, deprioritization of mental health in the setting of infection control measures, decreased access to community and religious supports, interpersonal conflict, unemployment, poverty, loneliness, and hopelessness [10–12].

Thus far, data on the association between COVID-19 and psychiatric illness includes case reports and case series, as well as data from anonymized health records, surveys, and publicly available entities [13–17]. Among psychiatric illnesses described as occurring or worsening in the setting of the pandemic, depression, anxiety, post-traumatic stress disorder, substance use disorders (SUDs), and general distress are all represented [3, 18–20]. One study found a twofold increase in emergency department visits for opioid overdoses as compared to the previous year during the first four months of the COVID-19 pandemic [21].

As one of the first states to suffer a major coronavirus outbreak in the United States, Massachusetts serves as an important test case in assessment of the relationship between the global pandemic and psychological distress. On February 1, 2020, Boston reported its first confirmed case of COVID-19, and by March 10, Governor Charles Baker declared a state of emergency in response to the coronavirus spread [22]. We used the period following the initial outbreak of COVID-19 in Massachusetts to assess the association between the pandemic and patients presenting with certain psychiatric complaints (a number of selected mood, psychotic, and substance use disorders, as well as suicidal ideation) to a tertiary care hospital Emergency Department (ED).

Methods

Study design

In this study, we employed a two-group, two-period cross-sectional (i.e., between-subject) study design, using data from electronic health records (EHR) of patients seen in the ED at Massachusetts General Hospital, a large, tertiary care hospital in Boston, Massachusetts, United States. The Partners Institutional Review Board approved this work (approval #2020P001482) and waived the requirement for informed consent. We used an interrupted time series design to compare changes in the proportion of patients seen with certain psychiatric presentations before and after the COVID-19 outbreak. We aimed to quantify the relationship between COVID-19 and the associated economic shut-down and quarantine orders on psychiatric ED presentations.

Comparative interrupted time series (CITS) design uses data collected over time, usually recorded at regular intervals [23]. We used a two-group (pre- and post-exposure), two time period (2018–2019 and 2019–2020) CITS [24] study design for a number of reasons. First, adequately capturing the evolution of certain psychiatric presentations before COVID-19 reduces the chance of history bias [25]. Second, a comparison series allows for adjustment for trends unrelated to the coronavirus pandemic. For example, presentation of some psychiatric disorders varies throughout the year [12, 26]. To capture changes in psychiatric presentation associated with COVID-19, we used changes in presentation in the previous year as our comparator. These data can identify underlying trends, and, when an exposure occurs at a known time, post-exposure trends can be studied to assess changes from pre-exposure trends. The change in the comparisons series thus can serve as the counterfactual, or what would happen regardless of the exposure. Third, unlike a standard difference-in-differences approach, general CITS design allows for modeling a pre-period trend differential across the two groups. Finally, we used a general CITS, rather than a linear CITS, to minimize the parametric assumptions made regarding the outcome’s evolution. A linear CITS assumes that the outcome is evolving linearly in the pre- and post-period [24]. On the other hand, general CITS assumes that any difference between the two groups is evolving linearly but does not make any assumption about the outcome’s parametric form (i.e., does not assume a linear, quadratic, or other form). Additionally, we conducted an event study to visually assess the plausibility of these assumptions and found differential trends (i.e., non-parallel) in the pre-period suggesting that a standard difference-in-difference approach would not be appropriate.

We collected cross-sectional data from the hospital EHR for patients who received care from the ED psychiatry consult service. In this ED, patients receive emergency psychiatric consultation at the request of Emergency Medicine providers, and generally in response to presentations related to acute risk to self or others, significantly altered mental status and/or behavior, and/or presentations with extreme psychological distress. Data for the comparison time series were recorded from charts of patients seen December 2018 through May 2019. Data for the time series assessing the coronavirus outbreak were recorded from charts of patients seen December 2019 through May 2020.

To reduce chart burden, we abstracted charts on Mondays, Wednesdays, and Fridays of each week in the two study periods. Three days per week provided adequate power to detect an effect. Our data are at the person-day level. Patients were included if they received a psychiatric consultation in any one of the four observation periods (pre- and post-period for comparator time series and pre- and post-period for COVID-19 time series). Patients were allowed to appear in more than one of these four study periods. We adjusted the models for the total number of ED visits per patient per study period. Patients that did not receive a psychiatric consultation (even if one was requested) were excluded from this study. Additionally, patients with incomplete demographic or diagnostic records were excluded from analyses that used the characteristic of interest, including CITS regression modeling (approximately 90 records).

Race, ethnicity, gender, insurance, and legal status were available directly from statistical information maintained by the ED triage team and recorded in the EHR. Of note, ED staff received education in the July 2019 department newsletter regarding the importance of capturing race and ethnicity data accurately. Additionally, reminders were sent about the importance of accurately capturing these data during the initial COVID-19 wave in Boston. Psychiatric presentation diagnosis and presence of suicidal ideation were abstracted by clinician author chart review.

We defined the exposure as the time period from February 26, 2020 –March 6, 2020 to indicate the coronavirus outbreak in Massachusetts and the initial public health response. We refer to this period in this study as the “exposure.” This period accounts for the growth in cases across Massachusetts, potentially associated with the Biogen conference held in Boston, which began on February 26, 2020 [27]. The first patient with a documented case of COVID-19 in Massachusetts presented to the study site hospital on March 1, 2020. The comparator series pre-period runs December 24, 2018–February 25, 2019 and the COVID-19 series pre-period runs December 23, 2019–February 24, 2020. The post-period for the comparator time series is March 11, 2019–May 13, 2019 and for the COVID-19 series is March 9, 2020–May 17, 2020. The timeline of COVID-19 cases in Massachusetts and actions taken to mitigate spread of the virus, as well as the study period (including the exposure), are detailed in Fig 1.

Outcome measures

From each medical record, we extracted information both automatically on patient demographics, as well as manually regarding psychiatric consultation findings. In other words, the medical records could be queried for summary information for some variables, but information had to be manually extracted by searching each record individually for other variables. Charts were reviewed by six psychiatrists (three residents and three attendings), and chart reviews that were ambiguous were adjudicated by other members of authorship team at bimonthly team meetings. The presence or absence of suicidal ideation as determined by text in the psychiatric consultation note, as well as up to three psychiatric diagnoses (not including SUDs), were recorded for each patient presentation. Reviewers included relevant and current diagnoses identified in the consult notes, as well as interpreted symptoms mentioned with Diagnostic and Statistical Manual of Mental Disorders 5^th edition (DSM-5) criteria in mind.

Our outcomes of interest were the proportion of patients who presented with suicidal ideation (SI), SUDs (including alcohol, opioid, cocaine, and stimulant use disorders, as well as substance-induced mood and psychotic disorders), psychotic disorders (including schizophrenia-spectrum illnesses and brief psychotic disorder), and affective disorders (including major depressive disorder, and bipolar affective disorder) among all patients seen by the psychiatric consult service. We only included present suicidal ideation in our suicidal ideation outcome variable. For psychiatric diagnoses, we generally used the diagnosis given by the evaluating clinician. Historical labels were only used if they were deemed correct by the evaluating clinician researcher and relevant to the current presentation. For SUDs, historical diagnoses were used if substances were involved in the current presentation or if there was compelling evidence of a current SUD in the consult note. We included suicidal ideation, SUDs, psychotic disorders, and affective disorders as our outcomes of interest because, based on clinical experience, these diagnoses are reliably and consistently captured in this ED setting. Anxiety disorders, although coded by the reviewers, represent a small proportion of ED visits and unless they are the primary reason for presentation, they are less likely to be consistently captured.

Data analysis

First, we compared demographic and clinical characteristics among the pre-period of each time series (December 24, 2018–February 25, 2019 and December 23, 2019–February 24, 2020) using a two-sided t-test to measure baseline similarity between the two groups. Then, we estimated linear probability models (LPMs) (ƒlines) for each of our four outcomes. LPMs are preferred to logit or probit regression models with binary outcomes for both CITS and difference-in-difference models, due to the ease of interpreting interactions in LPMs relative to logit or probit models [28]. Additionally, LPMs estimate unbiased and consistent treatment effects for events that occur with a probability between 0.20 and 0.80. Regression models were adjusted with patient age, sex, insurance, and total number of psychiatric consults. Additionally, we adjusted with the patient’s race and ethnicity because of the disproportionate impact of COVID-19 on Black and Latinx communities as a result of structural racism [29, 30]. We also included day-of-week and month fixed effects to account for the evolution of psychiatric presentations over the course of a week and year. We do not present data not adjusted by demographic factors, as the demographic composition of cohorts may change in different ways in the pre- to post-periods.

For suicidal ideation, we also adjusted for the presence of an affective disorder, psychotic disorder, and SUD. As a sensitivity analysis, we estimated the same regression for suicidal ideation in S3 Table, omitting the other diagnostic categories as covariates and adding back one of the diagnostic categories, since they may be on the causal pathway between the coronavirus exposure and suicidal ideation.

We estimated a treatment effect for each week in the post-period, rather than an average, to account for the temporal changes in care-seeking behavior during and after the initial surge in COVID-19 cases in Boston. In other words, our regression specification included an indicator variable for each post-period week. We interacted the indicator variable with the treatment variable, and each of these interactions represents the differential change for the week between the COVID-19 series and comparator series, as compared to the average pre-period levels. Additionally, we checked for autocorrelation in the time series and found no evidence of suspected autocorrelation structures, including AR1 or AR3. Additionally, the statistical significance of our results did not change using the most conservative method of multiple comparison adjustment, Bonferroni corrections. Our analyses used a 2-sided P<0.05 value as a threshold for reporting statistical significance and were performed using R statistical software.

Results

To assess overall volume of ED visits requiring psychiatric consultations, we compared unadjusted trends in consults from the comparator (March-May 2019) post-period with consults from the COVID-19 (March-May 2020) post-period. This comparison showed a 36.2% decrease in psychiatric consultations (Fig 2) (561 ED visits requiring psychiatric consultations in the 2019 post-period, as compared to 358 in the 2020 post-period). By comparison, the total number of ED arrival volumes decreased by 28.8% (20,198 total ED presentations in the 2019 post-period, as compared to 14,358 in the 2020 post-period).

In exploring pre-period differences, we found that the COVID-19 pre-period (December 23, 2019–February 24, 2020) cohort had a higher proportion of psychiatric ED consultations involving people from a minority racial group than did the comparator pre-period (December 24, 2018–February 25, 2019) (Table 1). We further explore these differences in S1 Table by presenting an unadjusted comparison of racial/ethnic groups stratified by psychiatric presentation in the pre-period. No statistically significant pre-period differences were found when comparing group variables including sex, median age, and insurance payer.

Table 1. Pre-period unadjusted comparison of demographic characteristics.

Characteristic	2018–2019 Cohort (n = 489)	2019–2020 Cohort (n = 467)	Test-statistic (p-value)
Age, median, years	38	39	--
Sex, n (%)
Male	281 (57.5)	301 (58.0)	0.2 (0.9)
Female	208 (42.5)	196 (42.0)	0.2 (0.9)
Racial Classification, n (%)
White	383 (78.3)	320 (68.5)	3.4 (<0.001)
Black	55 (11.2)	79 (16.9)	-2.5; 0.01
Asian	9 (1.8)	17 (3.6)	-1.7; 0.09
American Indian/Alaska Native	5 (1.0)	2 (0.4)	1.1; 0.3
Other	32 (6.5)	38 (8.1)	-0.9; 0.02
Hispanic or Latino/a/x, n (%)
Yes	47 (9.9)	53 (11.9)	-1.0; 0.20
No	426 (80.1)	391 (78.1)	-1.0; 0.20
Insurance, n(%)
MA Medicaid	194 (29.7)	177 (37.9)	0.6; 0.6
Private Insurance	140 (28.6)	136 (29.1)	-0.2; 0.9
Medicare	92 (18.8)	107 (22.9)	-1.6; 0.1
Uninsured	24 (.9)	13 (2.8)	1.7; 0.09
Other public coverage	39 (8.0)	34 (7.3)	0.4; 0.7

Open in a new tab

We found that rates of psychiatric presentation were similar between the two study groups in the pre-period for patients presenting with suicidal ideation, SUDs, and affective disorders (Table 2). We found that fewer patients with psychotic disorders were presenting in the 2019–2020 pre-period cohort, as compared with the 2018–2019 pre-period cohort (p = 0.04).

Table 2. Pre-period unadjusted comparison of psychiatric presentations.

Psychiatric Presentation n (%)	2018–2019 Cohort (n = 489)	2019–2020 Cohort (n = 467)	Test statistic; p-value
Suicidal Ideation	293 (59.9)	274 (58.7)	0.4; 0.7
Substance Use Disorder	202 (41.3)	184 (39.4)	0.6; 0.5
Affective Disorder	172 (35.2)	184 (39.4)	-1.35; 0.2
Psychotic Disorder	83 (17.0)	57 (12.2)	2.1; 0.04

Open in a new tab

Our adjusted models for each of the four psychiatric presentation groups (suicidal ideation, affective disorders, psychotic disorders, and SUDs) for the two study periods, which run from December to May in 2018–2019 and 2019–2020, respectively, are summarized in Fig 3, with tabular data presented in S2 Table. These results suggest that after the COVID-19 pandemic began spreading in Boston, there was no significant differential change in the proportion of ED presentations for affective disorders or psychotic disorders at any point in the COVID-19 time series relative to the comparator time series.

In the adjusted models, we found differential changes in the model proportions for both presentations with suicidal ideation, as well as SUDs. For psychiatric consultations with suicidal ideation present, we found a significant differential increase six weeks after the exposure, or the week of April 13–17, 2020 (36.4 percentage points change; 95% CI: 5.3, 67.6). Based on the comparison series, we would have expected an increase of 16.0 percentage points for suicidal ideation at week six. Relative to what we would have expected based on presentations from the previous year during week six, we observed a 127.5% increase in presentations with suicidal ideation following the coronavirus outbreak. Additionally, at nine weeks after the exposure (May 4–9, 2020), there was a differential decrease in the proportion of psychiatric presentations to the ED with suicidal ideation (34.9 percentage points; 95% CI; -69.5, -0.3), though this change was not statistically significant (p = 0.06). We calculated the percent increase for suicidal ideation presentations (as well as for SUD presentations below) by dividing the differential change in the post-period (i.e., the treatment effect estimate) by the difference in means from the pre-period to the post-period in the comparison group (i.e., the counterfactual).

We present data in S3 Table for suicidal ideation, sequentially removing and re-adding other certain psychiatric presentations to the model. Our results change modestly because of this different regression specification. There is a significant increase in presentations with suicidal ideation at weeks three and six. However, once adjustment for SUD is included in the model at week three, the differential change is no longer significant because this increase was mediated by the co-occurring SUD diagnoses. However, even after the adjustments for co-occurring psychiatric conditions are added at week six, the differential change persists, as noted above.

There is also a differential increase in the adjusted proportion of presentations with SUD in the COVID-19 time series relative to the comparison time series at all time points after the exposure. This differential increase is statistically significant at three weeks following the exposure, or the week of March 23–27, 2020 (32.8 percentage points; 95% CI: 4.0, 61.6). Based on the comparison series, we would have expected a decrease of 6.2 percentage points for substance use at week three. Relative to what we would have expected based on diagnoses from the previous year during week three, we saw a 629.0% increase in SUD presentations following the coronavirus outbreak.

Discussion

While overall ED psychiatric presentations decreased during the COVID-19 pandemic in Boston, the proportion of psychiatric presentations with suicidal ideation and SUDs increased.

The proportion of presentations with suicidal ideation increased six weeks after the exposure (April 13–17, 2020) or approximately three weeks after the March 23, 2020 shelter-in-place order. This was also the week leading up to the peak of the initial surge of COVID-19 cases in Massachusetts (4,946 new cases reported on April 24, 2020) [31]. This increase in the proportion of psychiatric presentations with suicidal ideation may reflect increasing stress as a result of the lockdown and isolation [1, 2]. The influence of having COVID, knowing people who had COVID, or fears about being infected with the coronavirus may also be related to this finding. While another recent Massachusetts-based study suggests that overall suicide rates did not increase during the pandemic, it appears that the experience of suicidal ideation may have increased at certain time points following the outbreak [13]. This finding regarding increased pandemic-related suicidal ideation, as well as our finding regarding increased presentations with SUD, is corroborated by another recent large-scale cross-sectional study in the United States [32].

Of note, in the adjusted analyses, at nine weeks after the exposure (May 4–9, 2020), there was a meaningful differential decrease in the proportion of psychiatric presentations to the ED with suicidal ideation. This decrease in suicidal ideation presentations parallels Governor Charles Baker’s announcement of the Reopening Advisory Board (April 28, 2020) [33] and the subsequent four-phase reopening plan announcement on May 11, 2020 [34]. Additionally, the month of May brought to Boston weather with an increased average daily temperature compared to the previous month, potentially facilitating outdoor socializing and connection. Unlike the trends noted in SUDs presentations, ED psychiatric presentations with suicidal ideation were dynamic across the study period.

Trends in presentations for SUDs suggest that these illnesses may have worsened during the pandemic, consistent with expert predictions and other findings [11, 21, 35, 36]. Some patients may have escalated substance use as a means of coping with the stress of the pandemic and shutdown [11]. Additionally, SUDs are highly prevalent in populations experiencing houselessness [37]. Given the higher risk of COVID-19 in communities without stable housing, several shelters in Boston changed their access policies to mitigate the spread of disease [35, 38]. Though important for reducing viral spread, these policies potentially resulted in housing barriers for communities experiencing houselessness, which may have disproportionately affected persons with SUDs [35, 38]. Furthermore, individuals with SUDs may have faced challenges accessing substances during quarantine, leading to increased presentations for withdrawal-related symptoms. Additionally, patients may have struggled accessing treatment, as detoxification and residential care programs were initially limiting access to patients without a known COVID-19 status and not accepting patients who tested positive for COVID, even if asymptomatic. Most concerning, perhaps, are cases involving individuals who experienced decreased tolerance to substances in the setting of weeks of reduced access and thus were more likely to unintentionally overdose during subsequent substance use [36]. Additionally, our sensitivity analyses suggest that SUD presentations seem to explain the increase in suicidal ideation presentations in week 3 in the unadjusted models presented in S3 Table, given that estimates become non-significant once SUD was included as an adjustment. This aligns with our findings that there is a significant increase in SUD presentations following the COVID-19 outbreak.

We found pre-period differences between the 2018–2019 and 2019–2020 cohorts, with a higher proportion of psychiatric ED consultations involving people from a minority racial group in the latter 2019–2020 cohort. When we further explore this difference by separating the groups by psychiatric presentations, statistically significant differences are isolated to fewer White patients presenting with psychotic disorders and more Asian patients presenting with affective disorders in the 2019–2020 cohort. The reason behind these pre-period differences remains unclear, though perhaps ED staff training regarding documentation improved the accuracy of recording data on race and ethnicity for the 2019–2020 cohort.

A better understanding of overall mental health resource utilization during the pandemic is needed, particularly in the setting of decreased psychiatric ED presentations following the outbreak, as well as documented expansion of telehealth services [39]. Patients in need of psychiatric care may have worried about potential risk for coronavirus infection associated with ED presentation and may have chosen to access treatment through other modalities, such as telehealth services, crisis support lines, and employee assistance programs. Another grimmer possibility is that patients went without treatment for weeks until access improved or until their symptoms became so burdensome that they could no longer be ignored [40]. Given the dynamic trends in suicidal ideation observed during the study period, further work is needed to better characterize phenotypes of patients presenting with suicidal ideation during the pandemic, including the triggers for suicidal ideation, underlying co-morbidities, and the presence of plan, intent, and attempts. Additionally, identification of patient populations vulnerable to suicidal ideation and attempt is requisite to appropriately target interventions.

Moving forward, guaranteed access to community-based mental health care and addiction services will be important [11, 12]. Specifically, maintenance of accessibility to resources, as well as insurance coverage for telehealth services, is prudent. We recommend raising awareness surrounding the potential deleterious relationship between the pandemic and mental illness, SUDs, and suicidal ideation.

Limitations

Our study has limitations. First, circumstances unique to Boston may have contributed to our findings, including those that we could not identify, potentially resulting in less generalizability outside of this area. Second, our study did not capture suicidal ideation, intent, plan, and attempt in patients who did not present for care. Third, we stopped collecting data in May, consistent with the end of the initial surge in Massachusetts. Because of this data limitation, we do not capture trends of psychiatric presentations that occurred after the surge, which may still have been related to the initial outbreak. Fourth, in our generation of outcome variables, we relied on clinician reviewer interpretation of consultation notes, which also represent interpretations of clinical presentations. We hope that the reviewers added a second layer of assessment, thus improving diagnostic clarity, though understand that the additional review may have introduced reviewer bias. Fifth, while the confidence intervals for some of our results are large, they generally do not suggest that we are underpowered to detect an effect (i.e., an unstably estimated zero) except for presentations with psychosis. In most cases, we can rule out large estimates in the opposite direction of the treatment effect, suggesting that the estimated treatment effects may be under-estimates of the increases or decreases seen in our study. Additionally, with a general CITS, estimates further from the exposure typically have larger confidence intervals, as uncertainty increases over the time period when estimating a between-group difference from the data. Finally, there were many interventions to address the spread of COVID-19 during our study period. While we use a robust quasi-experimental design to capture changes in psychiatric presentations to ED unrelated to COVID-19, our pattern of findings may be attributed to more than the policy changes cited. Thus, caution is warranted when interpreting our results as causal.

Conclusions

The COVID-19 pandemic is associated with changes in ED presentations requiring psychiatric consultation, particularly those involving suicidal ideation and SUDs.

Supporting information

S1 Material

(DOCX)

Click here for additional data file.^{(13.1KB, docx)}

S1 Table. Pre-period unadjusted comparison of racial/ethnic groups by psychiatric presentation.

(DOCX)

Click here for additional data file.^{(14KB, docx)}

S2 Table. Full regression output for differential changes in emergency department psychiatric presentations in the 2019 versus 2020 post-periods.

(DOCX)

Click here for additional data file.^{(22.1KB, docx)}

S3 Table. Differential change in suicidal ideation presentations between COVID-19 time series (Dec 2019–May 2020) and comparator time series (Dec 2018–May 2019) from pre-period (Dec–Feb) to post-period (Mar–May) among all ED psychiatric presentations, varying adjustment for other co-occurring psychiatric conditions.

(DOCX)

Click here for additional data file.^{(16.4KB, docx)}

Acknowledgments

All contributors to this manuscript met criteria for authorship.

Data Availability

There are ethical restrictions on sharing our data set, as our data contain potentially identifying and sensitive patient information per the Partners Institutional Review Board. Data is available upon request. Requests may be sent to the Partners Institutional Review Board protocol administrator, Virginia Rodriguez, at VGRODRIGUEZ@PARTNERS.ORG. For reference, our protocol number for the review board is 2020P—1482.

Funding Statement

Author: MM Grant #: R25 MH094612, Funder: National Institute of Mental Health Website: https://www.nimh.nih.gov/index.shtml Sponsor did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Gunnell D, Appleby L, Arensman E, Hawton K, John A, Kapur N, et al. Suicide risk and prevention during the COVID-19 pandemic. Lancet Psychiatry. 2020;7: 468–471. doi: 10.1016/S2215-0366(20)30171-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Yao H, Chen J-H, Xu Y-F. Patients with mental health disorders in the COVID-19 epidemic. Lancet Psychiatry. 2020;7: e21. doi: 10.1016/S2215-0366(20)30090-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Sherman AC, Williams ML, Amick BC, Hudson TJ, Messias EL. Mental health outcomes associated with the COVID-19 pandemic: Prevalence and risk factors in a southern US state. Psychiatry Res. 2020;293: 113476. doi: 10.1016/j.psychres.2020.113476 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Chua SE, Cheung V, McAlonan GM, Cheung C, Wong JWS, Cheung EPT, et al. Stress and psychological impact on SARS patients during the outbreak. Can J Psychiatry Rev Can Psychiatr. 2004;49: 385–390. doi: 10.1177/070674370404900607 [DOI] [PubMed] [Google Scholar]
5.Wasserman IM. The impact of epidemic, war, prohibition and media on suicide: United States, 1910–1920. Suicide Life Threat Behav. 1992;22: 240–254. [PubMed] [Google Scholar]
6.Cheung YT, Chau PH, Yip PSF. A revisit on older adults suicides and Severe Acute Respiratory Syndrome (SARS) epidemic in Hong Kong. Int J Geriatr Psychiatry. 2008;23: 1231–1238. doi: 10.1002/gps.2056 [DOI] [PubMed] [Google Scholar]
7.Oyesanya M, Lopez-Morinigo J, Dutta R. Systematic review of suicide in economic recession. World J Psychiatry. 2015;5: 243–254. doi: 10.5498/wjp.v5.i2.243 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Van Orden KA, Witte TK, Cukrowicz KC, Braithwaite SR, Selby EA, Joiner TE. The interpersonal theory of suicide. Psychol Rev. 2010;117: 575–600. doi: 10.1037/a0018697 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.VanderWeele TJ, Li S, Tsai AC, Kawachi I. Association Between Religious Service Attendance and Lower Suicide Rates Among US Women. JAMA Psychiatry. 2016;73: 845–851. doi: 10.1001/jamapsychiatry.2016.1243 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Wasserman D, Iosue M, Wuestefeld A, Carli V. Adaptation of evidence-based suicide prevention strategies during and after the COVID-19 pandemic. World Psychiatry. 2020;19: 294–306. doi: 10.1002/wps.20801 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Laderman M, Komaromy M, Moses J, Jarrett M. Under the cover of COVID-19 news, the epidemic of opioid overdoses silently surges. In: Becker’s Hospital Review [Internet]. 30 Oct 2020. [cited 4 Nov 2020]. Available: https://www.beckershospitalreview.com/opioids/under-the-cover-of-covid-19-news-the-epidemic-of-opioid-overdoses-silently-surges.html. [Google Scholar]
12.Reger MA, Stanley IH, Joiner TE. Suicide Mortality and Coronavirus Disease 2019—A Perfect Storm? JAMA Psychiatry. 2020;77: 1093. doi: 10.1001/jamapsychiatry.2020.1060 [DOI] [PubMed] [Google Scholar]
13.Faust JS, Shah SB, Du C, Li S-X, Lin Z, Krumholz HM. Suicide Deaths during the Stay-at-Home Advisory in Massachusetts. medRxiv. 2020; 2020.10.20.20215343. doi: 10.1101/2020.10.20.20215343 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Halford EA, Lake AM, Gould MS. Google searches for suicide and suicide risk factors in the early stages of the COVID-19 pandemic. PloS One. 2020;15: e0236777. doi: 10.1371/journal.pone.0236777 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Czeisler MÉ. Mental Health, Substance Use, and Suicidal Ideation During the COVID-19 Pandemic—United States, June 24–30, 2020. MMWR Morb Mortal Wkly Rep. 2020;69. doi: 10.15585/mmwr.mm6932a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Sher L. COVID-19, anxiety, sleep disturbances and suicide. Sleep Med. 2020;70: 124. doi: 10.1016/j.sleep.2020.04.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Taquet M, Luciano S, Geddes JR, Harrison PJ. Bidirectional associations between COVID-19 and psychiatric disorder: retrospective cohort studies of 62 354 COVID-19 cases in the USA. Lancet Psychiatry. 2020;0. doi: 10.1016/S2215-0366(20)30462-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Rossi R, Socci V, Talevi D, Mensi S, Niolu C, Pacitti F, et al. COVID-19 Pandemic and Lockdown Measures Impact on Mental Health Among the General Population in Italy. Front Psychiatry. 2020;11. doi: 10.3389/fpsyt.2020.00790 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Farhoudian A, Baldacchino A, Clark N, Gerra G, Ekhtiari H, Dom G, et al. COVID-19 and Substance Use Disorders: Recommendations to a Comprehensive Healthcare Response. An International Society of Addiction Medicine (ISAM) Practice and Policy Interest Group Position Paper. Basic Clin Neurosci. 2020;11: 133–146. doi: 10.32598/bcn.11.covid19.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Pedrosa AL, Bitencourt L, Fróes ACF, Cazumbá MLB, Campos RGB, de Brito SBCS, et al. Emotional, Behavioral, and Psychological Impact of the COVID-19 Pandemic. Front Psychol. 2020;11. doi: 10.3389/fpsyg.2020.566212 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Ochalek TA, Cumpston KL, Wills BK, Gal TS, Moeller FG. Nonfatal Opioid Overdoses at an Urban Emergency Department During the COVID-19 Pandemic. JAMA. 2020;324: 1673. doi: 10.1001/jama.2020.17477 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Coronavirus Disease (COVID-19) timeline. In: Boston.gov [Internet]. 15 Mar 2020 [cited 4 Nov 2020]. Available: https://www.boston.gov/departments/public-health-commission/coronavirus-timeline.
23.Humphreys DK, Gasparrini A, Wiebe DJ. Evaluating the Impact of Florida’s “Stand Your Ground” Self-defense Law on Homicide and Suicide by Firearm: An Interrupted Time Series Study. JAMA Intern Med. 2017;177: 44–50. doi: 10.1001/jamainternmed.2016.6811 [DOI] [PubMed] [Google Scholar]
24.Fry CE, Hatfield LA. Do Methodological Birds of a Feather Flock Together? ArXiv200611346 Econ Stat. 2020 [cited 2 Jul 2020]. Available: http://arxiv.org/abs/2006.11346.
25.Shadish WR, Cook TD, Campbell DT. Quasi-Experimentation: Interrupted Time Series Designs. 2nd ed. Experimental and Quasi-Experimental Designs for Generalized Causal Inference. 2nd ed. Cengage; 2002. doi: 10.1037/1082-989x.7.1.3 [DOI] [Google Scholar]
26.Woo J-M, Okusaga O, Postolache TT. Seasonality of Suicidal Behavior. Int J Environ Res Public Health. 2012;9: 531–547. doi: 10.3390/ijerph9020531 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Lemieux J, Siddle KJ, Shaw BM, Loreth C, Schaffner S, Gladden-Young A, et al. Phylogenetic analysis of SARS-CoV-2 in the Boston area highlights the role of recurrent importation and superspreading events. medRxiv. 2020; 2020.08.23.20178236. doi: 10.1101/2020.08.23.20178236 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Karaca-Mandic P, Norton EC, Dowd B. Interaction terms in nonlinear models. Health Serv Res. 2012;47: 255–274. doi: 10.1111/j.1475-6773.2011.01314.x [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Health Equity Considerations & Racial & Ethnic Minority Groups. In: www.cdc.gov [Internet]. 30 Apr 2020 [cited 28 Oct 2020]. Available: https://www.cdc.gov/coronavirus/2019-ncov/community/health-equity/race-ethnicity.html.
30.Yancy CW. COVID-19 and African Americans. JAMA. 2020;323: 1891. doi: 10.1001/jama.2020.6548 [DOI] [PubMed] [Google Scholar]
31.Trends in Number of COVID-19 Cases in the US Reported to CDC, by State/Territory. In: www.cdc.gov [Internet]. [cited 4 Nov 2020]. Available: https://www.cdc.gov/covid-data-tracker.
32.Holland KM, Jones C, Vivolo-Kantor AM, Idaikkadar N, Zwald M, Hoots B, et al. Trends in US Emergency Department Visits for Mental Health, Overdose, and Violence Outcomes Before and During the COVID-19 Pandemic. JAMA Psychiatry. 2021. doi: 10.1001/jamapsychiatry.2020.4402 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Baker-Polito Administration Extends Non-Essential Business Closures to May 18th, Announces Reopening Advisory Board | Mass.gov. [cited 12 Nov 2020]. Available: https://www.mass.gov/news/baker-polito-administration-extends-non-essential-business-closures-to-may-18th-announces.
34.Baker-Polito Administration Announces Four-Phase Approach to Reopening and Publishes Mandatory Workplace Safety Standards | Mass.gov. [cited 12 Nov 2020]. Available: https://www.mass.gov/news/baker-polito-administration-announces-four-phase-approach-to-reopening-and-publishes-mandatory.
35.Slavova S, Rock P, Bush HM, Quesinberry D, Walsh SL. Signal of increased opioid overdose during COVID-19 from emergency medical services data. Drug Alcohol Depend. 2020;214: 108176. doi: 10.1016/j.drugalcdep.2020.108176 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Wakeman SE, Green TC, Rich J. An overdose surge will compound the COVID-19 pandemic if urgent action is not taken. Nat Med. 2020;26: 819–820. doi: 10.1038/s41591-020-0898-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Torchalla I, Strehlau V, Li K, Aube Linden I, Noel F, Krausz M. Posttraumatic stress disorder and substance use disorder comorbidity in homeless adults: Prevalence, correlates, and sex differences. Psychol Addict Behav. 2014;28: 443–452. doi: 10.1037/a0033674 [DOI] [PubMed] [Google Scholar]
38.Baggett TP, Keyes H, Sporn N, Gaeta JM. Prevalence of SARS-CoV-2 Infection in Residents of a Large Homeless Shelter in Boston. JAMA. 2020;323: 2191–2192. doi: 10.1001/jama.2020.6887 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Wosik J, Fudim M, Cameron B, Gellad ZF, Cho A, Phinney D, et al. Telehealth transformation: COVID-19 and the rise of virtual care. J Am Med Inform Assoc. 2020;27: 957–962. doi: 10.1093/jamia/ocaa067 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Castro VM, Perlis RH. Electronic Health Record Documentation of Psychiatric Assessments in Massachusetts Emergency Department and Outpatient Settings During the Coronavirus Disease 2019 (COVID-19) Pandemic. JAMA Netw Open. 2020;3: e2011346. doi: 10.1001/jamanetworkopen.2020.11346 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0253805.r001

Decision Letter 0

Corstiaan den Uil

18 Mar 2021

PONE-D-21-02006

Evaluating the Association Between COVID-19 and Psychiatric Presentations, Suicidal Ideation in an Emergency Department: An Interrupted Time Series Study

PLOS ONE

Dear Dr. McDowell,

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.

Please submit your revised manuscript by May 02 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're 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.

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). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

We look forward to receiving your revised manuscript.

Kind regards,

Corstiaan den Uil

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

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for including your ethics statement:

"- Mass General Brigham

- #2020P001482

- The IRB has determined that this project meets the criteria for exemption 45 CFR

46.101(b)(#)

- Consent was not obtained".

a. Please amend your current ethics statement to include the full name of the ethics committee/institutional review board(s) that approved your specific study.

b. Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research

3. Please provide additional details regarding participant consent.

If you are reporting a retrospective study of medical records or archived samples, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent.

Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to " ext-link-type="uri" xlink:type="simple">http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research."

4. n your Methods section, please ensure you have provided sufficient details to replicate the analyses such as a description of any inclusion/exclusion criteria that were applied to participant inclusion in the analysis.

Moreover, please specify the name of the centre where the research took place.

5. Please refer to the specific statistical analyses performed as well as any post-hoc corrections to correct for multiple comparisons. If these were not performed please justify the reasons.

Please refer to our statistical reporting guidelines for assistance (https://journals.plos.org/plosone/s/submission-guidelines.#loc-statistical-reporting).

6. 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.

[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

**********

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

Reviewer #1: 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

**********

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

**********

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: This study examines changes in the prevalence of suicidality and selected psychiatric disorders in a busy ED during the early phase of the COVID-19 pandemic, relative to earlier time periods. The investigation addresses a concern of obvious interest and relevance. Several studies have reported changes in psychiatric case rates in various settings, but very few within a US ED. The paper is well written.

The investigators might consider some of the following comments to further enhance the manuscript:

1. In the study aims section (p. 4, last paragraph of the Introduction), it may be helpful to specify the study outcomes (even if only in parentheses) —“psychiatric presentations” is somewhat vague.

2. Throughout the Methods and Results, it might be useful to clarify the nature of the dependent variables. Did the analyses examine changes in the PERCENTAGES of psychiatric cases in the ED presenting with a given psychiatric diagnosis or SI, (e.g., the percentage of psychiatric cases with an affective disorder), or the raw NUMBER of these cases? Similarly, it may be helpful to indicate more explicitly for readers whether the repeated reference to “rates” signifies percentages. The same basic point applies to the Figures.

3. In the Outcome Measures section (p. 7), it may be helpful to define the outcomes more explicitly. For example, although the text on line 150 refers to “suicidality,” line 153 refers more specifically to “suicidal ideation.” More importantly, it would be useful to indicate how reports of previous as opposed to current SI were handled (presumably these were not counted); how intermittent use of substances were differentiated from diagnosed substance use disorders; and whether or how previous psychiatric disorders were differentiated from current ones. (In other words, was a patient who presented with a prior hx of recurrent major depression in full remission counted the same as one presenting with an active depressive disorder?).

4. In the Methods section (p. 7), it may be helpful to offer a brief rationale for the diagnoses selected as the outcomes for this study (e.g., why were mood disorders analyzed but not anxiety disorders?).

5. The regression model regarding suicidal ideation adjusted for various psychiatric disorders— it would be interesting as well to see the findings without these adjustments. Similarly, the analyses of psychiatric disorders adjusted for selected demographic variables and total number of psychiatric consultations; it might be useful to note the results without these adjustments (but of course retaining the time-related covariates).

6. The Results section (p. 8, line 175) reports that the overall volume of psychiatric consultations declined in the COVID-19 post period compared with the comparator post-period. Did the number of non-psychiatric consultations decline significantly as well? This information would provide readers with greater context for understanding the change in number of psychiatric consultations.

7. The Results section (p. 9, lines 190-191) indicates that “the rates of psychiatric presentation were similar between the two study groups in the pre-period.” However, Table 2 appears to indicate a significant change in the number of psychotic disorders; if so, it would be useful to mention this in the text.

8. In the Results section (p. 12, 1st paragraph), I confess that I had some difficulty following the calculation of percent increases (e.g., 30.7% vs. 558%). Perhaps the investigators might wish to expand the explanation a bit.

9 In the Discussion section, the authors might consider integrating their findings with other studies regarding changes in volume of psychiatric cases associated with the COVID-19 pandemic.

10. The study limitations section (pp. 15-16) makes some good observations. However, the large confidence intervals receive only fleeting reference, which does not really seem to do justice to this issue— it may be helpful to address this concern a bit more fully, including the ramifications of the very large CI’s for interpreting the findings.

11. Table 2 refers to “mean (%)” in the first column, but evidently only percentages are reported— it would be useful to include the frequencies as well, and to specify how the percentages were calculated (percentage of total number of psychiatric cases during that interval, percentage of total [psychiatric and non-psychiatric] ED visits during that interval, etc.?).

Table 1 reports percentages. It would be conventional to report the frequencies as well (though I appreciate that journals vary in their preferences and the editor may have a different view).

I was unable to find Appendix in the material I reviewed.

MINOR comments:

1. The use of the term “intervention” throughout the paper may be confusing to readers, since there is no intervention in the traditional clinical sense. Perhaps the authors might consider “exposure” instead.

2. On p. 7 (lines 146-147), the text states that information was extracted “automatically” for some variables and “manually” for others. It is not entirely clear what this means (e.g., the medical records could be queried for summary information regarding some variables, but information had to be manually extracted by searching each record individually for other variables?).

3. On p. 11 (lines 196-197), the text notes that Figure 3 summarizes the “adjusted models for each of the four psychiatric presentation groups.” It may be useful to specify the time period for these data (e.g., evidently the entire study period from Dec. to May).

4. On p. 11, in the last paragraph regarding changes in SI and SUDS, it would be helpful to reiterate that these analyses were adjusted (this was mentioned in the previous paragraph but not in this one).

5. On p. 3, line 57, it would be useful to note that the 30% in suicide rates associated with the 2003 SARS epidemic was reported in a study conducted in Hong Kong, if that is accurate (the previous clause refers to research in the US). Similarly, on p. 12 (lines 236), the last line on the page summarizes findings from the Faust et al. study-- it may be useful to specify that this investigation focused on rates in Massachusetts.

**********

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

[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.]

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 PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Jun 30;16(6):e0253805. doi: 10.1371/journal.pone.0253805.r002

Author response to Decision Letter 0

30 Apr 2021

Manuscript: PONE-D-21-02006

Title: “Evaluating the Association Between COVID-19 and Psychiatric Presentations, Suicidal Ideation in an Emergency Department: A Comparative Interrupted Time Series Study”

Response to Reviewers

Dear Editor:

We thank you for the thoughtful feedback on our manuscript. In the letter below, we outline how we have addressed the reviewers’ questions and concerns for our submission to PLOS ONE. We have additionally edited the manuscript to ensure that it follows guidelines outlined in your letter.

Again, thank you for your consideration of our manuscript.

Respectfully,

Michal J. McDowell, MD MPH

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

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Thank you for these helpful references. We have amended our article title, subtitle, and short title to be written in sentence case. We have amended our Level 2 headings to use sentence case, as well.

2. Thank you for including your ethics statement:

"- Mass General Brigham

- #2020P001482

- The IRB has determined that this project meets the criteria for exemption 45 CFR

46.101(b)(#)

- Consent was not obtained".

a. Please amend your current ethics statement to include the full name of the ethics committee/institutional review board(s) that approved your specific study.

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research

Thank you for this helpful reference. We have amended our ethics statement to include the full name of the institutional review board that approved our study. Additionally, we have added this statement to the first paragraph of the Methods section of the manuscript: “The Partners Institutional Review Board approved this work (approval #2020P001482) and waived the requirement for informed consent.”

3. Please provide additional details regarding participant consent.

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research."

Thank you for this helpful reference. We have amended our ethics statement to specify that the institutional review waived the requirement for informed consent. Additionally, we have added this statement to the first paragraph of the Methods section of the manuscript: “The Partners Institutional Review Board approved this work (approval #2020P001482) and waived the requirement for informed consent.”

Moreover, please specify the name of the centre where the research took place.

Thank you for this amendment. In the first paragraph of our methods section, we now specify the center at which the research took place, Massachusetts General Hospital. We also now clarify our inclusion criteria on page 6 of our Methods section, “Patients were included if they received a psychiatric consultation in any one of the four observation periods (pre and post-period for comparator time series or pre and post-period for COVID time series). Patients were allowed to appear in more than one of these periods. Patients that did not receive a psychiatric consultation (even if one was requested) were excluded from this study. Additionally, patients with incomplete demographic or diagnostic records were excluded from analyses that used the characteristic of interest, including CITS regression modeling.”

5. Please refer to the specific statistical analyses performed as well as any post-hoc corrections to correct for multiple comparisons. If these were not performed please justify the reasons.

Please refer to our statistical reporting guidelines for assistance (https://journals.plos.org/plosone/s/submission-guidelines.#loc-statistical-reporting).

Thank you for this helpful reference.

Regarding statistical details, we now specify the following:

• Our statistical testing was executed between subjects, as we now indicate in the Study design sub-section of the Methods section.

• Regarding assumptions, we now make further clarifications about our assumptions and checks surrounding linearity and normality.

o Regarding linearity of the outcomes over the study period, we now clarify our assumption of linearity at the end of the second paragraph of the Study design sub-section of our Methods section: “A general CITS assumes that any differential growth between the two groups is linear, and we conducted an event study to visually assess this assumption.”

o Regarding normality, we assume that the distribution of our outcomes is normal and that the relationship between the outcome and predictors is linear and additive. Typically, researchers use a non-linear regression model (e.g., logistic regression) when the outcome is binary, which assumes that the relationship between the predictors and outcome is multiplicative. We chose to use linear regression, as opposed to logistic regression to facilitate easy interpretation of the treatment effect estimates, which are interactions. In the first full paragraph of the Data analysis sub-section of our Methods section, we now give further background on how and why we estimated the linear probability models: “LPMs are preferred to logit or probit regression models with binary outcomes for both CITS and difference-in-differences, due to the ease of interpreting interactions in LPMs relative to logit or probit models.”

• There were no explicit outliers in our data. Regarding how we handled missing data, we now clarify that observations with missing data were excluded (fifth paragraph of the Study design sub-section of our Methods section): “Patients that did not receive a psychiatric consultation (even if one was requested) were excluded from this study. Additionally, patients with incomplete demographic or diagnostic records were excluded from analyses that used the characteristic of interest, including CITS regression modeling.”

• We also now clarify that we adjust for multiple comparisons testing with Bonferroni corrections (see paragraph 3 of the Data analysis sub-section of our Methods section).

• For our statistical tests, we now add descriptions our statistical tests in the first paragraph of the Data analysis sub-section of the Methods section: “First, we compared demographic and clinical characteristics among the pre-period (December – March) of each time series using a two-sided t-test to measure baseline similarity between these two groups. Then, we estimated linear probability models (ƒ-lines) for each of our four outcomes.”

Additionally, we now include the full regression output in the supplemental materials (see Appendix Table 2).

Regarding p-values, we now ensure that exact values are reported. We now provide both test statistics and p-values for each test.

Regarding percentages, we now ensure that we provide the numerator and denominator for all percentages.

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

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

Thank you for this helpful reference.

We have revised our cover letter to address the prompts indicated: “There are ethical restrictions on sharing our data set, as our data contains potentially identifying and sensitive patient information per the Partners Institutional Review Board. Data is available upon request. Requests may be sent to the corresponding author or to the Partners Institutional Review Board protocol administrator, Virginia Rodriguez VGRODRIGUEZ@PARTNERS.ORG. For reference, our protocol number for the review board is 2020P—1482.”

[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?

Reviewer #1: Yes

Thank you for the support regarding the technical rigor of the manuscript.

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

Reviewer #1: Yes

Thank you for the support regarding the statistical analysis performance.

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

Reviewer #1: Yes

Thank you, we agree that the data presented in the findings is available.

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

Reviewer #1: Yes

Thank you for the endorsement of the manuscript presentation.

5. Review Comments to the Author

Many thanks for the praise and kind feedback.

The investigators might consider some of the following comments to further enhance the manuscript:

Thank you for this thoughtful feedback. We have modified from “psychiatric presentations” to “patients presenting with psychiatric complaints” in the last paragraph of the Introduction section. We also further clarify this point in the first paragraph of the Study design sub-section of the Methods section: “We used an interrupted time series design to compare changes in the proportion of patients seen with certain psychiatric presentations before and after the COVID-19 outbreak.”

Thank you for this helpful suggestion. We have taken steps to clarify the nature of the dependent variables.

Our analyses examined changes in the percentages of psychiatric cases in the ED. Specifically, the numerator for our proportion is patients presenting with a given psychiatric diagnosis and the denominator is all patients seen by the psychiatric consult service in the ED.

We now clarify this point in the Outcome measures sub-section in the Methods section: “Our outcomes of interest were the proportion of patients who presented with suicidal ideation (SI), SUDs (including alcohol, opioid, cocaine, and stimulant use disorders, as well as substance-induced mood and psychotic disorders), psychotic disorders (including schizophrenia-spectrum illnesses and brief psychotic disorder), and affective disorders (including major depressive disorder, and bipolar affective disorder) among all patients seen by the psychiatric consult service.”

We also agree that indicating explicitly for readers that the rates signify percentages is important. We have changed the language throughout the document from “rate” to “proportion” to clarify.

Additionally, we have updated Fig 2 to add a clarification that the Y-axis is number based. We have updated Fig 3 to clarify that we indicate differential changes in proportion.

Thank you for this comment, which we believe helps clarify our outcome measures.

We agree that “suicidality” is not clear, so we have changed this term to “suicidal ideation” throughout the manuscript.

We agree that clarification surrounding various outcome variables is necessary. We now touch on how we handled past versus present suicidal ideation, substance use versus substance use disorder, and past versus present psychiatric diagnoses in the Outcome measures sub-section of the Methods section: “We only included present suicidal ideation in our outcome variable. For psychiatric diagnoses, we generally used the diagnosis given by the evaluating clinician. Historical labels were only used if they were deemed correct by the evaluating clinician and relevant to the current presentation. For SUDs, historical diagnoses were used if substances were involved in the current presentation or if there was compelling evidence of an SUD in the consult note.”

Thank you for this thoughtful suggestion – we agree that offering a rationale for the diagnoses we selected as the outcomes for this study would be useful for readers.

We now make this clarification in the Outcome measures sub-section of the Methods section: “We included suicidal ideation, SUDs, psychotic disorders, and affective disorders as our outcomes of interest because, based on our clinical experience, these diagnoses are reliably and consistently captured in our ED setting. Anxiety disorders, although coded by the reviewers, represent a small proportion of ED visits and unless they are the primary reason for presentation, they are more likely to be inconsistently captured.”

We also added further clarification regarding the means by which reviewers identified diagnoses in this sub-section: “Reviewers included relevant and current diagnoses identified in the consult notes, as well as interpreted symptoms mentioned with Diagnostic and Statistical Manual of Mental Disorders 5th edition (DSM-5) criteria in mind.”

Additionally, we included an addition to our limitation section regarding potential bias in reviewer interpretation of clinical notes, and thus outcome variables: “Fourth, in our generation of outcome variables, we relied on clinician reviewer interpretation of consultation notes, which are also represent interpretations of clinical presentations. We hope that the reviewers added a second layer of assessment, thus improving diagnostic clarity, though understand that the additional review may have introduced reviewer bias.”

We have also removed the bottom three rows of Table 2 (which detailed personality disorder, PTSD, and generalized anxiety disorder) to streamline outcome measure reporting.

Thank you for this thoughtful suggestion.

We agree that unadjusted analyses are often useful to explore.

We now include the regression model for suicidal ideation presentations for various psychiatric disorders with varying adjustments in Appendix Table 3. We now clarify this inclusion in the Data analysis sub-section of the Methods section: “For suicidal ideation, we also adjusted for the presence of an affective disorder, psychotic disorder, and SUD. As a sensitivity analysis, we estimate the same regression for suicidal ideation in Appendix Table 3, omitting all of the other diagnostic categories as covariates and adding back one of the diagnostic categories, since they may be on the causal pathway between exposure and suicidal ideation. Our results change modestly as a result of this different regression specification.”

We do not report findings without demographic adjustments because the demographic composition of cohorts may be changing differentially from pre- to post-period. We agree that clarifying why we do not present this analysis would be useful for the reader. We now include a sentence to this end in the Data analysis sub-section of the Methods section: “We do not present data not adjusted by demographic factors, as the demographic composition of cohorts may be change in different ways in the pre- to post-period.”

Of note, we corrected an error in Table 1 that we identified in reporting unadjusted comparisons of ethnicity in the pre-period. We corrected data related to this error throughout the manuscript.

Thank you for this excellent addition – we agree that understanding of total consultations would be helpful to understand as well.

We reached out to the EHR data team for data regarding total ED volumes, which we now include in the first paragraph of our Results section.

7. The Results section (p. 9, lines 190-191) indicates that “the rates of psychiatric presentation were similar between the two study groups in the pre-period.”

However, Table 2 appears to indicate a significant change in the number of psychotic disorders; if so, it would be useful to mention this in the text.

Thank you for highlighting this. We now have separated the paragraph in the Results section discussing Table 2. We draw the reader’s attention to the significant difference in psychotic disorder presentations in the pre-period for the 2018-2019 cohort as compared to the 2019-2020 cohort: “We found that fewer patients with psychotic disorders were presenting in the 2019-2020 pre-period cohort, as compared with the 2018-2019 pre-period cohort (p = 0.04).”

We agree that this was difficult to follow as written – thank you for highlighting. We have changed the wording in the two paragraphs following Figure 3 in the Results section and hope the language is now more accessible:

“We found differential changes in the adjusted proportion of presentations with suicidal ideation and SUDs. We also present unadjusted data in Appendix Table 3 for suicidal ideation. For psychiatric consultations with suicidal ideation present, we found a significant, differential increase six weeks after the exposure, or April 13-17, 2020 (36.4 percentage points change; 95% CI: 5.3, 67.6). Based on the comparison series, we would have expected an increase of 16.0 percentage points for suicidal ideation at week six. Relative to what we would have expected based on presentations from the previous year during week six, we observed a 127.5% increase in presentations with suicidal ideation following the coronavirus outbreak. We calculated the percent increases for suicidal ideation and SUD presentation by dividing the differential change in the post-period (i.e., the treatment effect estimate) by the difference in means from the pre-period to the post-period in the comparison group (i.e., the counterfactual).

There is also a differential increase in the adjusted proportion of presentations with SUD in the COVID-19 time series relative to the comparison time series at all time points after the exposure. This differential increase is statistically significant at three weeks following the exposure, or March 23-27, 2020 (32.8 percentage points; 95% CI: 4.0, 61.6). Based on the comparison series, we would have expected a decrease of 6.2 percentage points for substance use at week three. Relative to what we would have expected based on diagnoses from the previous year during week three, we saw a 629.0% increase in SUD presentations following the coronavirus outbreak.”

9 In the Discussion section, the authors might consider integrating their findings with other studies regarding changes in volume of psychiatric cases associated with the COVID-19 pandemic.

Thank you for this valuable addition. We agree that highlighting the literature exploring changes in volume of psychiatric cases associated with the COVID-19 pandemic is important.

We have extended our discussion of the relevant literature in the Discussion section: “While another recent Massachusetts-based study suggests that overall suicide rates did not increase during the pandemic, it appears that the experience of suicidal ideation may have increased at certain time points following the outbreak (13). This finding regarding increased pandemic-related suicidal ideation, as well as our finding regarding increased presentations with SUD, is corroborated by another recent large-scale cross-sectional study (32).”

Thank you for highlighting this important point. We agree, that emphasizing the large confidence intervals is important. However, after further examination and thought, we believe that these confidence intervals do not suggest that the study is underpowered (with the possible exception of the analysis with psychosis), as the treatment effect estimates do not appear to be unstably estimated zeroes. Nor do the confidence intervals preclude us from ruling out large effects in the opposite direction (i.e., in most instances, we can rule out large effects in one direction, but the effect may be larger than the point estimate).

Additionally, we noticed that the size of the confidence intervals grows over the course of the post-period, which is likely an effect of using a general CITS. Because general CITS uses data to estimate a growing difference between the two groups, this design introduces greater uncertainty than designs that assume no growing difference (e.g., a difference-in-differences), and this uncertainty grows as estimates get further from the intervention and the linearity assumption does more work.

We have attempted to address both points more fully in our Limitations sub-section in the Discussion section: “Fifth, while the confidence intervals for some of our results are large, they generally do not suggest that we are underpowered to detect an effect (i.e., an unstably estimated zero) with the exception of presentations with psychosis. In most cases, we can rule out large estimates in the opposite direction of the treatment effect, suggesting that the estimated treatment effects may be under-estimates of the increases or decreases seen in our study. Additionally, with a general CITS, estimates further from the intervention typically have larger confidence intervals, as uncertainty increases over the time period when estimating a between-group difference from the data.”

Thank you for this suggestion – we have amended the table to indicate “n (%)” for the values reported. We also now specify how the percentages were calculated, with the n value in the column and the total n in the column head to indicate numerator and denominator respectively.

12. Table 1 reports percentages. It would be conventional to report the frequencies as well (though I appreciate that journals vary in their preferences and the editor may have a different view).

Thank you for this recommendation – we agree that inclusion of the n, in addition to the percentages, will be helpful. We now include the n values in Table 1.

I was unable to find Appendix in the material I reviewed.

Thank you – we now include three Appendix tables in the Supplementary Materials.

MINOR comments:

We agree that the term “intervention” may be confusing to readers; we have changed this term to “exposure” throughout the manuscript.

Thank you for this helpful suggestion – we have amended the language in the Outcome measures sub-section in the Results section to clarify our data collection process: “In other words, the medical records could be queried for summary information for some variables, but information had to be manually extracted by searching each record individually for other variables.”

We agree that specifying the study period will be useful. We have amended the paragraph following Table 2 in the Results section to make this clarification: “Our adjusted models for each of the four psychiatric presentation groups (suicidal ideation, affective disorders, psychotic disorders, and SUDs) for the two study periods, which run from December to May in 2018-2019 and 2019-2020, respectively, are summarized in Fig 3, with tabular data presented in the Appendix Table 2.” Thank you for this helpful suggestion.

Thank you for this thoughtful comment – we have amended the last two paragraphs of the Results section to clarify that we are discussing “adjusted” proportions.

Thank you for this useful feedback. We now clarify that the 30% increase in suicide rates associated with the 2003 SARS epidemic was reported in a study conducted in Hong Kong. We also now specify that the Faust et al. study was conducted in Massachusetts.

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

Thank you for your thoughtful review of this manuscript.

Attachment

Submitted filename: Response to reviewers.docx

Click here for additional data file.^{(34.5KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0253805.r003

Decision Letter 1

Corstiaan den Uil

24 May 2021

PONE-D-21-02006R1

Evaluating the association between COVID-19 and psychiatric presentations, suicidal ideation in an emergency department

PLOS ONE

Dear Dr. McDowell,

Please submit your revised manuscript by Jul 08 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're 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.

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). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-emailutm_source=authorlettersutm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Corstiaan den Uil

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

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

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: (No Response)

**********

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

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?

Reviewer #1: Yes

**********

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

Reviewer #1: Yes

**********

6. Review Comments to the Author

Reviewer #1: I think the authors have done a commendable job with the revision— the clarifications and additional information were very helpful. This paper will be a nice contribution to the literature in this area.

I have only a few remaining minor comments for the investigators’ consideration. The page numbers refer to the tracked version rather than the clean version:

1. As previously noted, the study aims section (p. 4, last paragraph of the Introduction) is a bit vague and it seems that outcomes are only clarified for the reader 5 pages later (p. 10). It might be useful to change “psychiatric complaints” to “certain psychiatric complaints, and then to note in parentheses “(a number of selected substance use, mood, and psychotic disorders or suicidal ideation)” or similar phrasing-- in other words, provide a clearer sense of the study endpoints. Just a thought.

2. The statistical analysis section (top of p. 7) indicates that “a general rather than a linear CITS” was used to reduce assumptions, but the subsequent newly added sentence states that “A general CITS assumes that any differential growth… is linear.” For readers unschooled in CITS, like myself, that might seem at first blush like a logical contradiction, and perhaps the authors might take another look at those 2 sentences for clarity. Additionally, the new text on the bottom of the page states that “patients were allowed to appear in more than one of these periods,” and it might be helpful to briefly assure readers that this does not introduce problematic dependency in the data.

3. In the Results section regarding findings for the post-exposure periods, I think it would be important to indicate more explicitly that separate analyses were run for each of 10 weeks, if I understand correctly. That is, the text describing the significant results for SI at six weeks (p. 15, line 282), and the passage describing significant results for SUD at three weeks (p. 16, line 302) should note more clearly that each of these findings represent one of 10 analyses for each outcome, if that is accurate. (This approach is mentioned in the Data Analysis section of the Methods but should be specified in the Results.) .

4. On p. 15 in the Results section, there is ambiguity regarding the type of analysis described in the 3rd sentence on the page (lines 281-285), since the 1st sentence refers to adjusted model and the 2nd sentence refers to the unadjusted analyses. It may be helpful to clarify for readers that the 3rd sentence is once again referring to the adjusted analyses. In addition, should the text mention that there was an unexpected marginally significant decrease in SI at week 9? This finding does not seem to be mentioned in the Results section, and instead emerges for the first time in the Discussion (p. 18, lines 235-360). Was there a significant decrease in SI for those presenting with psychotic disorders, and was there a significant increase in SI for those from “other” racial/ethnic backgrounds relative to whites (as seems to be suggested in Appendix 2)?

5. The new text in the Results section includes a helpful summary of changes in SI based on unadjusted analyses (p. 16). Should the text also mention that there was a significant increase in SI at week 6 in the unadjusted model (as suggested in Appendix 3)?

6. In Appendix Table 2 and 3, it is not clear whether the model as a whole was significant for each outcome (i.e., the sample size and R square is presented for each outcome in Table 2, which is great, but not the significance level for the overall model for each outcome). More generally, I had a bit of difficulty interpreting this table, no doubt due to my limited familiarity with CITS, but the authors might wish to explain this Appendix more clearly (e.g., how does comparison level differ from differential level?).

7. The Discussion section indicates that there was a differential decrease in SI at nine weeks after the exposure (p. 18, lines 357-360). It might be useful to clarify that the text is referring to the adjusted analyses.

Other MINOR comments:

1. It might be helpful to specify the CITS abbreviation the first time the term is used in the paper (e.g., insert “CITS” in parentheses after “Comparative interrupted time series” on p. 6 line 100).

2. The change in terminology from “intervention” to “exposure” is appreciated. To avoid confusion, the authors might wish to make the same change in the remainder of the manuscript (i.e., p. 6, lines 110-112; p. 8, line 150; p. 11, lines 207 and 225; p. 15, line 293; p. 22, lines 439-441; Appendix Table 3) though they

3. On p. 6, line 107, it may be helpful to change “psychiatric presentation” to “selected psychiatric presentations” or “certain psychiatric presentations” or comparable language. On p. 11, the authors might wish to change the phrasing in the first and last paragraphs to the past tense.

4. On p. 12, for clarity, it would be useful to reiterate the time intervals for readers, e.g., by inserting in parentheses “(Dec. 2019- Feb. 2020)” after “COVID-19 pre-period” and by inserting “(Dec. 2018- Feb. 2019)” after “comparator pre-period.”

5. On p. 18, line 355, to facilitate interpretation it may be helpful to note the location of the study cited in this sentence (e.g., perhaps insert “in the US” after “cross-sectional study”).

6. On p. 20, it may be helpful to clarify the direction of the effects (perhaps by changing “White” to “fewer White,” and changing “Asian” to “more Asian” on line 397, and by inserting “in the 2019-2020 cohort” after “with affective disorders” on line 398).

7. Appendix Table 3, it may be useful to specify the demographic/temporal covariates that were included in all these analyses, in a note at the bottom of the table, to avoid a misimpression. In the title, after “Differential change in suicidal ideation presentations,” the authors might consider inserting “from ____ to ____” (i.e., specify the comparison conditions).

Good luck with your work!

**********

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

PLoS One. 2021 Jun 30;16(6):e0253805. doi: 10.1371/journal.pone.0253805.r004

Author response to Decision Letter 1

8 Jun 2021

Many thanks for your generous comments on our manuscript! We believe they have significantly improved the quality of the work we submit now.

I have only a few remaining minor comments for the investigators’ consideration. The page numbers refer to the tracked version rather than the clean version:

We agree that our aims in the last paragraph of the introduction could use clarification. We appreciate your suggestions and have included the word “certain” to qualify “psychiatric disorders.” We also now indicate in parentheses the outcomes of interest: “We used the period following the initial outbreak of COVID-19 in Massachusetts to assess the association between the pandemic and patients presenting with certain psychiatric complaints (a number of selected mood, psychotic, and substance use disorders, as well as suicidal ideation) to a tertiary care hospital Emergency Department (ED).”

Thank you for highlighting these points, which we agree would all benefit from clarification.

A linear CITS assumes that the evolution of the outcome (i.e., the probability of a patient showing up to the ED with a psychiatric presentation) is linear over the course of the study, which makes a parametric assumption about the post-period outcomes (i.e., that it follows the form of a line). However, a general CITS does not make an assumption about the post-period outcomes follow any shape (liner, quadratic, etc.). In a general CITS, the outcome is not assumed to be of a specific functional form. Instead, general CITS makes the assumption that the difference between the two groups is evolving linearly.

To clarify this point to our readers, we have included the following changes in the manuscript text in the second full paragraph of the Methods section, Study design sub-section: “Finally, we use a general CITS, rather than a linear CITS, to minimize the parametric assumptions made regarding the outcome’s evolution. A linear CITS assumes that the outcome is evolving linearly in the pre- and post-period (24). On the other hand, a general CITS assumes that any difference between the two groups is evolving linearly but does not make any assumption about the outcome’s parametric form (i.e., does not assume a linear, quadratic, or other form). We conducted an event study to visually assess the plausibility of these assumptions and found differential trends (i.e., non-parallel) in the pre-period suggesting that a standard difference in difference approach would not be appropriate.”

Regarding the issue of multiple presentations, we treat each of the four periods (2018-2019 pre, 2018-2019 post, 2019-2020 pre, 2019-2020 post) as a cohort in this cross-sectional study. We do make adjustments for patients presenting multiple times in one of the study periods, which we now specify in the antepenultimate paragraph in the Methods section, Study design sub-section: “We adjusted the models for the total number of emergency department visits per patient per study period.”

Thank you for this helpful suggestion. We agree that this point certainly needs clarification.

We did not run a separate analysis for each of the 10 weeks in the post-period. Rather, our regression specification includes an indicator variable for each post-period week. We include the indicator variable and an interaction between the treatment variable and the weekly indicator variable. Thus, each of these interactions represent the differential change for this week between the COVID-19 series and comparator series from the average pre-period level. We include our single regression formula at the beginning of the manuscript’s Supplemental Material to provide further explanation of the regression analysis conducted. In addition, we removed the “s” from “general CITS regression(s)” at the beginning of our Supplemental materials section to clarify that the analyses presented derive from a single regression specification.

We have added a sentence in the last paragraph of the Methods section, Data analysis sub-section to further clarify this point, as well: “In other words, our regression specification included an indicator variable for each post-period week. We and interacted the indicator variable with the treatment variable, and each of these interactions represent the differential change for the week between the COVID-19 series and comparator series as compared to the average pre-period levels.”

We agree that the first sentences in the Results section paragraph following Fig 3 were confusing to follow. We have improved clarity by reordering the sentences such that we mention the unadjusted models first, and then discuss the adjusted models following.

Thank you for the suggestion that we name the non-significant decrease in SI at week 9 in the Results section. We now do this at the end of the Results section paragraph following Fig 3: “Additionally, at nine weeks after the exposure (May 4-9, 2020), there was a differential decrease in the proportion of psychiatric presentations to the ED with suicidal ideation (34.9 percentage points; 95% CI; -69.5, -0.3), though this change was not statistically significant (p = 0.06).”

Your questions regarding Appendix Table 2 are helpful. In the “Any SUD,” “Any affective disorder,” “Any psychotic disorder” and “Other” race/ethnic background rows, you are correct that the confidence intervals do not cross 0 for the suicidal ideation column. This indicates that, over the course of the study, after accounting for all other covariates, including the exposure (COVID-19 outbreak), there is a relationship between suicidal ideation and other psychiatric conditions and/or demographic characteristics. In other words, the probability of presentation with suicidal ideation increases with the presence of co-occurring SUD, increases with co-occurring affective disorders, decreases with co-occurring psychotic disorders, and increases in people with “other” race/ethnic backgrounds. These changes do not have do to with COVID.

We are happy to provide additional text clarifications on Appendix Table 2 at the Editor and Reviewer’s discretion.

Thank you for highlighting this point, which could certainly use clarification. We now include more explanation regarding Appendix Table 3 in the paragraph following Fig 3 in the Results section: “We present data in Appendix Table 3 for suicidal ideation, sequentially removing and re-adding other certain psychiatric presentations to the model. Our results change modestly as a result of this different regression specification. Specifically, there is a significant increase in presentations with suicidal ideation at weeks three and six. However, once adjustment for SUD is included in the model at week three, the differential change is no longer significant because this increase was mediated by the co-occurring SUD diagnoses. However, even after the adjustments for co-occurring psychiatric conditions are added at week six, the differential change persists.”

We agree that Appendix Tables 2 and 3 may be challenging to follow.

First, we now include the global F-test statistic, p-value, and degrees of freedom to Appendix Table 2 and 3. In Appendix Table 2, the F-test statistics demonstrate that for suicidal ideation, affective disorders, and SUD, the models are statistically significant. For psychotic disorders, the overall model is not statistically significant, suggesting that this model does not explain the variation in the outcome better than the intercept-only model, suggesting that changes in the presence of psychotic disorder may not be attributable to COVID-19 and may be more complex than the current regression models can explain.

Second, we have re-arranged Appendix Table 2 to first display the quantities of interest (i.e., the treatment effects).

Third, we have added text in the notes of both Tables to help readers understand what these estimates represent: “This is the difference between the change in proportion of ED visits with a psychiatric presentation from the pre-period to post-period for the COVID-19 series and the change in the proportion of ED visits with a psychiatric presentation from the pre-period to the post-period in the comparison series (quantity of interest).”

We agree that this clarification would prove helpful. We now include the qualifier “in the adjusted analyses,” in this sentence.

Other MINOR comments:

Thank you for this catch! We now include the abbreviation after the first time the term is used in the paper.

Thank you for this helpful suggestion – we have changed the terms from “intervention” to “exposure” throughout the manuscript.

Thank you, we agree these clarifications would be useful for readers. We have amended to include “certain.” We have also amended the two indicated paragraphs to read in the past tense.

Thank you for this helpful suggestion. We now include the following clarifications: In exploring pre-period differences, we found that the COVID-19 pre-period (December 23, 2019 – February 24, 2020) cohort had a higher proportion of psychiatric ED consultations involving people from a minority racial group than did the comparator pre-period (December 24, 2018 – February 25, 2019) (Table 1).”

5. On p. 18, line 355, to facilitate interpretation it may be helpful to note the location of the study cited in this sentence (e.g., perhaps insert “in the US” after “cross-sectional study”).

We agree that this would be a helpful clarification. The sentence now reads: “This finding regarding increased pandemic-related suicidal ideation, as well as our finding regarding increased presentations with SUD, is corroborated by another recent large-scale cross-sectional study in the United States (32).”

We certainly agree that this clarification would be helpful to readers. We have made the suggested edits and the sentence now reads: “When we further explore this difference by separating the groups by psychiatric presentations, statistically significant differences are isolated to fewer White patients presenting with psychotic disorders and more Asian patients presenting with affective disorders in the 2019-2020 cohort.”

Thank you for this helpful suggestion. We have amended the title of the table to better represent what we are showing in the table and have added notes to the table to further explain our covariate adjustment and the meaning of the coefficients in the table.

Good luck with your work!

Many thanks for your patience and generosity in reviewing and offering suggestions to our paper.

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(22.1KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0253805.r005

Decision Letter 2

Corstiaan den Uil

14 Jun 2021

Evaluating the association between COVID-19 and psychiatric presentations, suicidal ideation in an emergency department

PONE-D-21-02006R2

Dear Dr. McDowell,

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

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. 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 help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- 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.

Kind regards,

Corstiaan den Uil

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0253805.r006

Acceptance letter

Corstiaan den Uil

23 Jun 2021

PONE-D-21-02006R2

Evaluating the association between COVID-19 and psychiatric presentations, suicidal ideation in an emergency department

Dear Dr. McDowell:

I'm 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 let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, 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.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Corstiaan den Uil

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Material

(DOCX)

Click here for additional data file.^{(13.1KB, docx)}

S1 Table. Pre-period unadjusted comparison of racial/ethnic groups by psychiatric presentation.

(DOCX)

Click here for additional data file.^{(14KB, docx)}

S2 Table. Full regression output for differential changes in emergency department psychiatric presentations in the 2019 versus 2020 post-periods.

(DOCX)

Click here for additional data file.^{(22.1KB, docx)}

(DOCX)

Click here for additional data file.^{(16.4KB, docx)}

Attachment

Submitted filename: Response to reviewers.docx

Click here for additional data file.^{(34.5KB, docx)}

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(22.1KB, docx)}

Data Availability Statement

[pone.0253805.ref001] 1.Gunnell D, Appleby L, Arensman E, Hawton K, John A, Kapur N, et al. Suicide risk and prevention during the COVID-19 pandemic. Lancet Psychiatry. 2020;7: 468–471. doi: 10.1016/S2215-0366(20)30171-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref002] 2.Yao H, Chen J-H, Xu Y-F. Patients with mental health disorders in the COVID-19 epidemic. Lancet Psychiatry. 2020;7: e21. doi: 10.1016/S2215-0366(20)30090-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref003] 3.Sherman AC, Williams ML, Amick BC, Hudson TJ, Messias EL. Mental health outcomes associated with the COVID-19 pandemic: Prevalence and risk factors in a southern US state. Psychiatry Res. 2020;293: 113476. doi: 10.1016/j.psychres.2020.113476 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref004] 4.Chua SE, Cheung V, McAlonan GM, Cheung C, Wong JWS, Cheung EPT, et al. Stress and psychological impact on SARS patients during the outbreak. Can J Psychiatry Rev Can Psychiatr. 2004;49: 385–390. doi: 10.1177/070674370404900607 [DOI] [PubMed] [Google Scholar]

[pone.0253805.ref005] 5.Wasserman IM. The impact of epidemic, war, prohibition and media on suicide: United States, 1910–1920. Suicide Life Threat Behav. 1992;22: 240–254. [PubMed] [Google Scholar]

[pone.0253805.ref006] 6.Cheung YT, Chau PH, Yip PSF. A revisit on older adults suicides and Severe Acute Respiratory Syndrome (SARS) epidemic in Hong Kong. Int J Geriatr Psychiatry. 2008;23: 1231–1238. doi: 10.1002/gps.2056 [DOI] [PubMed] [Google Scholar]

[pone.0253805.ref007] 7.Oyesanya M, Lopez-Morinigo J, Dutta R. Systematic review of suicide in economic recession. World J Psychiatry. 2015;5: 243–254. doi: 10.5498/wjp.v5.i2.243 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref008] 8.Van Orden KA, Witte TK, Cukrowicz KC, Braithwaite SR, Selby EA, Joiner TE. The interpersonal theory of suicide. Psychol Rev. 2010;117: 575–600. doi: 10.1037/a0018697 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref009] 9.VanderWeele TJ, Li S, Tsai AC, Kawachi I. Association Between Religious Service Attendance and Lower Suicide Rates Among US Women. JAMA Psychiatry. 2016;73: 845–851. doi: 10.1001/jamapsychiatry.2016.1243 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref010] 10.Wasserman D, Iosue M, Wuestefeld A, Carli V. Adaptation of evidence-based suicide prevention strategies during and after the COVID-19 pandemic. World Psychiatry. 2020;19: 294–306. doi: 10.1002/wps.20801 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref011] 11.Laderman M, Komaromy M, Moses J, Jarrett M. Under the cover of COVID-19 news, the epidemic of opioid overdoses silently surges. In: Becker’s Hospital Review [Internet]. 30 Oct 2020. [cited 4 Nov 2020]. Available: https://www.beckershospitalreview.com/opioids/under-the-cover-of-covid-19-news-the-epidemic-of-opioid-overdoses-silently-surges.html. [Google Scholar]

[pone.0253805.ref012] 12.Reger MA, Stanley IH, Joiner TE. Suicide Mortality and Coronavirus Disease 2019—A Perfect Storm? JAMA Psychiatry. 2020;77: 1093. doi: 10.1001/jamapsychiatry.2020.1060 [DOI] [PubMed] [Google Scholar]

[pone.0253805.ref013] 13.Faust JS, Shah SB, Du C, Li S-X, Lin Z, Krumholz HM. Suicide Deaths during the Stay-at-Home Advisory in Massachusetts. medRxiv. 2020; 2020.10.20.20215343. doi: 10.1101/2020.10.20.20215343 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref014] 14.Halford EA, Lake AM, Gould MS. Google searches for suicide and suicide risk factors in the early stages of the COVID-19 pandemic. PloS One. 2020;15: e0236777. doi: 10.1371/journal.pone.0236777 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref015] 15.Czeisler MÉ. Mental Health, Substance Use, and Suicidal Ideation During the COVID-19 Pandemic—United States, June 24–30, 2020. MMWR Morb Mortal Wkly Rep. 2020;69. doi: 10.15585/mmwr.mm6932a1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref016] 16.Sher L. COVID-19, anxiety, sleep disturbances and suicide. Sleep Med. 2020;70: 124. doi: 10.1016/j.sleep.2020.04.019 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref017] 17.Taquet M, Luciano S, Geddes JR, Harrison PJ. Bidirectional associations between COVID-19 and psychiatric disorder: retrospective cohort studies of 62 354 COVID-19 cases in the USA. Lancet Psychiatry. 2020;0. doi: 10.1016/S2215-0366(20)30462-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref018] 18.Rossi R, Socci V, Talevi D, Mensi S, Niolu C, Pacitti F, et al. COVID-19 Pandemic and Lockdown Measures Impact on Mental Health Among the General Population in Italy. Front Psychiatry. 2020;11. doi: 10.3389/fpsyt.2020.00790 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref019] 19.Farhoudian A, Baldacchino A, Clark N, Gerra G, Ekhtiari H, Dom G, et al. COVID-19 and Substance Use Disorders: Recommendations to a Comprehensive Healthcare Response. An International Society of Addiction Medicine (ISAM) Practice and Policy Interest Group Position Paper. Basic Clin Neurosci. 2020;11: 133–146. doi: 10.32598/bcn.11.covid19.1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref020] 20.Pedrosa AL, Bitencourt L, Fróes ACF, Cazumbá MLB, Campos RGB, de Brito SBCS, et al. Emotional, Behavioral, and Psychological Impact of the COVID-19 Pandemic. Front Psychol. 2020;11. doi: 10.3389/fpsyg.2020.566212 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref021] 21.Ochalek TA, Cumpston KL, Wills BK, Gal TS, Moeller FG. Nonfatal Opioid Overdoses at an Urban Emergency Department During the COVID-19 Pandemic. JAMA. 2020;324: 1673. doi: 10.1001/jama.2020.17477 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref022] 22.Coronavirus Disease (COVID-19) timeline. In: Boston.gov [Internet]. 15 Mar 2020 [cited 4 Nov 2020]. Available: https://www.boston.gov/departments/public-health-commission/coronavirus-timeline.

[pone.0253805.ref023] 23.Humphreys DK, Gasparrini A, Wiebe DJ. Evaluating the Impact of Florida’s “Stand Your Ground” Self-defense Law on Homicide and Suicide by Firearm: An Interrupted Time Series Study. JAMA Intern Med. 2017;177: 44–50. doi: 10.1001/jamainternmed.2016.6811 [DOI] [PubMed] [Google Scholar]

[pone.0253805.ref024] 24.Fry CE, Hatfield LA. Do Methodological Birds of a Feather Flock Together? ArXiv200611346 Econ Stat. 2020 [cited 2 Jul 2020]. Available: http://arxiv.org/abs/2006.11346.

[pone.0253805.ref025] 25.Shadish WR, Cook TD, Campbell DT. Quasi-Experimentation: Interrupted Time Series Designs. 2nd ed. Experimental and Quasi-Experimental Designs for Generalized Causal Inference. 2nd ed. Cengage; 2002. doi: 10.1037/1082-989x.7.1.3 [DOI] [Google Scholar]

[pone.0253805.ref026] 26.Woo J-M, Okusaga O, Postolache TT. Seasonality of Suicidal Behavior. Int J Environ Res Public Health. 2012;9: 531–547. doi: 10.3390/ijerph9020531 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref027] 27.Lemieux J, Siddle KJ, Shaw BM, Loreth C, Schaffner S, Gladden-Young A, et al. Phylogenetic analysis of SARS-CoV-2 in the Boston area highlights the role of recurrent importation and superspreading events. medRxiv. 2020; 2020.08.23.20178236. doi: 10.1101/2020.08.23.20178236 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref028] 28.Karaca-Mandic P, Norton EC, Dowd B. Interaction terms in nonlinear models. Health Serv Res. 2012;47: 255–274. doi: 10.1111/j.1475-6773.2011.01314.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref029] 29.Health Equity Considerations & Racial & Ethnic Minority Groups. In: www.cdc.gov [Internet]. 30 Apr 2020 [cited 28 Oct 2020]. Available: https://www.cdc.gov/coronavirus/2019-ncov/community/health-equity/race-ethnicity.html.

[pone.0253805.ref030] 30.Yancy CW. COVID-19 and African Americans. JAMA. 2020;323: 1891. doi: 10.1001/jama.2020.6548 [DOI] [PubMed] [Google Scholar]

[pone.0253805.ref031] 31.Trends in Number of COVID-19 Cases in the US Reported to CDC, by State/Territory. In: www.cdc.gov [Internet]. [cited 4 Nov 2020]. Available: https://www.cdc.gov/covid-data-tracker.

[pone.0253805.ref032] 32.Holland KM, Jones C, Vivolo-Kantor AM, Idaikkadar N, Zwald M, Hoots B, et al. Trends in US Emergency Department Visits for Mental Health, Overdose, and Violence Outcomes Before and During the COVID-19 Pandemic. JAMA Psychiatry. 2021. doi: 10.1001/jamapsychiatry.2020.4402 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref033] 33.Baker-Polito Administration Extends Non-Essential Business Closures to May 18th, Announces Reopening Advisory Board | Mass.gov. [cited 12 Nov 2020]. Available: https://www.mass.gov/news/baker-polito-administration-extends-non-essential-business-closures-to-may-18th-announces.

[pone.0253805.ref034] 34.Baker-Polito Administration Announces Four-Phase Approach to Reopening and Publishes Mandatory Workplace Safety Standards | Mass.gov. [cited 12 Nov 2020]. Available: https://www.mass.gov/news/baker-polito-administration-announces-four-phase-approach-to-reopening-and-publishes-mandatory.

[pone.0253805.ref035] 35.Slavova S, Rock P, Bush HM, Quesinberry D, Walsh SL. Signal of increased opioid overdose during COVID-19 from emergency medical services data. Drug Alcohol Depend. 2020;214: 108176. doi: 10.1016/j.drugalcdep.2020.108176 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref036] 36.Wakeman SE, Green TC, Rich J. An overdose surge will compound the COVID-19 pandemic if urgent action is not taken. Nat Med. 2020;26: 819–820. doi: 10.1038/s41591-020-0898-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref037] 37.Torchalla I, Strehlau V, Li K, Aube Linden I, Noel F, Krausz M. Posttraumatic stress disorder and substance use disorder comorbidity in homeless adults: Prevalence, correlates, and sex differences. Psychol Addict Behav. 2014;28: 443–452. doi: 10.1037/a0033674 [DOI] [PubMed] [Google Scholar]

[pone.0253805.ref038] 38.Baggett TP, Keyes H, Sporn N, Gaeta JM. Prevalence of SARS-CoV-2 Infection in Residents of a Large Homeless Shelter in Boston. JAMA. 2020;323: 2191–2192. doi: 10.1001/jama.2020.6887 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref039] 39.Wosik J, Fudim M, Cameron B, Gellad ZF, Cho A, Phinney D, et al. Telehealth transformation: COVID-19 and the rise of virtual care. J Am Med Inform Assoc. 2020;27: 957–962. doi: 10.1093/jamia/ocaa067 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0253805.ref040] 40.Castro VM, Perlis RH. Electronic Health Record Documentation of Psychiatric Assessments in Massachusetts Emergency Department and Outpatient Settings During the Coronavirus Disease 2019 (COVID-19) Pandemic. JAMA Netw Open. 2020;3: e2011346. doi: 10.1001/jamanetworkopen.2020.11346 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Evaluating the association between COVID-19 and psychiatric presentations, suicidal ideation in an emergency department

Michal J McDowell

Carrie E Fry

Mladen Nisavic

Mila Grossman

Charles Masaki

Emily Sorg

Suzanne Bird

Felicia Smith

Scott R Beach

Roles

Abstract

Objective

Methods

Results

Conclusions

Introduction

Methods

Study design

Fig 1. Study period timeline.

Outcome measures

Data analysis

Results

Fig 2. Comparison of unadjusted trends in psychiatry consults between 2019 and 2020 cohorts.

Table 1. Pre-period unadjusted comparison of demographic characteristics.

Table 2. Pre-period unadjusted comparison of psychiatric presentations.

Fig 3. Differential changes in ED psychiatry consult presentations in the COVID versus non-COVID post periods.

Discussion

Limitations

Conclusions

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Corstiaan den Uil

Roles

Author response to Decision Letter 0

Decision Letter 1

Corstiaan den Uil

Roles

Author response to Decision Letter 1

Decision Letter 2

Corstiaan den Uil

Roles

Acceptance letter

Corstiaan den Uil

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases