International Journal of Academic Medicine

: 2021  |  Volume : 7  |  Issue : 1  |  Page : 22--29

Emergency mental health calls to first responders following a natural disaster: Examining the effects from Hurricane Harvey

John Saunders1, Deepa Dongarwar2, Jason Salemi3, Joan Schulte4, David Persse4, Asna Matin1, Sophia Banu1, Asim Shah1,  
1 Menninger Department of Psychiatry, Baylor College of Medicine, Tampa, FL, USA
2 Baylor College of Medicine Center of Excellence in Health Equity and Research, Tampa, FL, USA
3 Department of Family and Community Medicine, Baylor College of Medicine; Department of Gynecology, College of Public Health, Morsani College of Medicine, University of South Florida, Tampa, FL, USA
4 Department of Houston Health, Houston, Tx, USA

Correspondence Address:
Dr. John Saunders
One Baylor Plaza, Houston, TX 77030


Introduction: Hurricane Harvey which made landfall on August 25, 2017 was a devastating storm that dumped unprecedented amount of rainfall on the area including Houston, Texas, United States of America. There are limited data about emergency service mental health utilization following disaster events. The goal for this project was to examine mental health calls to emergency medical services (EMS) and to the Houston Police Department following Hurricane Harvey. An analysis looking at this utilization following a natural disaster represents an understudied area and can potentially provide information about city services and community psychiatric services in the acute period following the event. Materials and Methods: Total number of calls to the police department and mental health calls to the police department described as crisis intervention calls (Crisis Intervention Team) were obtained from August 1, 2016 to October 31, 2016 and January 1, 2017 to December 31, 2017. Emergency detention orders (EDO) per date were obtained from January 1, 2017 to December 31, 2017. Data for mental health calls based on the primary impression of mental health complaint were obtained from the Houston Fire Department for EMS from January 1, 2016 to December 31, 2017. Results: There was a statistically significant increase in the number of mental health calls to the police following Hurricane Harvey. When comparing the prestorm, active storm, and poststorm period, there was not a statistically significant difference in the number of EDOs or the number of EMS mental health calls. Conclusions: The increase in police mental health calls suggests that there may have been an increase in the acuity of the mental health calls to EMS around in evaluating calls surrounding the period of Hurricane Harvey. The following core competencies are addressed in this article: Medical Knowledge and Patient Care.

How to cite this article:
Saunders J, Dongarwar D, Salemi J, Schulte J, Persse D, Matin A, Banu S, Shah A. Emergency mental health calls to first responders following a natural disaster: Examining the effects from Hurricane Harvey.Int J Acad Med 2021;7:22-29

How to cite this URL:
Saunders J, Dongarwar D, Salemi J, Schulte J, Persse D, Matin A, Banu S, Shah A. Emergency mental health calls to first responders following a natural disaster: Examining the effects from Hurricane Harvey. Int J Acad Med [serial online] 2021 [cited 2021 Apr 12 ];7:22-29
Available from:

Full Text


Hurricane Harvey was a devastating event for the region of Southeast Texas. It made landfall in the latter portion of August 2017 as a category 4 storm and dumped over 50 inches of rain in some portions of the affected area. Hurricane Harvey damaged large and small communities alike and included unprecedented rainfall in the Houston area. Some have estimated that the damage done by Hurricane Harvey exceeded $150 billion.[1] During the storm, two large centers were used as shelters: George R Brown Convention Center (GRB) and NRG Stadium (NRG). At one time, up to 10,000 people were sheltered at GRB.[2] Individuals with mental health needs, including mood disorders, schizophrenia and schizoaffective disorder, anxiety disorder, substance use disorders, and posttraumatic stress disorder (PTSD) were able to receive treatment while at the shelters.[2] Previous work looking at mental illness following Hurricane Katrina found significant increases in the prevalence of mild-to-moderate as well as serious mental illnesses.[3] Although preexisting substance use disorders can be exacerbated, it has been suggested that PTSD is the disorder most often associated with disasters.[4]

Considering the potential for escalations in mental health acuity and the significant strain that can be imposed by massive flooding events like Hurricane Katrina, an established health-care infrastructure is critical during natural disasters.[5] (1) An element of that infrastructure that is vital are first responders, specifically police and emergency medical services (EMS). These first responders are the first point of contact for many people in mental health emergencies. In the United States, it has been estimated that one-third of patients seen in emergency rooms are transported for behavioral health concerns.[6] Moreover, previous studies estimated that 30% of the over 7 million annual mental health visits are transported by EMS.[7] To understand the general scale of these interactions, in Boston in 2017, it was noted that of the 681,546 total police calls for service in 5953 specifically involved people with mental illness, though it was suggested that this may undercount the total number of police interactions with those with a mental illness.[8] A report involving 174 police departments throughout the United States that serve populations of over 100,000 people indicated that 7% of all police contacts involve persons who were believed to have a mental illness.[9] A review article noted that one in ten people with mental illness may interact with police as they attempt to assess services.[10] These interactions have contributed to the formation of various police models to address the mental health needs of the people they serve and these models having varying levels of perceived effectiveness.[9] One of the best-known models is the Crisis Intervention Team (CIT) model, which is used by multiple police departments including the Houston Police Department. It promotes the development of techniques for officer de-escalation, communication, suicide prevention, and witness assistance.[11] One of the goals of the CIT model is to divert individuals with mental illnesses from the criminal justice system into treatment, if possible,[12] often by using the “emergency commitment.” The grounds for emergency commitment vary across states, but all states recognize the need to mandate emergent treatment for mental health crises at times.[13] This speaks consistently to the level of acuity of a person's presentation, as the vast majority of these laws require a component of dangerousness.[13] Within Texas, peace officers are empowered by Texas State Statute 573.001, which allows for the transportation and evaluation of persons with a mental illness, determined to be a risk to themselves or others, or with evidence of severe mental distress or deterioration. Those orders are referred to as Emergency Detention Orders (EDO).[14]

Not all disasters are natural, and man-made disasters including terrorist events affect mental health. Previous studies have shown that in the initial period after the terrorist attacks on September 11, 2001, there was an increase in both mental health-related 911 calls and involuntary psychiatric evaluations initiated by law enforcement. Both of these studies observed this effect outside of New York City and in other parts of the United States.[15],[16] However, considering the relative rarity of natural and man-made disasters, we lack comprehensive information regarding community response to the mental health crises imposed by these events. The purpose of this study was to evaluate changes in the volume of mental health-related calls to EMS and police within Houston, Texas, before, during, and after Hurricane Harvey in 2017, to better understand the mental health utilization of first responders following a natural disaster.


Data sources

Data specific to incoming emergency calls were collected from both the Houston Fire Department for EMS electronic patient care record system (EPCRS) and the Houston Police Department. From the EMS EPCRS, for all mental health-related calls made between May 24, 2017 and December 31, 2017, we were able to extract call-level information including the chief complaint, call date and time, address from which the call was placed, and the age, race, and gender of the caller. Mental health-related calls were defined using the caller's primary complaint as determined by on-site EMS responders.

We also collected call-level information from the Houston Police Department on all mental health-related calls made between January 1, 2017 and December 31, 2017. Data elements included call type, disposition of the caller, the address from which the call was placed, the beats assigned, and whether an EDO was written by responding officers. Mental health-related calls were defined as those flagged as “CIT calls.” We were also able to collect aggregated daily call volume for all call types (not restricted to CIT calls) to the police department. Considering the possibility of nonhurricane-related seasonality to impact temporal changes in call volume, we were also able to ascertain similar information on CIT calls and any type police calls in the year before the hurricane, from August 1, 2016 to October 31, 2016.

Hurricane Harvey-relevant time periods

Since the primary focus of this study was to assess the extent to which Hurricane Harvey impacted mental health-related call volume, we created three time periods based on the date of the call made to Houston Fire Department EMS or the Houston Police Department. The “prestorm” period was defined as any call made prior on or before August 22, 2017; for Houston Fire Department and EMS data, this starts on May 24, 2017. The Houston Police Department data “prestorm” period include data from August 1, 2016 to October 31, 2016, as well as January 1, 2017 to August 22, 2017. The “active storm” period was defined as calls made between August 23, 2017 and September 15, 2017, inclusive. This timeframe corresponds to the dates used to define the Hurricane Harvey incident period identified by the Federal Emergency Management Agency, DR 4332.[17] The “poststorm” period was defined as calls made on or after September 16, 2017 to December 31, 2017 for both Houston Fire Department EMS and Houston Police Department data.

Statistical analyses

Descriptive statistics were used to compare mental health-related call characteristics across the prestorm, active storm, and poststorm periods. For both EMS and Police Department calls, we assessed total call volume and mean daily call volume, using a one-way analysis of variance to determine whether there were statistically significant differences in mean mental health-related call volume across the three time periods. T-tests with a Bonferroni's correction for multiple group comparisons were used to further explore differences between specific time groups (e.g., pre vs. active, active vs. post, and pre vs. post). For police department calls, we also assessed differences in mean daily call volume for any reason (mental health or otherwise) across the three time periods and compared the extent to which the proportion of all calls that were flagged as CIT changed across the hurricane-relevant time periods.

In addition to daily call volume, for EMS calls, we also compared characteristics of mental health-related calls across the three time periods. Call location was defined as “home” or “other,” call gender was male, female, or other, caller race/ethnicity grouped as non-Hispanic (NH) White, NH-Black, Hispanic, and other, and caller age in years was categorized as younger than 16, 16–25, 26–35, 36–45, 46–55, 56–65, and 66 or older. The incident complaint reported that by dispatch complaints, diagnoses were broken down into six categories for analysis as taken from the complaint listed in the report (1) psychiatric, (2) cardiovascular or respiratory problems which comprised of heart problems or breathing problems, (3) neurologic which comprised of seizures, convulsions, unconsciousness, stroke, cerebrovascular accident, syncope, fall victim, or headache, (4) substance use or ingestion which included cases of overdose or alcohol related, (5) injury comprising of lacerations, injuries, assault, sexual assault, transportation accident, hemorrhage, lacerations, fall, hanging, sting, or venomous bite, and (6) other which included cases of diabetic problem, abdominal pain, sick person, unknown problem, walk in, assist police with transport, allergic reaction, and nontraumatic back pain. Chi-square tests of statistical independence were used to assess statistically significant differences in the distribution of caller characteristics over time.

All statistical analyses were performed using R (version 3 · 5 · 2) RStudio (version 1.1.463) (Rstudio Inc., Boston, Massachusetts, USA). All tests of hypotheses were two-tailed with type-1 error rate set at 5%. The study was approved by the Institutional Review Board of Baylor College of Medicine and the protocol number was H-42593.


Police mental health calls

There were 1,131,149 total calls to the police department during the calendar year 2017. Of these, 729,514 (64.5%) were during the prestorm period, 79,016 (7.0%) during the active storm period, and 322,619 (28.5%) during the poststorm period [Table 1]. The mean daily call volume did not change significantly between the prestorm and active storm periods; however, there was a statistically significant decrease observed between prestorm and poststorm periods (P < 0.01). Approximately 3% of total police calls were mental health related (i.e., CIT calls). Moreover, there was a small but statistically significant (P < 0.01) increase in mean CIT calls per day from the prestorm (88) to the active storm (91) and poststorm (94) periods. When we compared the 3-month period between August 1, 2016–October 31, 2016 (the year before Hurricane Harvey) and August 1, 2017–October 31, 2017 (when Harvey occurred), there was little difference in overall police calls but an increase in 2017 for CIT calls [Figure 1] and [Figure 2]. EDOs, for which there was an average of one in every 125 police calls, did not change in mean daily calls between the pre-, active, and poststorm periods (26, 27, and 26, respectively) [Table 1].{Figure 1}{Figure 2}{Table 1}

Emergency medical services mental health calls

When comparing the prestorm (May 24, 2017–August 22, 2017), active storm (August 23, 2017–September 15, 2017), and poststorm (September 16, 2017–December 31, 2017) periods based on the availability of EMS data, we did not observe statistically significant differences in the average daily number of mental health calls, which varied between 5.1 and 6.2 calls per day [P = 0.056, [Table 2]].{Table 2}

Although there was a shift in the location of EMS calls from being 46.2% made from home versus another location during the prestorm period to 56.8% from home during the active storm, and then 44.0% from home during the poststorm period (P = 0.04), there were no differences in the distribution of EMS calls by age, gender, race, or dispatch complaint across Harvey-relevant time periods [Table 3].{Table 3}


Disasters can cause stress in the community and exacerbate existing mental health disorders.[3] Previously, it was found that following Hurricane Katrina, those with mental health needs had disruptions in care.[18] Having a greater understanding of mental health crisis service utilization can be critical in preparing for future events. We observed small differences in the average number of police calls per day and CIT identified calls per day when comparing the prestorm, active storm, and poststorm periods. There was not a statistically significant difference in the average number of mental health calls to EMS in comparing the three periods. Initially, calls sent to EMS are triaged, and police are sent if requested or deemed necessary; otherwise, calls default to EMS. Depending on the call, services from EMS, police, or both may be required. An increase in police involvement could happen if the calls were deemed more acute or more likely to require police presence. We did not observe an increase in EDOs when comparing the pre/active/poststorm periods. This suggests that though the calls may have been more acute, they may not have required the completion of EDO. There is limited information available about the role of police related to mental health crisis services after a major disaster making comparisons to other events challenging. Our findings are different from other studies that investigated involuntary detainments following the terrorist attacks in the United States on September 11, 2001. Previous work by Catalano suggests in the immediate period after that event that in acute situations, there may be an increase in involuntary detainments.[15],[16] Differences between the study differences are likely attributable to the changes that can be expected after a terrorist event compared to those following a natural disaster. Another possibility is that a substantial number of people with severe mental health crisis were present at the shelter sites where mental health staff may have been able to identify them for interventions.[19] The initial decrease of police mental health calls during the active storm phase is also consistent with a previous evaluation of EMS total calls, suggesting possible issues with the ability to call during the initial storm period.[20] Previous work looking at EMS calls during Hurricane Sandy showed an increase 2 days after the storm but did not demonstrate the same decrease at the initial duration of the storm.[21] This difference could be related to differences in specific EMS calls versus calls for police services related to a mental health crisis. Specifically, there may be a difference in the reasons for those calls that are reflected in these data. The difference between call locations when comparing pre/active/poststorm may be related to people calling from home during the storm as they are less likely to be out with active flooding, and the call location data return to similar to prestorm levels in the poststorm period.

The fact that there was no statistically significant difference in EMS mental health calls appears to be consistent with some previous work that looked at psychiatric calls following a hurricane. In an evaluation of EMS calls following Hurricane Ike, there was no significant change in psychiatric calls noted.[22] That study was for a shorter duration and was looking at EMS calls in total, though psychiatric calls were identified, while this study extends for several months after the active storm period and that may explain the differences. There was not a statistically significant difference in the dispatch reported incident complaints of EMS mental health calls when comparing the pre/active/storm periods. Although not statistically significant, there appeared to increase in injuries, mental health, neurologic, and substance use calls. In an earlier study looking at Hurricane Harvey and EMS calls, there was a large percentage of calls for injuries observed during the peak call volume days from the Galveston, Texas, Area Ambulance Authority during Hurricane Harvey.[20] Furthermore, Hurricane Ike, a previous natural disaster in the region demonstrated increase in falls and lacerations surrounding the storm.[20],[22] The percentage of EMS calls that included transportation in our study seemed to be higher than a recent study looking at mental health EMS calls from 2018.[23] This difference may be secondary to limited behavioral health alternatives between the two settings, and possible stigma associated with receiving mental services, affecting the proportion of affected people willing to seek care.

Study limitations

There are limitations of this study. This study does not address individuals who presented directly to the emergency room and did not contact the police or EMS. The presence of two EMS databases limits the ability to evaluate calls in the pre-Harvey period before May 1, 2017. The long-term effects were not included as this study is limited to December 31, 2017 as the endpoint, and it is possible that changes in utilization may occur further from the event. This date was chosen as the end of an acute period after storm and a nature breakpoint as the end of that calendar year. A previous study suggested that there may be significant differences in behavioral health problems among frequent utilizers.[24] As the personal information of the service seekers was restricted, information about frequent versus nonfrequent utilizers was not included in this study. It is possible that any visible changes may be related to seasonality and temporality and not related to Hurricane Harvey. This is ultimately an observational study and subsequently, causation cannot be determined about changes in mental health crisis services used in this study. There is no formal diagnostic information to correlate any observed changes with specific psychiatric illnesses. Given the nature of the data, it was also not possible to determine if there were a large number of unique individuals or a small number of individuals with multiple presentations, as the police data do not include individual data for this type of analysis.

Despite these limitations, our study has great strengths. To our knowledge, this is the first study to utilize first-hand police and EMS data to evaluate the impact of Hurricane Harvey on the mental health of the impacted individuals. When the physical health is impacted, as in the case of any natural calamity of such severity as Harvey, little to no attention is given to the mental health of the population. Moreover, the mental health impact of such disasters can be overpowering and long lasting. Therefore, our study can act as the pathway for future studies and policymakers for disaster preparedness and to address and mitigate mental health concerns at the onset, during, and after such disasters hit the humankind. Future studies could also examine the change in mental health-related visits to the ED and hospitalizations during and postcalamity.


In the face of future disasters, it will be crucial to anticipate the mental health-care needs of the community and to recognize the role of first responders including EMS and police to provide initial crisis services. Immediately after Hurricane Harvey, there was an increase in mental health calls to the police; this increase in mental health police calls may be related to an increase in the acuity and subsequent triage of mental health calls following Hurricane Harvey within Houston, and this may provide vital information for planning for further events. Providing mental health services in the large-scale shelters and the fact that there was a large amount of flooding may have affected the mental health call volume during the active storm period. As mental health crisis developed, there was an increase in police mental health calls in the poststorm period with these calls making up a larger percentage of overall police calls. Expansion of crisis mental health services in anticipation of changes immediately following a natural disaster provides an opportunity to be proactive in that aspect of mental health planning.


The authors would like to thank the Houston police and EMS department for making the data available for this study.

Financial support and sponsorship

This research was funded by U.S. Department of Health and Human Services and Health Resources and Services Administration, grant number D34HP31024.

Conflicts of interest

There are no conflicts of interest.

Ethical standards

The authors assert that this scientific work complies with reporting quality, formatting, and reproducibility guidelines set forth by the EQUATOR Network. The authors also assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. Institutional Review Board of Baylor College of Medicine and the protocol number was H-42593.


1The Texas General Land Office Community Development and Revitalization Program. The State of Texas Plan for Disaster Recovery: Hurricane Harvey; 2018. Available from: [Last accessed on 2018 Feb 19].
2Storch EA, Shah A, Salloum A, Valles N, Banu S, Schneider SC, et al. Psychiatric diagnoses and medications for hurricane harvey sheltered evacuees. Community Ment Health J 2019;55:1099-102.
3Kessler RC, Galea S, Jones RT, Parker HA, Hurricane Katrina Community Advisory Group. Mental illness and suicidality after Hurricane Katrina. Bull World Health Organ 2006;84:930-9.
4North CS, Pfefferbaum B. Mental health response to community disasters: A systematic review. JAMA 2013;310:507-18.
5Berggren RE, Curiel TJ. After the storm Health care infrastructure in post-Katrina New Orleans. N Engl J Med 2006;354:1549-52.
6Prener C, Lincoln AK. Emergency medical services and “psych calls”: Examining the work of urban EMS providers. Am J Orthopsychiatry 2015;85:612-9.
7Trivedi TK, Glenn M, Hern G, Schriger DL, Sporer KA. Emergency medical services use among patients receiving involuntary psychiatric holds and the safety of an out-of-hospital screening protocol to “Medically Clear” Psychiatric Emergencies in the Field, 2011 to 2016. Ann Emerg Med 2019;73:42-51.
8Morabito MS, Savage J, Sneider L, Wallace K. Police response to people with mental illnesses in a major U.S. city: The Boston experience with the co-responder model, victims & offenders. Vict Offender 2018;13:1093-105.
9Deane MW, Steadman HJ, Borum R, Veysey BM, Morrissey JP. Emerging partnerships between mental health and law enforcement. Psychiatr Serv 1999;50:99-101.
10Livingston JD. Contact between police and people with mental disorders: A review of rates. Psychiatr Serv 2016;67:850-7.
11Council of State Governments. Criminal Justice/Mental Health Consensus Project; 2002. Available from: [Last accessed on 2019 May 23].
12Watson AC, Ottati VC, Morabito M, Draine J, Kerr AN, Angell B. Outcomes of police contacts with persons with mental illness: The impact of CIT. Adm Policy Ment Health 2010;37:302-17.
13Hedman LC, Petrila J, Fisher WH, Swanson JW, Dingman DA, Burris S. State laws on emergency holds for mental health stabilization. Psychiatr Serv 2016;67:529-35.
14Texas State Legislature. Texas Statute Health and Safety Code 573.001 2017. Available from: [Last accessed on 2019 May 27].
15Catalano RA, Kessell E, Christy A, Monahan J. Involuntary psychiatric examinations for danger to others in Florida after the attacks of September 11, 2001. Psychiatr Serv 2005;56:858-62.
16Catalano RA, Kessell ER, McConnell W, Pirkle E. Psychiatric emergencies after the terrorist attacks of September 11, 2001. Psychiatr Serv 2004;55:163-6.
17Federal Emergency Management Agency. Texas Hurricane Harvey (DR-4332); 2017. Available from: [Last accessed on 2019 Apr 22].
18Wang PS, Gruber MJ, Powers RE, Schoenbaum M, Speier AH, Wells KB, et al. Disruption of existing mental health treatments and failure to initiate new treatment after Hurricane Katrina. Am J Psychiatry 2008;165:34-41.
19Shah AA, Valles N, Banu S, Storch EA, Goodman W. Meeting the mental health needs of hurricane harvey evacuees. Am J Psychiatry 2018;175:13-4.
20Crutchfield AS, Harkey KA. A comparison of call volumes before, during, and after Hurricane Harvey. Am J Emerg Med 2019;37:1904-6.
21Bucher J, McCoy J, Donovan C, Patel S, Ohman-Strickland P, Dewan A. EMS Dispatches during Hurricanes Irene and Sandy in New Jersey. Prehosp Emerg Care 2018;22:15-21.
22Cooper E, Langabeer JR 2nd, Alqusairi D, Persse D. Impact of Hurricane Ike on the call volumes of Houston Fire Department emergency medical services. Am J Disaster Med 2012;7:137-44.
23Roggenkamp R, Andrew E, Nehme Z, Cox S, Smith K. Descriptive analysis of mental health-related presentations to emergency medical services. Prehosp Emerg Care 2018;22:399-405.
24Knowlton A, Weir BW, Hughes BS, Southerland RJ, Schultz CW, Sarpatwari R, et al. Patient demographic and health factors associated with frequent use of emergency medical services in a midsized city. Acad Emerg Med 2013;20:1101-11.