Introduction

Hurricane Sandy was an unprecedented natural disaster for New York City that caused widespread power outages, shutdowns of public transportation, and property damage. Several New York City hospitals experienced evacuations and closures lasting weeks to months including New York University Langone Medical Center, the Manhattan campus of the VA New York Harbor Healthcare System, Hospital, Bellevue Hospital Center, and Coney Island Hospital. Storm damage compounded by the medically vulnerability of patients at these clinics created unexpected medical risks, especially for opioid-dependent patients. During prior comparable states of emergency, such as Hurricane Katrina in New Orleans and 9/11 in New York City, relapse rates for opioid-maintained patients increased due to discontinuity of care, and clinician uncertainty regarding emergency procedures and access to health information.1,2

A 2006 data analysis estimates there to be over 92,000 illicit opioid users in New York City,3 and the New York State Office of Alcoholism and Substance Abuse Services (OASAS) estimates that more than 75 % of patients at 112 substance abuse treatment programs in New York City had interruptions in treatment due to program closures during Hurricane Sandy.4 Previously, during Hurricane Katrina and 9/11, substance abuse treatment was disrupted for many reasons. Maxwell and colleagues found that during Hurricane Katrina, methadone dispensing was difficult because of the low number of publicly funded clinics available to low-income patients and the lack of a centralized information system documenting dosing treatment histories.5 Increases in drug use occurred during Hurricane Katrina for especially vulnerable populations such as low income African Americans.1 During disasters, opioid treatment programs are especially burdened by ambiguous emergency laws and strict regulations.6 In contrast, flexible regulation implementation in clinics has been correlated with resilience during natural disasters.7 During 9/11, methadone programs stayed open, but lacked telephone communication and therefore the capacity to verify doses, leading to difficulty making decisions regarding take-home methadone doses and to high stress and anxiety among staff.8 While a higher incidence of positive urine toxicology screenings were reported after 9/11,9 changes in drug use after 9/11 were varied and affected by income, geography, and access to services.2

In the case of Hurricane Sandy, New York City subways were suspended and highways were closed down due to damage for days to weeks, complicating patients’ commute to their clinics. Although this affected both methadone and buprenorphine patients, transportation issues may have disproportionately affected methadone maintenance patients in outpatient programs who were required to come into clinic daily for their dosing, while many buprenorphine treatment patients come less regularly for prescription refills.

Buprenorphine and methadone have different dispensing methods that are dictated by their narcotic classifications and corresponding Federal oversight. Since 1971, US regulations require methadone, a schedule II drug, to be dispensed for maintenance treatment on site in specialty clinics. In contrast, buprenorphine, a schedule III, was US FDA approved for office based treatment by prescription in 2002. Qualified physicians can apply for a Federal waiver to prescribe buprenorphine after completing 8 hours of training in buprenorphine management.10

Given the recent approval of buprenorphine for treatment of opiate dependence relative to methadone, few direct comparisons of buprenorphine and methadone within health service systems have been made. Hurricane Sandy presented a natural experiment to test the relative advantages and disadvantages of methadone and buprenorphine regulations and dispensing methods in the face of a major disruption of service. To analyze the effects of regulatory differences between methadone and buprenorphine on the continuity of care after Hurricane Sandy, we interviewed providers and administrators in New York City public sector clinics and practices that provide buprenorphine and/or methadone maintenance treatment.

Methodology

In an attempt to better understand the effects of Hurricane Sandy on substance abuse treatment in New York City, we analyzed semi-structured interviews with providers and administrators that offer pharmacological treatment for opioid dependence using methadone and/or buprenorphine in public clinics and in Medicaid accepting community practices. Face-to-face interviews with providers and administrators were conducted 8 to 20 weeks after the storm. Interviews lasted 30–60 min and addressed topics such as the impact of regulations for both methadone and buprenorphine on service provision after the storm, barriers to continuity of care, communication with regulators and agency leadership as well as patients and other care providers, and navigation of hospital emergency plans.

Public sector buprenorphine prescribers were recruited during site visits to the nine municipal hospitals and three VA medical centers in New York City that offer outpatient methadone and/or buprenorphine treatment. Participants were also recruited from a list of New York City buprenorphine certified prescribers, provided online by the Substance Abuse and Mental Health Services Administration (SAMHSA), cross-referenced with a list of physicians who accept New York State Medicaid. Index participants were asked to identify additional public sector buprenorphine prescribers for inclusion in the study in a modified snowball sampling technique in which new participants were solicited until no new names were offered. Of the 119 public sector buprenorphine and methadone prescribers contacted for participation in the study, 50 completed interviews between January and March of 2013 and 69 could not reached or declined participation. Trained graduate-level interviewers conducted face to face semi-structured interviews lasting 30–60 min regarding regulation barriers and facilitators pertaining to methadone and/or buprenorphine, as well as on diversion and barriers to treatment.

Interview transcripts were analyzed using iterative thematic coding techniques, including continuous comparison and pragmatic adaptation of grounded theory in order to develop relevant coding categories.11–14 Multiple coders were used for all transcripts to check inter-coder reliability. Discrepancies between coders were resolved through team discussion and consensus.

This research was conducted with oral informed consent procedures, data storage techniques designed to safeguard the confidentiality of participants’ identities, and participant protection from court subpoena of the study’s data as provided by a US Health and Human Services Certificate of Confidentiality. The study was approved by New York University School of Medicine’s Institutional Review Board.

Results

Eight administrators, 41 providers, and 1 provider/administrator agreed to the interview regarding their clinical experiences during and after Hurricane Sandy (Table 1).

TABLE 1 Number of administrator and provider participants
Full size table

Methadone Regulations

Methadone providers and administrators faced challenges in navigating regulatory agencies and providing continuity of care. Four themes emerged from our interviews: (1) poor communication with regulatory agencies, (2) clinic relocation problems, (3) lack of emergency preparedness strategies, and (4) dosage and patient status verification difficulties (see Table 2).

TABLE 2 Methadone provider interview themes
Full size table
  1. 1.

    Communication with regulatory agencies

    Methadone providers and administrators were more likely than buprenorphine providers and administrators to report communication problems with regulatory agencies such as the New York State Office of Alcoholism and Substance Abuse Services (OASAS), the US Drug Enforcement Agency (DEA), and the New York City municipal hospital administration. Problems included lack of contact with these agencies. Administrators and providers wanted greater clarity on the guidelines regarding releasing extra methadone doses before Hurricane Sandy. Overall, providers and administrators reported that if regulatory agencies had communicated with them prior to the hurricane, they would have released additional take-home doses of methadone to last patients through the disaster, therefore easing their continuity of care (see Table 2). Providers and administrators also commented on certain regulatory agencies’ inflexibility and their difficulty navigating agency guidelines for methadone. In one instance, because of a technical compliance issue, an agency shut down a methadone clinic in the middle of clinic hours. During a visit to another clinic, an agency demanded the hospital “cease and desist” the clinic’s emergency methadone dispensation as the three-day emergency dosing window had lapsed. Providers and administrators attempted to relocate the patients to another methadone program the same day and were dosing methadone until 7 pm that night.

  2. 2.

    Clinic relocation problems

    Many complications emerged after providers began to realize that Hurricane Sandy’s impact would last much longer than the DEA’s three-day maximum emergency distribution of methadone. Some public hospital clinics had to be relocated to other public hospitals which hosted their patients and staff. The largest issue pertaining to clinic relocation was transitioning the large patient population, their records, and obtaining the proper accreditation for residents and fellows to assist with methadone treatment. The large number of patients physically overcrowded hospital facilities and staff were overwhelmed. Attempts to relocate staff from closed facilities were met with accreditation problems for trainees with the Accreditation Council for Graduate Medical Education (ACGME); one of the providers we interviewed stated that he was the only doctor for a week before the addiction fellows and residents were approved to assist by the ACGME. As many as one third of clinic patients fell out of treatment due to clinic relocation and breakdowns in communication.

  3. 3.

    Lack of emergency preparedness strategies

    Other methadone providers and administrators described uncertainty regarding emergency preparedness and emergency procedures. In particular, there was no clear indication of when the emergency state ended. One respondent reported that due to lack of clarity surrounding emergency procedures, the pharmacy had to bend rules in order to maintain continuity of care. Administrators from hospitals that had to move their methadone program to a new facility after the storm worried that their license did not cover the new facility.

    Another concern was that patients crowded emergency department by turning to them for medication. This resulted in suboptimal, and sometimes dangerous dosing of methadone patients by emergency room doctors who either refused to dose methadone patients, or who, upon patient request, gave patients higher doses of methadone than their usual dose, risking fatal overdose. One provider reported that when patients had nowhere else to go for their medication, she would dose them out of her office. During this crisis, wait hours were so long, patients arrived as early as 5:30 am to wait to be dosed.

  4. 4.

    Dosage and patient status verification difficulties

    Difficulty verifying dose was another problem faced by methadone providers. Despite there being no regulatory guidelines that demanded dose verifications, methadone providers unsuccessfully attempted to verify doses. Some regulatory agencies recommended that in emergency conditions, providers take patients at their word, but providers felt great discomfort in doing so. Providers attempted to verify doses by accessing patient databases at other programs to improve upon the “just dose them and trust them” strategy. Patients presented used methadone bottles that did not always have doses or dates printed on them. Providers ultimately had to make judgment calls regarding dosing, and warned patients about the dangers of overdose.

Buprenorphine Regulations

Buprenorphine providers faced fewer regulatory issues than methadone providers. The emergent themes regarding buprenorphine included (1) buprenorphine clinic relocation, (2) lack of emergency preparedness strategies, and (3) dose verification (see Table 3).

TABLE 3 Buprenorphine provider interviews emergent themes
Full size table
  1. 1.

    Buprenorphine clinic relocation

    One issue mentioned was the difficulty of relocating to a new buprenorphine clinic. One buprenorphine provider described going through many bureaucratic steps before relocating. However, another buprenorphine provider stated that once buprenorphine treatment was relocated, physician accreditation was quick and they were able to resume care almost immediately.

  2. 2.

    Lack of emergency preparedness strategies

    A common report was that public buprenorphine clinics ordinarily relied heavily on medical residents and fellows to provide buprenorphine, but following Hurricane Sandy, clinics that were forced to relocate had difficulty transferring their medical residents with them. Since accreditation of resident and fellow training facilities had to be approved by the ACGME on a national level, the accreditation process was slow and delayed restoration of buprenorphine services.

  3. 3.

    Dose verification

    One provider mentioned that a barrier to prescribing buprenorphine to patients was a lack of electronic system with centralized records that would facilitate dosing during an emergency. Another provider noted that a lack of such a record led to dose reductions. The relocation clinic did not dispense film strips, therefore dose verification was also complicated by the switch from buprenorphine film strips to tablets due to different dispensation procedures required for each.

    Patients who relied on Medicaid were disproportionately affected since they could only fill out their prescriptions at the municipal hospital, which was shut down, while those who were privately insured could go to outside pharmacies to refill their prescriptions.

Discussion

Methadone providers reported more barriers to continuity of patient care than buprenorphine providers, including lack of clarity around emergency regulations, lack of communication with regulators, as well as lack of centralized electronic systems to confirm dosage. Methadone licenses only provided Methadone programs three days to dispense methadone in emergency facilities, and programs often needed the flexibility to do so for longer periods of time. During Hurricane Sandy, there was no established central registry to verify doses, despite this having been a major concern shortly after the similar crisis of Hurricane Katrina in New Orleans.

Buprenorphine treatment centers generally faced fewer problems providing continuity of care in the wake of Hurricane Sandy than methadone programs. Because patients could be prescribed up to a month’s supply of buprenorphine, they often had sufficient supplies during their dislocation from their clinics. Buprenorphine providers reported a lack of cross-coverage by colleagues. Their reliance on resident physicians in public clinics was hampered by the need for temporary clinics’ training credential approval. Buprenorphine providers also reported that their patients often had difficulties filling out their prescriptions in pharmacies whose supplies were affected by the storm. The flexibility of buprenorphine dispensing was an advantage under emergency conditions, but the guest dosing policies of methadone clinics provided a network of cross-coverage that was unavailable for buprenorphine patients.

This study had several limitations, including that only municipal and VA hospitals in four of the five boroughs of New York City were included in this study, which may limit the generalizability of our findings. There were many other areas affected by Sandy, including Staten Island, Long Island, and New Jersey, as well as private methadone and buprenorphine programs in New York City with providers who may have had different experiences than those interviewed. In addition, just under half of the physicians identified as public sector buprenorphine and methadone prescribers agreed to participate in the study, which may have biased the results.

Despite these limitations, our findings suggest a number of interventions that would improve continuity of care for opioid-maintained patients during times of service disruptions. They include equipping methadone treatment for flexible dosing, creating a central registry for buprenorphine dosing, a central database for methadone maintenance programs, strengthening cross-covering networks of buprenorphine providers and allowing for credential exemption for medical trainees during emergency settings.

Enhancing methadone maintenance program flexibility to approach that of buprenorphine treatment, such as facilitating guest dosing in alternate clinics, would not only facilitate care in the event of an emergency, but would also enhance daily care for patients who often experience treatment interruptions due to difficulties traveling to clinics because of comorbidities or lack of resources, for example. Central databases for buprenorphine and methadone treatment would similarly allow for comprehensive knowledge of dosing histories and treatment plans while easing administrative hurdles to continuity of care.

Stronger networks of buprenorphine providers have eased buprenorphine dispense in the past and can help improve the continuity of care for buprenorphine patients who are displaced from their primary clinic.15 Such networks promise not only to enhanced cross-coverage during routine and emergency service disruptions, but also to address a widespread concern among current and potential buprenorphine prescribers about the lack of professional support and mentoring available to prescribers in managing buprenorphine patients.16 During Hurricane Sandy, providers depended on the prescribing capabilities of residents and fellows, and credentialing delays for these trainees became increasingly cumbersome for continuity of care of patients at host hospitals. The ACGME’s creation of exemptions for trainees working under emergency conditions would benefit underserved areas where teaching hospitals often preside. The implementation of centralized database for opioid maintenance programs, provider cross-coverage networks, and flexible methadone treatment would also prevent delays in treatment due to emergencies and due to routine system disruptions in resource poor settings, therefore decreasing patients’ risk of withdrawal and relapse.