News & Events

Use of Mental Health Services Among Disaster Survivors

Posted to the Web: Wednesday, July 02 , 2008
Jorge J. Rodriguez; Robert Kohn Author Information 07/02/08

Abstract and Introduction

Abstract

Purpose of Review: A sizable proportion of individuals following a disaster develop mental health problems. The consequences of these disorders can be long lasting. Only recently has research focused on mental health service delivery following disasters. This review examines the rates, predictors, and barriers to mental health service utilization following a disaster.
Recent Findings: Most of the data on mental health service delivery come from three sources: a fireworks disaster in The Netherlands, the September 11, 2001, attack on New York City, and hurricane Katrina. Most survivors of disasters are reluctant to utilize mental health services and face barriers to accessing care. Even among disaster victims who are severely mentally ill, only a minority receive treatment. Among those who do receive assistance, more than half drop out shortly thereafter. Mental health service utilization following a disaster is influenced by a set of predisposing characteristics, enabling resources, and perceived need. The model for mental healthcare delivery following a disaster that has gained acceptance is Psychological First Aid.
Summary: Research is evolving on mental health service utilization. It is limited however to developed countries, although most disasters occur in developing countries. More research is needed, particularly among populations with scarce resources.

Introduction

Although most individuals following a disaster do not develop psychiatric problems, a sizable proportion do.[1] Among those who develop emotional and psychiatric disorders, the consequences can be lasting.[2–4] Disasters that should have minimal consequence for mental health on the population level other than transient stress reactions are those where injury and death are rare, property destruction does not outstrip community resources, when social support systems remain intact, and the event does not take on more symbolic meaning. Although rare, some disasters do result in severe, lasting and pervasive psychological consequences. Such disasters are characterized by extreme and widespread property damage, resulting in ongoing financial problems for the community, are frequently caused by human intent rather than by a natural event, and there is a high prevalence of injuries, threat to life and loss of life. Delivery of mental health services in such situations is a challenge.[5,6] Numerous studies have been conducted on disasters, investigating their mental health consequences. Few studies have actually examined postdisaster mental health services (MHS) utilization.

Rates of Mental Health Service Utilization After a Disaster

Most studies that provide information on MHS use following disasters have been limited to examining the rates of service utilization in victims, but not the antecedents of service utilization. Service utilization studies can be divided into those that examine MHS use among the direct victims or those that examine MHS use in the affected community. Most disaster studies have focused on MHS utilization at the level of the individual survivor. The studies conducted in The Netherlands following a fireworks disaster[7•] and hurricane Mitch[1] are examples of survivor studies. The series of studies conducted regarding the September 11, 2001, World Trade Center attack and hurricane Katrina[8••] are examples of community-based MHS utilization studies.

A 4-year longitudinal study of 649 survivors of a fireworks disaster that destroyed 500 houses, injured approximately 1000 people, and killed 23 in The Netherlands found that large groups of people with mental health problems do not receive treatment. At 18 months following the disaster of those with posttraumatic stress disorder (PTSD) or depressive symptoms, only 43.6% received treatment. Of those, 73% had contact with specialized MHS, and 53.9% were treated psychopharmacologically.[7•] At 4 years, 40.8% of those who were symptomatic were receiving MHS. Of those individuals, 71.3% were using psychiatric medications and 60% specialized MHS.

A community household survey of 800 adults directly exposed to hurricane Mitch in Honduras where MHS are limited found that 5.3% sought out MHS. Of those who consulted anyone for their ‘nerves’ (8.8% of the respondents), 10.2% saw a psychiatrist, 19.8% a psychologist, 3.0% a nurse, and 8.2% clergy. Most saw their primary care physician (49.1%). The remainder sought help from friends and family.[9]

In a random digit-dialing survey of MHS utilization of 1043 pre-hurricane Katrina residents of Alabama, Mississippi and Louisiana following the disaster, 16% had sought MHS:[8••] 4% sought a mental health specialist and 11% a general medical provider. Among those with mild to moderate anxiety and depression, 23% sought assistance, and among those with a serious mental illness, 46% sought help. Only 18% of those with a new-onset disorder sought treatment. Psychopharmacological treatments were used by 12% of the sample, 19% with mild to moderate mental illness, and 37% with severe mental illness. The most commonly prescribed psychopharmacological agent was antidepressants, followed by benzodiazepines. Psychotherapy was utilized by 7%, 12%, and 24%, respectively. Of those who did receive psychotherapy following the disaster, 60% had dropped out of treatment. Fourteen percent of the respondents who did not seek treatment reported that they were in need of treatment. Among respondents with preexisting mental disorders who reported using MHS in the year before the hurricane, over one-fifth experienced reduction in or termination of treatment after hurricane Katrina.[10••]

A number of studies from the September 11, 2001, terrorist attack have specifically examined MHS. Some studies on MHS utilization from the September 11, 2001, terrorist attacks used a population-based continuous time-sampling procedure, in order to obtain a representative sample of not only the population, but also the time period under study. This method overcomes the limitations of a single time-point sample used in other studies, and allows generalization across a defined time period. A study of residents living near New York City in Connecticut with 4640 respondents 5–15 months following the attack found that 9% of the population sought help for problems associated with reaction to the events of September 11, 2001. Of those individuals, 20% sought help from peers, 13.5% from a priest, doctor, therapist, support group, or other health professional, and 8% received therapeutic services.[11] The rate of service utilization was higher than in a study conducted a year earlier, 1–3 months following the attack, on the same population (6%).[12]

One to two months following the World Trade Center disaster, a random digit-dialing study of 1008 persons living in Manhattan found that 19.4% with current PTSD or current depression had a professional visit for a mental health problem within 30 days after the attack.[13] An increase in mental health visits was found among 10%. Psychiatric medications usage increased after the attack; of the population who were not taking psychotropic medications, 3.3% were after September 11, 2001. In another similar study conducted 4–5 months after the attack throughout New York City among 20 001 individuals, 7.6% of the respondents reported use of psychiatric services, and 7.7% reported use of psychiatric medications in the past 30 days.[14] This later study suggested that there might have been a decrease in population-level utilization in the months following the disaster.

A study conducted 6 months after September 11 among 2752 residents of the Greater New York Metropolitan Area using random digit dialing found that 9% of the population, 15% of those directly affected, and 36% of those with probable PTSD or depression sought help from a professional after the attacks.[15] In another random digit-dialing interview study conducted over 1 year following September 11, 2001, of the 2368 respondents in New York City, 20% of the population had a mental health visit in the past year, and 12.9% of the visits were related to the disaster itself. Psychotropic medications were used by 4.5% of the respondents for issues related to the attack.[16] Among those with major depression or PTSD, 45% reported receiving MHS postdisaster, and 33.1% reported receiving services following the World Trade Center attack for symptoms due to the event.[17] As for psychotropic drug use, 26% were prescribed medications in the year following September 11, 2001, and 16% were prescribed for emotional problems associated with the attacks.

A face-to-face interview of 10 009 residents of Manhattan 3–5 months following the September 11 disaster found that 11.3% reported that they were currently receiving psychotherapy or other trauma-related or grief-related counseling that began after September 11.[18] A total of 1.4% received psychotropic medications. Of those who were most symptomatic, only 26.7% reported receiving any kind of treatment.

Few studies have examined correlates of MHS utilization. There are several factors that have been hypothesized to predict MHS utilization based on Andersen's Behavioral Model of Healthcare utilization:[19] first, predisposing characteristics, personal or sociocultural factors; second, enabling resources such as health insurance and the availability of services; and third, perceived need, the individual's motivation to get help due to emotional distress or associated behavioral or functional problems.

Predisposing Characteristics

At 18 months after The Netherlands fireworks disaster,[7•] the investigators found that those with higher education and with medium and high levels of optimism were less likely to use MHS. Immigrants and those relocated following the disaster were more likely to use MHS. Gender, age, and marital status were not significant predictors of MHS utilization at 18 months. When controlled in multiple regression analysis, however, relocation, education, optimism, and migrant status were not significant. At 4 years, this study found that female gender and immigrants were the only predisposing factors associated with MHS utilization. In addition, being single was noted to be associated with higher service utilization. Those who had only a medium exposure to the disaster were less likely to seek treatment.

In Honduras following hurricane Mitch, those with lower socio-economic status and females were more likely to seek help for their ‘nerves’ following the natural disaster.[9] A postal questionnaire of 231 evacuees conducted 10 months after the Miyake Island volcanic eruption in Japan found that younger and female victims sought help more from informal sources, while older and male victims sought help more frequently from formal mental health sources.[20]

The Connecticut September 11, 2001, study conducted 5–15 months after the attack found in the bivariate analysis that receiving formal help or therapeutic services was associated with not being married, being employed, current or increased smoking, and not currently drinking alcohol. None of these predisposing factors, however, remained significant in a multivariate analysis.[11] Age, ethnicity, and race were not associated risk factors. An earlier study on the same population conducted 1–3 months after the tragedy found that those who smoke were more likely to seek help; however, this did not remain significant in the multivariate analysis.[12]

In a study conducted 1–2 months after the attack using random digit dialing in Manhattan, no predisposing characteristics were found for MHS utilization. Predictors for receiving psychiatric medications were noted, including being female and being exposed to more than one lifetime trauma.[13,21] Predictors of increased MHS use 4–5 months after the attack in a similar study conducted throughout New York City were graduate school education, being white rather than African American or Hispanic, increased postdisaster alcohol use, involvement with recovery efforts, exposure to earlier lifetime traumatic events, and residing closer to the disaster site.[14] After adjustment in a multivariate model, only race and life events remained significant. A similar finding was noted for predisposing characteristics for being prescribed psychiatric medications. African Americans and Hispanic people and those who were younger were less likely to receive psychiatric medications, while those who were once married, had more stressful life events, and increased use of alcohol were more likely to receive medications.

A study conducted 6 months after the attack in Greater New York City by random digit dialing found that, among those with PTSD or depression, having a higher income was associated with help seeking.[15] The random digit-dialing study conducted 1 year after September 11, 2001, found no statistically significant predisposing factors for MHS utilization for those with major depression or PTSD, except for the degree of exposure.[17] Interestingly, when MHS utilization was examined for only September 11, 2001-related issues, African Americans were less likely to utilize MHS than white people. African Americans were also less likely to use medications postdisaster. Those aged 45–65 were the most likely to be treated with psychotropic medications. Exposure was not related to being treated with psychotropic medications 1 year after the disaster.

MHS utilization among displaced and nondisplaced individuals from hurricane Katrina had few predictors. Those who were married and middle-aged had higher rates of service utilization.

Enabling Resources

Although in The Netherlands having private compared with public health insurance does not lead to greater access to MHS, the fireworks study found that having private health insurance was associated with increased service utilization among those in need of assistance at 18 months.[7•] In addition, those who had received treatment at 18 months were more likely to receive treatment at 4 years.

Interestingly, in the two studies of the September 11, 2001, terrorist attacks conducted on residents of Connecticut, health insurance was not a predictive factor for MHS utilization.[11,12] A similar finding was noted in the random digit-dialing study in New York City 1 year after the attack for those with major depression and PTSD.[17]

Having a physician in the 1 year post-September 11th study, however, was predictive of seeking help in the bivariate, but not multivariate, analysis of September 11, 2001-related issues. Having a physician was predictive of being prescribed psychotropic medications. In an earlier study, using random digit dialing 1–2 months after September 11, 2001, in Manhattan, insurance status was associated with being prescribed medications.[13] In the random digit-dialing study conducted 4–5 months after the attack in a sample from New York City, having a primary care physician was predictive of seeking MHS in the bivariate, but not multivariate, analysis.[14]

Perceived Need

Enabling factors for increased MHS utilization following The Netherlands fireworks disaster included having comorbid PTSD with anxiety and depressive symptoms.[7•] At 18 months after the event, adults with predisaster psychological problems and those with three or more mental health problems were more likely to use MHS. Also at 4 years, adults with three or more mental health problems were more likely to use MHS.

In one of the few studies conducted in a developing country, in a study of 191 persons visiting primary care clinics 6 months after hurricane Mitch in a rural area in Nicaragua, the results of a logistic regression for seeking help from a mental health professional found that only having a previous mental health problem was a predictor.[22]

In the Miyake Island volcanic eruption study, severity of PTSD and depression symptoms were positively correlated with help-seeking from physicians, but not psychologists or mental health professionals.[20]

The Connecticut September 11, 2001, 5-month to 15-month study found only those reporting at least 1 day of poor mental health during the past 30 days was associated with receiving peer support, formal help, or therapeutic services in a multivariate analysis,[11] and there were no predisposing or enabling factors. Those with disability were found to have an association with utilizing MHS in bivariate, but not multivariate, analysis. The investigators also found a correlation between the intensity of help received and the likelihood of chronic mental problems. Interestingly, in the earlier 9/11 study, at least 1 day of poor mental health during the past 30 days was not associated with receiving help, after adjusting for covariates in a multivariate analysis. Sleep problems, increased substance use, and receiving alternative help were all associated with formal help-seeking.[12] In the first Connecticut study, receiving MHS was not clearly differentiated from seeking help for physical health issues.

The random digit-dialing study conducted 6 months after September 11, 2001, found that, among those with PTSD or depression, having a previous mental health problem prior to the terrorist attack was associated with seeking MHS.[15] Among those directly affected but without PTSD, having a prior mental health problem and having a regular doctor, and being in poorer physical health were associated with receipt of counseling in the 6 months following the attack. Those who received counseling prior to September 11, 2001, were more likely to also receive it after September 11, 2001. The other random digit-dialing study of those with PTSD or major depression 1 year after the attack found those with perievent panic attacks were more likely to have MHS utilization for World Trade Center disaster-related symptoms, as well as psychotropic medications.[17] Similarly, those with a perievent panic attack were more likely to receive psychiatric medications 1–2 months after September 11, 2001, as were those with PTSD.[13] PTSD and depression were also found to be predictive of MHS utilization 4–5 months after the World Trade Center attack. Perievent panic attacks, although significant for MHS utilization in bivariate analysis, were not significant in the multivariate model.

An internet-based survey of 704 presumed victims of September 11, 2001, found that those who received psychotropic prescription drugs after the terrorist attack and those who received grief counseling were more likely to have complicated grief.[23]

Barriers to Care

A number of studies have inquired about barriers to care. Stuber et al.[15] found that among individuals with PTSD or depression who did not consider seeking help following September 11, 2001, 17% indicated they lacked knowledge about how to get help, 27% were concerned about stigma associated with mental illness, 29% indicated they did not have the money, 42% said they did not have the time, and 58% stated they did not seek help because others needed the services more than themselves. Of those who considered but did not seek care, the most common reason given was they believed they could care for themselves; a few reported being too depressed to seek care, not trusting of mental health professionals, and a fear of talking about the September 11 attacks.

One year after the attack, 73% of individuals with PTSD and depression who did not seek treatment reported that they did not believe they had a problem.[17] Other reasons given for not seeking treatment included wanting to solve the problem on their own (5%), having problems accessing services (6%), having financial problems (4%), and having a fear of treatment (4%). For those who delayed for 2 weeks or more before seeking treatment, most reported having experienced an access problem (24%). The next most common reasons given were that they wanted to solve the problem on their own (14%), they did not think they had a problem (11%), and that they were afraid of treatment (10%).

An inquiry into barriers to care was also conducted in the hurricane Katrina study among those who did not seek MHS and were in need of services.[8••] Enabling factors were the most common reasons cited: financial (37%), inconvenience (18%), availability (13%), and transportation (8%). Need factors were the next most common: thought they would get better on their own (18%), and problem was not severe (5%). Predisposing factors included wanted to handle it themselves (17%), stigma (5%), and perceived ineffectiveness of treatment (2%). Table 1 summarizes the barriers to MHS utilization following a disaster.

Mental Health Service Implementation

The current literature suggests that most survivors of disasters are reluctant to utilize MHS, or face significant barriers to access MHS. This finding has been noted in the extensive studies conducted following the September 11, 2001, terrorist event and hurricane Katrina. The reluctance to seek mental healthcare appears to be present in small-magnitude disasters, such as the Beverly Hills Supper Club fire,[24] as well as ones that affect a sizable proportion of the population. Even among disaster victims who are severely mentally ill, only a minority receive treatment. Among those who do receive MHS, more than half drop out after a short period of time.

MHS utilization following a disaster is influenced by a set of predisposing characteristics, enabling resources, and perceived need. Awareness of these predictors for MHS utilization is useful in preparation of MHS planning for a disaster (see Table 2 ).

Longitudinal studies have suggested that individuals with high exposure to a disaster may continue to have poorer psychological wellbeing, suggesting that MHS for survivors of a community disaster should continue beyond the first year.[25] Individuals in less affected communities may be equally in need, however, as noted in a study from China where people in communities less affected by a major earthquake were more likely to develop PTSD and exhibited poorer recovery 3–6 months later than individuals in more severely affected communities who received sustained help, including MHS.[26] Even when MHS are readily available, awareness of their existence may be low.[27]

Following the experiences of September 11 and hurricane Katrina, it has been suggested that effective mental health planning requires assessment of population-level MHS needs.[28] Emergency mental health units stationed in devastated areas were shown to be useful following hurricane Katrina.[10••] Other novel strategies may be necessary to reach widely dispersed populations affected by a disaster and evacuees. For example, personnel outside affected areas could be utilized to remotely deliver services such as cognitive-behavioral therapy by telephone, internet, or telecommunications.[29] The development of geographical maps to identify at-risk populations that reside near potential disaster sites for MHS planning has been a recent proposal.[30]

Historically, a psychological debriefing procedure—Critical Incident Stress Debriefing—was commonly utilized; however, the benefits of this approach have been brought into question and its use is no longer recommended. Currently, crisis intervention specialists recommend an approach known as Psychological First Aid (PFA),[31] which is not a plan of delivery of MHS, but a set of principles of care. PFA recognizes that MHS delivery is complex and requires outreach and communication, and addresses barriers to care through a public health policy approach rather than only individual-focused interventions. The principles of this approach focus on meeting each individual's crisis-related psycho-bio-social needs in a practical manner, while developing an action plan for recovery. The basic objectives of PFA are to, first, establish a human connection in a nonintrusive, compassionate manner; second, enhance immediate and ongoing safety, and provide physical and emotional comfort; third, calm and orient emotionally overwhelmed or distraught survivors; fourth, help survivors to say specifically what their immediate needs and concerns are, and gather additional information as appropriate; fifth, offer practical assistance and information to help survivors address their immediate needs and concerns; sixth, connect survivors as soon as possible to social support networks, including family members, friends, neighbors, and community helping resources; seventh, support adaptive coping, acknowledge coping efforts and strengths, and empower survivors; encourage adults, children, and families to take an active role in their recovery; eighth, provide information that may help survivors cope effectively with the psychological impact of disasters; and ninth, be clear about your availability and link the survivor to another member of a disaster response team or to local recovery systems, MHS, public-sector services, and organizations.[32] Table 3 provides a summary of core actions involved in PFA and their goals. PFA includes conducting outreach and survivor engagement to make the community aware of available services, and engage them if appropriate in using these services. It also provides psychoeducation and postdisaster education, allowing survivors to understand posttrauma responses and view them as acceptable, as well as help them understand where to seek further treatment, how to better use social supports, and how to develop adaptive coping strategies. PFA directs professionals to talk to survivors about sources of social support. This model supports that use of brief crisis therapy, which focuses on offering survivors emotional support, education about reactions to disasters, coping support, problem solving, and direction to available resources.

Although nearly all MHS disaster research has been conducted in developed countries, most disasters occur in the developing world where MHS resources are scarce or nonexistent. In 2005, the Inter-Agency Standing Committee (IASC) established minimum response guidelines on mental health and psychosocial support in emergencies.

These guidelines mostly focus on protection and social support, but also include attention to preexisting or emergency-induced severe mental disorders, acute trauma-induced distress and substance use.[33] The guidelines suggest mobilizing groups of disaster-affected people to organize their own supports and participate fully in the relief effort. Local people are not considered passive beneficiaries but individuals with assets and resources. The guidelines also call for support to be provided from within the community, as well as by outsiders. Furthermore, the guidelines emphasize multisectoral action. The way any type of humanitarian response is implemented has implications, beneficial or detrimental, for the mental health and psychosocial wellbeing of the affected population. Effective or harmful supports may result from the ways in which aid is organized or delivered. Assessment and follow-up is conducted and, if need be, referral to mental health treatment for those with extreme problems, such as inability to function, suicidality, and alcohol dependence in the immediate aftermath of the crisis. Concern about risk of stigmatization and use of scarce resources suggests that referrals for unneeded therapy are to be prevented. Mental health and psychosupport is organized in a layered system of complementary supports that meets the needs of different groups in an intervention pyramid, with basic services and security at the base, followed by community and family supports, focused nonspecialized supports, and specialized services at the peak. Table 4 summarizes the action sheets for minimum response that were developed by the IASC.

More recently, based on the principles of PFA, an international group of eminent disaster researchers have synthesized the immediate and intermediate response to mass trauma into five basic points: promotion of sense of safety; promotion of calm; promotion of sense of self and collective efficacy; promotion of connectedness; and promotion of hope.[34••] The authors of this report provide a detailed outline of recommendations on how to achieve these goals, both at the public health and individual levels. Table 5 provides an abbreviated summary of their recommendations for the public health sector.

Conclusion

Research in recent years on MHS following disasters has advanced considerably. We currently have a better understanding of utilization patterns, antecedents and barriers to care, and have developed more promising models for delivering MHS.[35] The IASC guidelines will hopefully also provide improved MHS postdisaster to developing countries. The implementation of PFA, along with utilization of the five empirically supported intervention principles, following disasters will hopefully improve the psychological health of affected populations in the future. Much more research, however, is needed on MHS, in particular among populations with scarce resources.