PTSD and substance use: unrecognized sequelae of bioterrorism in primary care providers.

Background: Psychological casualties following public health emergencies are likely to significantly outnumber physical casualties. However, postevent psychological disorders may be underrecognized by primary care providers (PCPs).

Methods: Rural PCPs in northern and central Florida were interviewed using a series of open-ended questions to assess knowledge of likely mental disorders, their risk factors, and preferred treatment options following such events (n = 21).

Results: PTSD was identified by 14% and substance abuse by 10% of the sample. Physicians were significantly more likely to identify posttraumatic stress disorder (PTSD) as an expected postevent psychological disorder than nonphysician providers. PCPs were significantly more likely to endorse counseling (86%) than medications (43%) as a preferred treatment option.

Conclusions: Our findings support the need for increased education and training regarding the mental health consequences of bioterrorism in rural PCPs, particularly for nursing-level and other nonphysician providers. Improvements in knowledge may enhance preparedness for such emergencies.

Key Words: rural mental health, posttraumatic stress disorder, bioterrorism, rural primary care, substance use disorders.

**********

Recent work on the potential impact of terrorism, bioterrorism and similar public health emergencies on the primary care system has drawn attention to the mental health consequences of such events. (1,2) The US Department of Defense has estimated that psychological casualties following such crises are expected to outnumber physical casualties by 5 to 1. (3) Primary care providers (PCPs) will be on the frontline of mental healthcare following these major public health emergencies, particularly in rural areas. (1,2) The most frequently documented psychological disorders following disasters and mass casualty events include posttraumatic stress disorder (PTSD), depression and anxiety disorders. (4,5) Increased substance use has also been reported following the attacks of September 11, 2001, particularly among persons with comorbid PTSD (6,7) and depression. (7)

It has been estimated that approximately 28% of terrorism survivors will develop PTSD. (8) Although there has been less attention focused on postevent substance use, one study found that roughly 27% of respondents who lived near the World Trade Center (WTC) reported increased use of cigarettes, alcohol or marijuana 6 months after the September 11 attacks. (9) Prior studies that have not been specifically linked to disasters or terrorism have found low rates of recognition for PTSD among PCPs (10-12) and a recent US national survey, not directly tied to public health emergencies, found that PCPs query less than a third of their patients about alcohol or drug use. (13) To the best of our knowledge, PCP awareness of PTSD and substance use disorders has not been examined specifically in relation to bioterrorist attacks. Awareness of such risk factors may assist PCPs in identifying and triaging psychological casualties in the aftermath of these potentially catastrophic events.

Given the dearth of behavioral health specialists in rural communities, rural PCPs, even more than their urban counterparts, will be the main providers of mental healthcare following bioterrorism. (14,15) However, there may be reduced awareness of postevent mental health consequences among rural PCPs. For example, the study by D'Amico et al (13) found that patients in rural areas were less likely to be identified or advised by their PCP regarding alcohol use compared with those in urban areas. Although attention has focused largely on the vulnerability of urban centers to bioterrorism, rural areas are less well defended (ie, "soft targets") and their healthcare systems are less prepared (16) for such events. A major attack in an urban center may also lead survivors to converge on neighboring rural communities, thereby overwhelming the rural healthcare system, which is often understaffed and underfunded in many communities.

The purpose of the present study was to assess knowledge of likely mental disorders and the risk factors for the development of such disorders following bioterrorism in a sample of rural PCPs in northern and central Florida. Florida was considered a particularly appropriate location for the present study given that the state was the initial site of the anthrax attacks of 2001, and because roughly half of the state's counties are considered rural. Based on prior research, (10-12) we expected low recognition rates for PTSD as a likely postevent psychological disorder. We explored PCP knowledge of risk factors for the development of postevent mental disorders as well as knowledge of increased substance use disorders as an expected consequence of bioterrorism. Finally, we examined PCP knowledge of treatment options for postevent mental disorders. To investigate differences in responses based on provider training, we compared the responses of physicians with those of nonphysician providers. It was expected that physician knowledge of these aspects of postevent mental health would exceed that of nonphysician providers.

Materials and Methods

Participants and Procedure

Twenty-one primary care providers (PCPs) (67% female) from 13 clinics across northern and central Florida served as participants. The mean age was 45.5 years (SD = 8.7; range = 32-60 yr); ethnic composition was 91% Caucasian, 5% Asian, 5% African-American. Participants consisted of 38% physicians, 52% nurses and 10% physician assistants/advanced nurse practitioners. This study was approved by the UCLA, University of Florida, Florida State University, and VA GLAHS institutional review boards (IRBs).

Participants were interviewed by the investigators using a series of open-ended questions as part of a larger study examining preparedness for rural mental health needs following bioterrorism and other public health emergencies. Interviews were conducted between October 2003 and September 2005. Participants were told: "Now I'm going to ask you about the kinds of mental health problems you might encounter in patients following a terrorist attack such as the attacks on the World Trade center on 9/11/01, or a disease outbreak, or a biologic agent such as anthrax."

To assess knowledge of postevent mental disorders, respondents were asked: "What types of mental disorders do you think you would be likely to see in patients following such an event?" To assess knowledge of risk factors for such disorders, respondents were asked: "Are you aware of any risk factors for developing mental health problems following a bioterrorist attack?" and "Are you aware of certain patients or certain types of patients who might be more likely to develop mental disorders following such an event?" To evaluate knowledge of treatment options for postevent mental disorders, respondents were asked: "What do you think are the best treatment options for people who develop mental disorders following a bioterrorist attack?" Interview responses to these questions were tape recorded and transcribed verbatim. Responses were then categorized into relevant categories.

Statistical Analysis

Differences between the responses of physicians and nonphysicians were compared using independent sample Student's t tests for continuous data and [chi square] tests for categorical data. For the latter class of data, Fisher's exact tests were used in the event that one or more cells had an expected frequency of five or less. A standard probability level of P < 0.05 was used.

Results

Recognition of Expected Postevent Mental Disorders

Respondents identified between one and four likely postevent mental disorders (M = 1.9; SD = 0.9). As shown in the figure, anxiety (91%) and depression (57%) were the most commonly identified disorders, whereas PTSD was identified by only three participants (14%) and substance abuse was identified by only two participants (10%). In addition, manic depression, obsessive-compulsive disorder, and schizophrenia were each identified by one respondent (data not shown). Various other psychological sequelae (ie, symptoms of mental disorders, rather than mental disorders per se) were also identified, including anger, panic attacks, grief, sleep disturbance, somatization and paranoia (data not shown). Each of these sequelae was identified by one participant with the exception of panic attacks and anger which were each identified by two participants.

There were no significant differences in the likelihood of identifying anxiety or depression between physicians and nonphysician providers. However, physicians were significantly more likely to identify PTSD as a postevent mental disorder than nonphysician providers (Fisher's exact, P < = 0.042). Although the overall level of recognition of substance abuse was very low, the likelihood of identification did not differ between physicians and nonphysician providers (ie, only one respondent in each category identified substance abuse).

Awareness of Risk Factors for Postevent Mental Disorders

Most participants (76%) identified at least one possible risk factor for the development of a postevent mental disorder (M = 1.5; SD = 1.3; range = 0-5). Physicians (M = 2.3; SD = 1.4) identified significantly more potential risk factors than nonphysician providers (M = 1.1; SD = 1.1) (t[19] = 2.13, P < 0.05). The table lists the risk factors identified by respondents. As shown in the table, the most commonly cited risk factor was history of prior mental disorder, followed by lack of social support/social isolation. Notably, one respondent identified existing substance abuse as a potential risk factor for the development of postevent mental disorders. Roughly 24% of PCPs stated that they were not aware of any risk factors for the development of postevent mental disorders. There were no differences in the likelihood of identifying potential risk factors or in the lack of awareness of possible risk factors between physicians and nonphysician providers.

Treatment Options for Postevent Mental Disorders

When asked about the best treatment options for postevent mental disorders, most respondents (86%; n = 18) endorsed counseling, whereas fewer respondents (43%; n = 9) endorsed medication; two participants (11%) indicated that they did not know the best treatment options. A [chi square] test indicated that significantly more respondents endorsed counseling than medication (chi square (1) = 6.86, P < = 0.01). Just over a third of respondents endorsed both medication and counseling (38%; n = 8), although one-fourth of these participants indicated that counseling would be the superior approach. A number of respondents indicated that group counseling, either on its own (19%; n = 4) or in combination with individual counseling (14%; n = 3), would be a preferred approach. Physicians and nonphysicians did not differ on any of these responses.

Discussion

Consistent with our predictions, interview responses in this sample of rural PCPs revealed little awareness that PTSD and substance use disorders constitute likely mental health consequences of bioterrorism. PTSD was identified by only 14% of providers (Fig.). As hypothesized, providers' level of training was an important factor in recognition of PTSD--physicians were significantly more likely to identify PTSD than nonphysician providers (P [less than or equal to] 0.05). Given that PTSD is formally classified as an anxiety disorder, (17) it is possible that some respondents may have intended their responses to include any anxiety disorder, including PTSD. Indeed, over 90% of respondents identified anxiety as an expected postevent mental disorder. Nevertheless, for the purposes of this study we were specifically interested in determining whether PCPs would recognize that PTSD, apart from more generalized anxiety or panic disorders, would be a common postevent consequence, as the symptoms of PTSD (eg, nightmares; flashbacks) are quite different from those found in the other anxiety disorders. Our findings also indicated that substance abuse was identified by only two (10%) of the PCPs we interviewed (Figure). It should be noted that substance abuse is a mental disorder with established criteria (17); previous studies (6,7,9) have focused on increased postevent substance use that may not meet criteria for a formal diagnosis of substance abuse. One possibility is that some respondents may not have considered substance abuse to be a mental disorder. Alternatively, some respondents may have recognized increased substance use as a likely postevent consequence but did not regard it as rising to the level of a disorder. Nevertheless, our findings are disconcerting in light of previous work showing that approximately 27% of residents living near the WTC reported increased substance use 6 months after the September 11 attacks. (7)

[FIGURE OMITTED]

The present findings agree with prior work indicating a low diagnosis rate of PTSD by primary care physicians. (12) Our results are also consistent with recent research showing that less than a third of PCPs queried their patients about drug or alcohol use, (13) with rural patients being less likely to be identified and advised regarding such use compared with their urban counterparts. (13) It should be noted that the present study used open-ended questioning, in contrast to previous research which assessed provider knowledge using clinical vignettes. (10) It is unclear which method is preferable given the dearth of studies in this area. Clinical vignette studies have not generally assessed provider knowledge of mental disorders specifically in relation to certain types of events such as public health emergencies. Our study was intended to evaluate the extent to which specific postevent mental health consequences would be readily recognized by providers following bioterrorism and similar emergencies as presumably these would be the conditions that providers would be more likely to screen for and ask patients about. In the aftermath of such an emergency, healthcare providers will need to conduct rapid assessments of presenting patients, possibly without access to a complete medical history. Thus, existing PCP knowledge of common mental disorders, their key symptoms and risk factors, are important considerations in evaluating preparedness for bioterrorism and other emergencies.

Although more than three-quarters of the sample (76%) identified at least one possible risk factor for the development of a postevent psychological disorder (see Table), roughly one-fourth (24%) were not able to identify any such risk factors. As hypothesized, physicians identified a significantly greater number of possible risk factors for the development of a postevent mental disorder than nonphysician providers (P < 0.05). However, there were no differences in the kinds of risk factors identified by respondents based on level of training.

To the authors' knowledge, no existing studies have assessed PCP knowledge of risk factors for the development of postevent mental disorders. The majority of prior research has focused on risk factors for PTSD (18,19) which include female gender, low socioeconomic status, prior psychiatric history, prior exposure to traumatic events or chronic stress, living in a highly traumatized or disrupted community, and direct, severe exposure to the traumatic event (this list is not exhaustive). It is well recognized that women are roughly twice as likely to develop PTSD, as well as anxiety and depressive disorders, compared with men. (17) Men on the other hand are more prone to substance use disorders than women. (17) However, gender was not mentioned by PCPs as a likely risk factor for either disorder (see Table). On the other hand, just over half the sample stated that a prior history of mental disorder was an established risk factor for PTSD.

Notably, PCPs were significantly more likely to endorse counseling (86%) rather than medication (43%) as a preferred treatment option for postevent mental disorders (P < 0.01). Approximately one-third of the sample endorsed both medications and counseling, with no differences based on provider level of training. The stated preference for counseling is notable given that rural communities are much more likely than urban areas to have limited access to counseling services and other specialty mental health services. (20) The reasons for these findings are unclear. It may be that the PCPs we interviewed believed psychotropic medications to be less effective and/or less acceptable to their patients following bioterrorism. Alternatively, these findings could reflect a lack of comfort among PCPs in treating postevent mental disorders using psychotropic medications and/or the recognition of a need for social and community support following such events, despite the stigma often associated with mental health treatment in rural communities.

Caveats to the present study should be mentioned. The main limitation of this investigation is the small sample size. Replication of the current findings in a larger sample would increase confidence in the findings. It should be noted that the present sample was selected using purposive sampling (21) based on the participants' status as primary care providers in rural communities and was not designed to be representative of all rural PCPs across Florida. The current study took place in Florida, a state with a history of recurrent hurricanes, most notably the unusually active hurricane season of 2004. Two of our interviews took place after the hurricanes of 2004, whereas the remainder was conducted before the hurricane season. The impact of the 2004 hurricanes on PCP knowledge of postevent mental disorders is not known, and the limited sample size precludes examining its effect on this study. Although bioterrorism and natural disasters share some common characteristics, each type of event has its unique features and it is unclear if the present results would have differed if we had asked PCPs about likely mental disorders following natural disasters. Nevertheless, it can be argued that due to the recurrent possibility of hurricanes in the state, PCPs in Florida may be more likely to be aware of the potential psychological consequences of mass trauma events than PCPs in states or areas that are less prone to natural disasters. Further research is needed to examine the relationship between knowledge of mental health consequences related to recurrent events such as hurricanes in Florida and uncommon events such as bioterrorism.

Our findings support the need for increased education and training regarding the mental and behavioral health consequences of bioterrorism and other public health emergencies in rural PCPs, particularly for nursing-level and other nonphysician providers. In a recent study of family physicians, 95% believed bioterrorism to be a real threat, but only 26% reported that they would know what to do as physicians if such an event were to occur; only 18% reported they had received previous training in bioterrorism preparedness. (22) Educational efforts regarding postevent psychological sequelae will likely be particularly important in rural primary care clinics given the limited resources for mental healthcare in rural communities. (23) Such educational interventions should be tailored to rural PCPs and may be delivered via media that are accessible to remote clinic sites (eg, web-based formats). The development and widespread dissemination of evidence-based guidelines for postevent care may also improve diagnosis and treatment following a public health emergency. Notably, PCPs in the current sample recognized the importance of treatment, particularly counseling for postevent mental disorders. Thus, continued efforts should be directed toward facilitating referrals to mental health professionals and increasing the availability of such professionals in rural communities throughout the US. Improvements in knowledge of expected mental disorders and their risk factors may enhance preparedness for bioterrorism and other such emergencies by assisting rural PCPs in the identification and triage of immediate postevent psychological casualties, as well as the care of postacute and chronic mental health needs, following such events.

Conclusion

The under-recognition of PTSD and substance abuse as potential consequences following bioterrorism suggest that existing efforts to educate PCPs should be enhanced. Although rural communities may be particularly vulnerable to long-term mental and behavioral health needs following public health emergencies due to a lack of specialty mental health providers in many rural areas, our findings are also relevant to urban areas, as PCPs in urban areas are also likely to encounter an increased incidence of psychological disorders following these events. Additional research should focus on determining whether these results generalize to other geographic areas and on designing interventions that target these deficits in provider knowledge.

Acknowledgments

This project was supported by Grant 7 R04 RH01310 from the Health Resources and Services Administration awarded to the first author and Grant 7 U01 HS14355 from the Agency for Healthcare Research and Quality awarded to the second author. Dr. Dobalian is supported by a Veterans Administration Health Services Research and Development Merit Review Entry Program award (MRP 03-328).

References

1. Dobalian A, Tsao JCI, Putzer GJ, et al. A commentary on improving rural community preparedness for the chronic health consequences of bioterrorism and other public health emergencies. Journal of Public Health Management and Practice. (in press).

2. Lacy TJ, Benedek DM. Terrorism and weapons of mass destruction: managing the behavioral reaction in primary care. South Med J 2003; 96:394-399.

3. Warwick MC. Psychological effects of weapons of mass destruction. The Beacon, National Domestic Preparedness Office Newsletter 2001:1-8.

4. Norris FH. Epidemiology of trauma: frequency and impact of different potentially traumatic events on different demographic groups. J Consult Clin Psychol 1992;60:409-418.

5. Smith EM, North CS, McCool RE, et al. Acute postdisaster psychiatric disorders: identification of persons at risk. Am J Psychiatry 1990;147:202-206.

6. Grieger TA, Fullerton CS, Ursano RJ. Posttraumatic stress disorder, alcohol use, and perceived safety after the terrorist attack on the Pentagon. Psychiatr Serv 2003;54:1380-1382.

7. Vlahov D, Galea S, Resnick H, et al. Increased use of cigarettes, alcohol, and marijuana among Manhattan, New York, residents after the September 11th terrorist attacks. Am J Epidemiol 2002;155:988-996.

8. Gidron Y. Posttraumatic stress disorder after terrorist attacks: a review. J Nerv Ment Dis 2002;190:118-121.

9. Vlahov D, Galea S, Ahem J, et al. Sustained increased consumption of cigarettes, alcohol, and marijuana among Manhattan residents after September 11, 2001. Am J Public Health 2004;94:253-254.

10. Munro CG, Freeman CP, Law R. General practitioners' knowledge of post-traumatic stress disorder: a controlled study. Br J Gen Pract 2004; 54:843-847.

11. Samson AY, Bensen S, Beck A, et al. Posttraumatic stress disorder in primary care. J Fam Pract 1999;48:222-227.

12. Taubman-Ben-Ari O, Rabinowitz J, Feldman D, et al. Post-traumatic stress disorder in primary-care settings: prevalence and physicians' detection. Psychol Med 2001;31:555-560.

13. D'Amico EJ, Paddock SM, Burnam A, et al. Identification of and guidance for problem drinking by general medical providers: results from a national survey. Med Care 2005;43:229-236.

14. Hartley D, Korsen N, Bird D, et al. Management of patients with depression by rural primary care practitioners. Arch Fam Med 1998;7:139-145.

15. Rost K, Fortney J, Fischer E, et al. Use, quality, and outcomes of care for mental health: the rural perspective. Med Care Res Rev 2002;59:231-271.

16. Office of Rural Health Policy. Rural communities and emergency preparedness. Available at: ftp://ftp.hrsa.gov/ruralhealth/RuralPreparedness.pdf. Accessed October 25, 2005.

17. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Washington, DC, American Psychiatric Association, 1994.

18. Halligan SL, Yehuda R. Risk factors for PTSD. PTSD Res Q 2000;11:1-8. Available at: http://www.ncptsd.va.gov/publications/rq/rqpdf/V11N3.pdf. Accessed July 8, 2006.

19. Norris FH, Christopher M, Byrne ED, et al. Risk factors for adverse outcomes in natural and human-caused disasters: a review of the empirical literature. A National Center for PTSD Fact Sheet. Available at: http://www.ncptsd.va.gov/facts/disasters/fs_riskfactors.html. Accessed October 28, 2005.

20. Merwin E, Hinton I, Dembling B, et al. Shortages of rural mental health professionals. Arch Psychiatr Nurs 2003;17:42-51.

21. Patton MQ. Qualitative Evaluation and Research Methods (2nd edition). Newbury Park, CA, Sage Publications, 1990.

22. Chen FM, Hickner J, Fink KS, et al. On the front lines: family physicians' preparedness for bioterrorism. J Fam Pract 2002;51:745-750.

23. Tsao JC, Dobalian A, Wiens BA, et al. Posttraumatic stress disorder in rural primary care: improving care for mental health following bioterrorism. J Rural Health 2006;22:78-82.

The plea of necessity, that eternal argument of all conspirators.
--William Henry Harrison

Jennie C.I. Tsao, PhD, Aram Dobalian, PhD, JD, Brenda A. Wiens, PhD, Julius A. Gylys, PhD, Art Clawson, MS, and Robert Brooks, MD

From the Pediatric Pain Program, Department of Pediatrics, David Geffen School of Medicine at UCLA, Los Angeles, CA; the VA Greater Los Angeles Healthcare System (GLAHS) HSR & D Center of Excellence for the Study of Healthcare Provider Behavior and Department of Health Services, UCLA School of Public Health, Los Angeles, CA; the National Rural Behavioral Health Center, Department of Clinical and Health Psychology, University of Florida, Gainesville, FL; and the Division of Health Affairs, and Department of Family Medicine and Rural Health, Florida State University College of Medicine, Tallahassee, FL.

Reprint requests to Jennie C. I. Tsao, Pediatric Pain Program, Department of Pediatrics, David Geffen School of Medicine at UCLA, 10940 Wilshire Boulevard, Suite 1450, Los Angeles, CA 90024. Email: [e-mail].

Accepted April 25, 2006.

RELATED ARTICLE: Key Points

* Respondents in this sample of rural primary care providers (PCPs) revealed low levels of awareness that posttraumatic stress disorder (PTSD) (14%) and substance abuse (10%) constitute likely mental health consequences of bioterrorism.

* PCPs in the present study were significantly more likely to endorse counseling, (86%) rather than medications, (43%) as a preferred treatment option for postevent mental disorders, and one-third of the sample endorsed both medications and counseling.

* Our findings support the need for increased education and training regarding the mental and behavioral health consequences of bioterrorism and other public health emergencies in rural PCPs, particularly for nursing-level and other nonphysician providers.

* Improvements in knowledge of expected mental disorders and their risk factors may enhance preparedness for bioterrorism and other such emergencies by assisting rural PCPs in the identification and triage of immediate postevent psychological casualties, as well as the care of postacute and chronic mental health needs after such events.

* Increased efforts should be directed toward facilitating referrals to mental health professionals and increasing the availability of such professionals in rural communities throughout the U.S.

Table. Risk factors for development of mental disorders following
bioterrorist events as identified by rural primary care providers in
northern and central Florida (N = 21).

                                                       Number of
                                                       Respondents
                                                       Identifying the
Risk Factor                                            Risk Factor (%)

History of prior mental disorder                       11 (52)
Lack of social support/social isolation                 5 (24)
Life stress (e.g., loss of job; unstable environment)   3 (14)
Family history of mental disorder                       2 (10)
Children                                                2 (10)
Excessive worry                                         2 (10)
Substance abuse                                         1 (5)
Being directly affected by the event                    1 (5)
Low socioeconomic status (SES)                          1 (5)
Elderly                                                 1 (5)