Growing inequities in mental health crisis services offered to indigent patients in Puerto Rico versus the US states before and after Hurricanes Maria and Irma

Abstract

Objective

To assess changes in the availability of mental health crisis services in Puerto Rico relative to US states before and after Hurricanes Maria and Irma.

Data Sources/Study Setting

National Mental Health Services Surveys conducted in 2016 and 2020.

Study design

Repeated cross‐sectional design. The independent variable was mental health facility location in Puerto Rico or a US state. Dependent variables were the availability of three mental health crisis services (psychiatric emergency walk‐in services, suicide prevention services, and crisis intervention team services).

Data Collection/Extraction Methods

The proportion and per 100,000 population rate of facilities offering crisis services were calculated.

Principal Findings

The availability of crisis services at mental health facilities in Puerto Rico remained stable between 2016 and 2020. These services were offered less at indigent care facilities in Puerto Rico than US states (e.g., 38.2% vs. 49.5% for suicide prevention, p = 0.06) and the magnitude of difference increased following Hurricane Maria.

Conclusions

There are disparities between Puerto Rico and US states in the availability of mental health crisis services for indigent patients.

What is known on this topic

• Puerto Rico has experienced a series of emergencies (e.g., Hurricanes Maria and Irma, earthquakes, the COVID‐19 pandemic), which have had major mental health impacts on the island's population.

• Puerto Rico has experienced an extensive financial crisis and forced reformation of its health system by the US federal government in effort to reduce the territory's debt.

• Little prior research has focused on the availability of mental health crisis services or mental health services more broadly on the island of Puerto Rico.

What this study adds

• Availability of psychiatric emergency walk‐in, suicide prevention, and crisis intervention team services was higher or equivalent in Puerto Rico than in the US states before and after Hurricanes Maria and Irma.

• The availability of crisis services at indigent care facilities was lower in Puerto Rico than in the US states during this period.

• The magnitude of this inequity in crisis service offerings at indigent care facilities increased after Hurricanes Maria and Irma.

1. INTRODUCTION

Puerto Rico is a US territory that has experienced a series of major disasters and public health emergencies in recent years. ¹ , ² , ³ Hurricanes Irma and Maria were two Category 5 storms that struck the island in September 2017 and caused substantial death, infrastructure devastation, and health care system disruption. ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² In 2019, a cluster of earthquakes also caused major damage and distress across the island. Shortly after in 2020, the COVID‐19 pandemic caused significant mortality, morbidity, and psychological distress. In the years prior to these emergencies, the territory was in the midst of an economic crisis and its health care system was undergoing forced reformation by the US federal government to reduce its debt. ¹³ , ¹⁴ Economic hardship among the island's residents is pervasive; the 2020 poverty rate was about four times higher in Puerto than the US states (43.4% vs. 11.4%). ¹⁵ A study also observed that health insurance coverage decreased among the island's residents following Hurricanes Irma and Maria. ¹⁶

Disasters, public health emergencies, and economic stressors are well‐established risk factors for community‐level mental health problems, psychiatric emergencies, and suicidality. ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ Numerous studies have documented elevated rates of these issues among Puerto Rico's residents following Hurricanes Maria and Irma. ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² For example, the suicide rate on the island is estimated to have doubled 3 months following the storms, ¹⁰ and the prevalence of post‐traumatic stress disorder among island residents 6 months after the storms was estimated to be 44%. ¹¹ Over 20% of the island's residents reported that they or a household member received mental health services because of the storms, and an additional 17% believed that they needed these services but did not receive them. ¹² News media reports have suggested that mental health outcomes in Puerto Rico have further worsened in light of the earthquakes and COVID‐19 pandemic. ²¹ , ²² , ²³

Within this context of elevated need for mental health crisis services, it is important to characterize the mental health crisis services environment in Puerto Rico, understand how it might have changed after Hurricanes Maria and Irma, and describe how it compares to that of the US states. ²⁴ More specifically, given the substantially higher poverty rate in Puerto Rico than the US states ¹⁵ and that insurance providers, such as Medicaid, often do not cover mental health crisis services, ²⁵ , ²⁶ , ²⁷ , ²⁸ there is a need to assess the availability of mental health crisis services for low‐income patients who are unable to pay for care. Furthermore, understanding mental health crisis service environments is of elevated importance as the roll out of 9‐8‐8—the new three digit dialing code for the National Suicide Prevention Lifeline, which went into effect in July 2022—is projected to substantially increase demand for crisis services across the US states and its territories. ²⁵ , ²⁹ , ³⁰

Very little mental health services research has focused on Puerto Rico and the few studies focused on the topic were conducted more than 20 years ago. ³¹ , ³² , ³³ , ³⁴ , ³⁵ Qualitative studies have documented health care system disruptions on the island following Hurricanes Maria and Irma, ³⁶ , ³⁷ but little of this work has focused on mental health services. ³⁸ Quantitative studies have characterized trends in the offering of mental health services, ³⁹ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ including crisis services, ³⁹ , ⁴⁵ in the United States, but these studies have excluded Puerto Rico and have not generated Puerto Rico‐specific estimates.

This study will address these knowledge gaps through analyses of data from the 2016 (pre‐Hurricane Maria/Irma) and 2020 (post‐Hurricanes Maria/Irma) National Mental Health Services Surveys (N‐MHSS). The primary study objectives were to assess changes in the offering of crisis services at mental health facilities in Puerto Rico before and after Hurricane Maria/Irma and compare the magnitude of these changes to those of the US states. The secondary objective is to assess the differences among facilities that offer services to indigent clients.

2. METHODS

2.1. Data and study design

Data on services offered at mental health facilities came from the N‐MHSS, a web‐based, post‐mail, and telephone‐based census conducted annually by the Substance Abuse and Mental Health Services Administration. The questionnaire is sent to the census of all known mental health facilities in the US mainland and its territories. N‐MHSS datasets are publicly available and widely used in health services research. ³⁹ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ We analyzed 2016 (pre‐Hurricanes Maria/Irma) and 2020 (post‐Hurricanes Maria/Irma) N‐MHSS data. The 2016 survey aggregate response rate was 91% (n = 12,164) and the Puerto Rico response rate was 99% (n = 88). The 2020 survey aggregate response rate was 89% (n = 12,266), and the Puerto Rico response rate was 79% (n = 69). Although most facilities completed the N‐MHSS for both years, a repeated cross‐sectional design was used because the N‐MHSS does not contain a unique identifier for facilities across survey years.

2.2. Variables

The primary independent variable was whether each mental health facility was located in Puerto Rico or a US state. Facilities in other US territories were excluded from the analyses. The secondary independent variable was survey year (2016, pre‐Hurricanes Maria/Irma or 2020, post‐Hurricanes Maria/Irma). The primary dependent variables were separate dichotomous indicators of whether each facility offered three crisis services: (1) psychiatric emergency walk‐in services, (2) suicide prevention services, and (3) a crisis intervention team that handles acute mental health issues at this facility and/or off‐site (i.e., crisis intervention team services). The secondary dependent variable was a dichotomous indicator of whether each facility offered “treatment at no charge to clients who cannot afford to pay” and/or used a “sliding‐scale fee” based on “income and other factors” and offered crisis services. These facilities are hereafter referred to as “indigent care” facilities.

2.3. Analysis

The number and proportion of facilities providing each crisis service in Puerto Rico and all US states combined, respectively, were calculated and stratified by survey year. Chi‐squared tests assessed the significance of differences in the proportion of facilities offering each crisis service between Puerto Rico and the US states within each survey year and the significance of changes in the proportion of facilities offering each service between survey years within each geography. P‐values indicating the significance of differences between Puerto Rico and the US states and differences within Puerto Rico between years should be interpreted with caution because the total number of mental health facilities in Puerto Rico is relatively small (e.g., n = 69 in 2020, with the sample capturing 79% of all known mental health facilities on the island). Strict adherence to a p‐value threshold of 0.05 may result in a Type 2 error when assessing these differences because of limited statistical power as a result of the small and finite number of mental health facilities in Puerto Rico.

Consistent with prior work assessing the availability of crisis services in the US mainland, ³⁹ we estimated per 100,000 population rates of the number of mental health facilities offering each crisis service in Puerto Rico and the US states, respectively. Population size data were obtained from the 2016 American Community Survey, 2016 Puerto Rico Community Survey, and 2020 Decennial Census. For aggregate facility rates, total population size estimates were used. For indigent care facility rates per 100,000 population, the size of the population living below the federal poverty limit (FPL) was used. We used the size of the population below the FPL because it served as a more precise approximation of the size of the population that may not be able to pay for crisis services. When calculating rates, estimates of the number of mental health facilities offering each service were generated by multiplying the total number of facilities in Puerto Rico and the US states, respectively, by the percentage of facilities that indicated offering the service in the N‐MHSS surveys. Although this approach is based on the potentially invalid assumption that the proportion of facilities offering each service is identical among facilities that did and did not complete the N‐MHSS survey, it accounts for differences in N‐MHSS response rates between Puerto Rico and the US states each survey year. Rate ratios were calculated to compare rates between Puerto Rico and the US states each survey year.

3. RESULTS

Pre‐Hurricanes Maria/Irma in 2016, larger proportions of facilities in Puerto Rico than the US states offered psychiatric emergency walk‐in services (38.6% vs. 30.5%, p = 0.10), suicide prevention services, (75.0% vs. 53.9%, p < 0.001), and crisis intervention team services (68.2% vs. 47.8%, p < 0.001) (Table 1). After Hurricanes Maria/Irma in 2020, the portion of facilities offering each of these services remained higher or equivalent in Puerto Rico than the US states. Between 2016 and 2020, the proportion of facilities offering psychiatric emergency walk‐in services in Puerto Rico decreased by 6.7 percentage points, becoming equivalent with that in the US states (31.9% vs. 32.2%, p = 0.96). The proportion of facilities in Puerto Rico offering suicide prevention services increased by 0.4 percentage points from 2016 to 2020, while it increased by 12.1 percentage points in the US states but remained larger in Puerto Rico (75.4% vs. 66.0%, p = 0.01).

TABLE 1.

Proportions of mental health facilities offering crisis services: Puerto Rico and US States, 2016 and 2020

Crisis service	2016			2020			2016–2020 Change
	Number of facilities offering service	Percentage of facilities offering service	Within year, between geography χ2 p value	Number of facilities offering service	Percentage of facilities offering service	Within year, between geography χ2 p value	Between year, within geography percentage point difference	Between year, within geography χ2 p value
Psychiatric emergency walk‐in
US states	3685	30.5%	0.1	3908	32.2%	0.96	+1.70	0.01
Puerto Rico	34	38.6%		22	31.9%		−6.7	0.38
Suicide prevention
US states	6503	53.9%	<0.001	8026	66.0%	0.10	+12.1	<0.001
Puerto Rico	66	75.0%		52	75.4%		+0.4	0.96
Crisis intervention team
US states	5765	47.8%	<0.001	5789	47.7%	<0.001	−0.1%	0.94
Puerto Rico	60	68.2%		51	73.9%		+5.7	0.43

When focusing on indigent care mental health facilities, the proportions of those offering crisis services were similar between Puerto Rico and the US states in 2016 but then generally decreased in Puerto Rico in 2020 following Hurricanes Maria/Irma while they increased or remained constant in the US states (Table 2). Findings were similar when no charge and sliding‐scale facilities were disaggregated (Appendixes A and B, respectively). For example, in 2016, 25.6% of mental health facilities in Puerto Rico were indigent care facilities that offered psychiatric emergency walk‐in services, while the proportion was 24.7% in the US states. However, in 2020, this proportion had decreased by 9.4 percentage points to 16.2% % in Puerto Rico, while it increased by 0.8 percentage points to 21.0% in the US states (p = 0.08). In 2016, 44.2% of mental health facilities in Puerto Rico were indigent care facilities that offered suicide prevention services compared with 40.5% of facilities in the US states. However, in 2020, this proportion had decreased by 6.0 percentage points to 38.2% in Puerto Rico, while it increased by 9.0 percentage points to 49.5% in the US states (p = 0.06).

TABLE 2.

Proportions of mental health facilities offering crisis services, limited to “Indigent care” facilities that offer and services to clients who cannot afford to pay: Puerto Rico and US States, 2016 and 2020

Crisis service	2016			2020			2016–2020 Change
	Number of indigent care facilities offering service	Percentage of indigent care facilities offering service	Within year, between geography χ2 p value	Number of indigent care facilities offering service	Percentage of indigent care facilities offering service	Within year, between geography χ2 p value	Between year, within geography percentage point difference	Between year, within geography χ2 p value
Psychiatric emergency walk‐in
US states	2868	24.7%	0.84	2938	25.5%	0.08	+0.8%	0.15
Puerto Rico	22	25.6%		11	16.2%		−9.4%	0.16
Suicide prevention
US states	4712	40.5%	0.49	5711	49.5%	0.06	+9.0%	<0.001
Puerto Rico	38	44.2%		26	38.2%		−6.0%	0.46
Crisis intervention team
US states	4358	37.5%	0.62	4349	37.8%	0.68	+0.3%	0.64
Puerto Rico	30	34.9%		24	35.3%		+0.4%	0.96

When expressed as rates per 100,000 population, there were fewer mental health facilities offering crisis services in Puerto Rico than the US states, especially indigent care facilities offering these services (Table 3). When all facilities were considered in aggregate, per 100,000 population rates were slightly lower in Puerto Rico than the US states for psychiatric emergency walk‐in services and suicide prevention services. When focusing on indigent care mental health facilities, however, larger inequities emerged.

TABLE 3.

Rates of mental health facilities offering crisis services: Puerto Rico and US States, 2016 and 2020

Crisis service	All mental health facilities				“Indigent care” mental health facilities offering services to clients who cannot afford to pay
	2016 rate per 100,000 population	2016 within year, between geography rate ratio a	2020 rate per 100,000 population	2020 within year, between geography rate ratio a	2016 rate per 100,000 population in poverty	2016 within year, between geography rate ratio a	2020 rate per 100,000 population in poverty	2020 within year, between geography rate ratio a
Psychiatric emergency walk‐in
US states	1.14	1.13	1.20	1.24	7.27	4.72	8.34	7.32
Puerto Rico	1.01		0.97		1.54		1.14
Suicide prevention
US states	2.02	1.03	2.46	1.07	11.92	4.50	16.19	6.04
Puerto Rico	1.96		2.29		2.65		2.68
Crisis intervention team
US states	1.79	1.01	1.78	0.79	11.04	5.26	12.36	5.00
Puerto Rico	1.78		2.25		2.10		2.47

^a Reference group = Puerto Rico.

In 2016, there were 1.54 indigent care mental health facilities offering psychiatric emergency walk‐in services per 100,000 Puerto Rico residents living in poverty compared with 7.27 facilities per 100,000 US state residents living in poverty (rate ratio = 4.72). After Hurricanes Maria/Irma in 2020, the rate decreased to 1.14 per 100,000 in Puerto Rico and increased to 8.34 per 100,000 in the US states (rate ratio = 7.32). A similar inequity existed for indigent care mental health facilities offering suicide prevention and crisis intervention team services. In 2020, the number of indigent care facilities offering suicide prevention services per 100,000 residents living in poverty was more than six times higher in the US states than Puerto Rico (rate ratio = 6.04) and more than five times higher for crisis intervention team services (rate ratio = 5.00).

4. DISCUSSION

To our knowledge, this study is the first quantitative assessment of the availability of mental health crisis services in Puerto Rico. Despite the impact of Hurricanes Maria and Irma and other recent major public health emergencies on Puerto Rico's health care system, ³⁶ , ³⁷ , ³⁸ we did not find that the availability of crisis services at mental health facilities decreased meaningfully on the island between 2016 and 2020 in absolute terms or relative to the US states. However, when limiting our analyses to indigent care facilities, we found less mental health crisis service availability in Puerto Rico than the US states, and that the size of this difference increased after Hurricanes Maria and Irma. This finding is concerning given the substantially higher poverty rate in Puerto Rico than the US states. ¹⁵

Differences between Puerto Rico and the US states were largest for psychiatric emergency walk‐in services, a service for patients with immediate service need and severe psychiatric distress. For a person experiencing a psychiatric emergency who cannot afford to pay for services in Puerto Rico in 2020, there was less than one facility per 100,000 persons living in poverty, while there were nearly seven such facilities per 100,000 persons in poverty in the US states. Although emergency psychiatric care can be obtained regardless of the patient's ability to pay through general hospital emergency departments under the Emergency Medical Treatment and Active Labor Act, the care is likely of higher quality at mental health facilities. Unlike most emergency departments, these facilities are staffed with trained mental health professionals, and there are well‐established barriers to providing high‐quality care for psychiatric emergencies in general hospital emergency departments. ⁴⁶ , ⁴⁷ , ⁴⁸

The finding that indigent care mental health facilities offering crisis services were less available in Puerto Rico compared with the US states may reflect efforts to reform and privatize Puerto Rico's health care system. ¹³ , ¹⁴ These efforts, which stem from the 2016 Puerto Rico Oversight, Management, and Economic Stability Act, which created a federal Financial Oversight and Management Board to address the territory's debt crisis, may have had the consequence of reducing access to public mental health crisis services for Puerto Rican residents with low incomes during a time of elevated service need.

4.1. Limitations

N‐MHSS is the primary data source for tracking services offered at mental health facilities in the US mainland and its territories, but it has limitations. Facilities indicate whether they provided specific mental health services but not the volume of or capacity for specific services. ⁴⁹ For example, a facility that indicated offering crisis intervention team services could have multiple full‐time teams working 24/7 or one part‐time team. Such variation is not captured in the N‐MHSS data. The data also do not provide indication of service quality, and some reports suggest that the quality of psychiatric care provided to indigent patients in Puerto Rico is inferior to that provided to privately insured patients. ^50,51 The N‐MHSS uses a self‐report questionnaire and does not validate whether responses correspond with actual service offerings. Our study focused on comparing Puerto Rico to all US states combined. There is substantial geographic heterogeneity across US states and across the island of Puerto Rico in the availability of mental health crisis services, ³⁹ and the study did not compare Puerto Rico to individual US states or assess geographic differences across Puerto Rico. Finally, as noted above, p‐values should be interpreted with caution because the study may have been underpowered to detect significant differences at a threshold of 0.05, given that the total number of mental health facilities in Puerto Rico is relatively small.

5. CONCLUSIONS

The availability of crisis services at mental health facilities in Puerto Rico remained fairly stable between 2016 and 2020, the period before and after Hurricanes Maria and Irma. However, inequities between Puerto Rico and the US states exist in terms of the availability of mental health crisis services for low‐income patients. The size of these differences in services availability generally increased following Hurricanes Maria and Irma. Efforts to reform Puerto Rico's health care system should ensure that mental health crisis services are available for island residents living in poverty, which make up nearly half of Puerto Rico's population.

FUNDING INFORMATION

This study was supported by grants R01MD016426 and R01MD013866 from the National Institute on Minority Health and Health Disparities (NIMHD) at the National Institutes of Health (NIH).

Supporting information

Appendix S1. Supporting Information.

Purtle J, Rivera‐González AC, Mercado DL, et al. Growing inequities in mental health crisis services offered to indigent patients in Puerto Rico versus the US states before and after Hurricanes Maria and Irma. Health Serv Res. 2023;58(2):325‐331. doi: 10.1111/1475-6773.14092