Abstract
The New York City (NYC) Department of Health and Mental Hygiene (DOHMH) used multiple methods to provide guidance to healthcare providers on the management and prevention of Zika virus disease during 2016. To better understand providers' use of information sources related to emerging disease threats, this article describes reported use of information sources by NYC providers to stay informed about Zika, and patterns observed by provider type and practice setting. We sent an electronic survey to all email addresses in the Provider Data Warehouse, a system used to maintain information from state and local health department sources on all prescribing healthcare providers in NYC. The survey asked providers about their use of information sources, including specific information products offered by the NYC DOHMH, to stay informed about Zika during 2016. Trends by provider type and practice setting were described using summary statistics. The survey was sent to 44,455 unique email addresses; nearly 20% (8,711) of the emails were undeliverable. Ultimately, 1,447 (5.8%) eligible providers completed the survey. Most respondents (79%) were physicians. Overall, the most frequently reported source of information from the NYC DOHMH was the NYC Health Alert Network (73%). Providers in private practice reported that they did not use any NYC DOHMH source of information about Zika more frequently than did those working in hospital settings (29% vs 23%); similarly, private practitioners reported that they did not use any other source of information about Zika more frequently than did those working in hospital settings (16% vs 8%). Maintaining timely and accurate databases of healthcare provider contact information is a challenge for local public health agencies. Effective strategies are needed to identify and engage independently practicing healthcare providers to improve communications with all healthcare providers during public health emergencies.
The New York City (NYC) Department of Health and Mental Hygiene used multiple methods to provide guidance to healthcare providers on the management and prevention of Zika virus disease during 2016. To better understand providers' use of information sources related to emerging disease threats, this article describes reported use of information sources by NYC providers to stay informed about Zika and patterns observed by provider type and practice setting.
H
Adding to this challenge, especially in the context of emerging infectious diseases, there may be uncertainty or incomplete information about the pathogen, the risk to certain groups, or the effectiveness of prevention and treatment strategies available. Public health recommendations may rapidly change throughout the response to an event as more information becomes available. Recognizing the importance of effective communication to healthcare providers during public health emergencies, public health officials often prioritize the development of provider communications as a key element of public health emergency response. However, evidence supporting the effectiveness of various communication methods between public health authorities and healthcare providers is limited.
The response of the New York City (NYC) Department of Health and Mental Hygiene (DOHMH) to the emergence of Zika virus as a potential public health threat in 2016 illustrates the complexity of the information environment for healthcare providers. It also highlights the role of the local health department in providing timely and accurate information to the healthcare system during an emergency. In NYC's large and complex healthcare system, healthcare providers can receive communications from multiple public health sources in addition to the NYC DOHMH. These include the Centers for Disease Control and Prevention (CDC) Health Alert Network (HAN) and the New York State Department of Health (NYS DOH).
The NYC DOHMH uses a local Health Alert Network to distribute email messages to providers in New York City. Annually, the NYC DOHMH sends 40 to 50 numbered health alerts and advisories, providing time-sensitive information about public health threats. Health alerts provide high-priority, vital information that warrants immediate attention from clinicians, while health advisories provide important information about a specific incident or situation. While anyone with an email address can register online to receive the NYC DOHMH HAN, the health department also sends HAN information via email to NYC providers identified from several sources and maintained in the NYC DOHMH Provider Data Warehouse (PDW). These messages may also be filtered through or amplified by healthcare organizations, professional societies and associations, or a growing number of healthcare coalitions. Information about an emerging threat that is disseminated by healthcare organizations and professional societies or associations may be identical to, aligned with, or contradictory to public health messages and recommendations; as such, each additional information source represents an opportunity to either reinforce or further confuse public health messaging during a response to an emerging health threat.
During the health department's Zika response, beginning in 2016, the NYC DOHMH used the HAN to send 5 alerts and 11 advisories related to the clinical management of patients at risk for Zika. The first of these, issued on January 16, 2016, was a health advisory titled “Microcephaly Associated with Zika Virus.” 1 During this same period, the DOHMH hosted 7 provider conference calls open to all healthcare providers, 1 webinar, and 1 in-person conference for regional perinatal centers. DOHMH experts provided speakers for clinical audiences, such as grand rounds. The DOHMH specifically targeted certain physician specialties in an effort to reach those caring for women of childbearing age who might be at highest risk for the complications of Zika virus infection during pregnancy. Activities included conducting in-person outreach to obstetric practices in neighborhoods with high concentrations of residents born in countries affected by Zika; redistributing HAN announcements through local chapters of the American College of Obstetricians and Gynecologists, the American Academy of Family Physicians, and the American Academy of Pediatrics; and including messages at the top of each health alert asking recipients to share with colleagues in specialties that might encounter patients at risk for Zika virus infection. 1
Nevertheless, requests for Zika virus testing and reporting of Zika virus infections remained lower than expected, particularly in certain neighborhoods in NYC. 2 To better describe how healthcare providers in NYC attended to and used public health information sources about Zika, the DOHMH conducted a survey of all healthcare providers in NYC. This article describes how the use of information sources varies by provider type and practice setting in New York City.
Material and Methods
Survey Design and Data Collection
The survey instrument was developed to collect demographic and practice information, identify information sources used by providers to learn about current guidelines for assessment and treatment of Zika virus, and describe providers' preferred information sources for emerging health threats. Questions specific to NYC DOHMH provider communication messages and practices were designed to elicit information that could be used to improve DOHMH communications in future emergency responses.
Individual survey links were sent using SurveyMonkey from a DOHMH email address on March 31, 2017. Biweekly reminders were sent out 4 times between March 31 and June 1. To assist participants with recall, they were asked to consider their use of various information sources during all of 2016, the period of time during which most NYC DOHMH communications related to Zika were released, and representative images of each type of NYC DOHMH communications product were included for respondents to reference.
Undeliverable email addresses were extracted after each successive email campaign and excluded from subsequent rounds of reminders. In addition, in compliance with the 2003 federal CAN SPAM act, email recipients were provided the option in each email message to unsubscribe from future survey-related reminders. Unsubscribing from reminders did not unsubscribe a provider from other future DOHMH emergency communications. Live email support for questions pertaining to the survey was provided by survey design team staff.
Sampling Methods
The survey targeted healthcare providers working in NYC during the local health department response to the emergence of Zika as a health threat. Since no gold-standard list of healthcare providers exists, we distributed the survey electronically to all email addresses contained in the NYC DOHMH Provider Data Warehouse, attempting to reach all healthcare providers in New York City that are currently known to the NYC DOHMH. The data warehouse aggregates provider information, including email addresses and other data fields, from several sources: the New York State Health Commerce Communications Directory (ComDir), the New York State Provider Network Data System (PNDS), and the NYC DOHMH's e-City Health Information directory (e-CHI), which are all regularly extracted to update the data warehouse. Specifically, the ComDir data is updated monthly to the Provider Data Warehouse, and PNDS data is updated quarterly.
ComDir is a secure online communication system operated by New York State Department of Health. It contains contact information for healthcare providers and other professionals in New York State and is managed in the New York State Department of Health's Health Commerce System. 3 Healthcare providers in New York are required to register in ComDir to order New York prescription pads, subscribe to state public health alerts, and access state registries; registered providers are required to update their passwords in the system at least every 1 to 2 years. Because the Health Commerce system is used by individual providers as well as many types of healthcare organizations and local health departments, there are a wide range of provider types and roles that could be represented in this source.
The PNDS is an electronic system that captures data about providers and service networks contracted to health insurers operating in New York State. 4 Because contact information in this system is used by insurance companies for provider reimbursement, it is likely to be more timely and accurate than other sources. NYS DOH provides quarterly updates from this system to the NYC DOHMH.
The e-CHI distribution list is a subscriber database of online registration open to the general public operated by the NYC DOHMH. Subscribers receive an evidence-based clinical bulletin 8 to 10 times per year.
In addition, extractions of provider emails from part of the NYC DOHMH's Citywide Immunization Registry and the NYC Department of Education's (DOE) Automated Student Health Record are also included in the Provider Data Warehouse. The extraction from the NYC DOHMH Citywide Immunization Registry is updated weekly. Permission to use this contact information for emergency health communications is obtained at the time of inclusion in the Provider Data Warehouse. This project was determined by the NYC DOHMH Institutional Review Board to be exempt from human subject review.
Analysis
Summary analyses were conducted using SAS software version 9.4 and Stata software version 14. Analysis was restricted to eligible participants. Eligible participants were defined as healthcare providers who consented to participate in the survey and worked in New York City during 2016 (the period of time during which the NYC DOHMH released most provider communications specific to Zika). These were respondents that reported they either directly provided or indirectly coordinated health or medical services either full-time or part-time in New York City at any time during 2016. Other respondents were determined to be ineligible and excluded from this analysis. Completed surveys were those in which the respondent completed practice setting and provider type questions and completed survey questions of interest related to information source and communication system preference.
The overall response rate was calculated using the American Association for Public Opinion Research standard definitions. 5
Summary statistics were compiled for self-reported provider type and practice setting, and used to describe trends in use of information sources used to keep providers, their colleagues, and their patients informed about Zika during 2016. Provider type was determined based on the response to the question, “What category best describes your profession or occupation at [your primary practice site] during 2016?” Providers were reclassified as follows: physician (MD or DO), nurse professional (includes registered nurse, licensed practical nurse, certified nursing assistant, certified nurse midwife, other midwife, nurse practitioner, or self-described as a nurse professional, including nurse manager), or other health professional (includes administrator, clinical laboratory personnel, dentist, other dental professional, health educator, counselor, pharmacist, physician assistant, psychologist [PhD, PsyD, EdD], other mental health provider, social worker [LCSW, MSW, DSW, LMSW, etc], veterinarian). Researchers classified self-described responses into 1 of the 3 provider type categories: physician, nurse professional, or other health professional.
Practice setting was determined based on the response to the question, “What best describes your principal practice in 2016?” Settings were categorized as follows: solo or private practice, group practice, hospital-affiliated practice (includes inpatient and outpatient), community practice (includes community health centers, clinics, urgent care, residential sites [nursing home, assisted living, hospice, long-term care], and other community-based facilities), and other (includes self-reported facility types that could not be reclassified into 1 of the other categories).
Respondents were also asked to rank 6 preferred methods of receiving NYC DOHMH public health alerts and guidance (conference calls, email, hard copy or through regular mail, in-person presentations, online webinar sessions, and via hospital/clinic administrators or leadership), with 1 being the most preferred and 6 being the least preferred method.
The percent of respondents in each provider type and practice setting that reported using each information source to stay informed about Zika was calculated to provide a descriptive summary of the responses observed among survey respondents.
Results
Individual survey links were sent to 44,455 unique email addresses contained in the Provider Data Warehouse; of 3,239 respondents who consented to participate in the survey, 2,043 NYC healthcare providers were eligible based on screening criteria, and 1,447 completed the survey questions relevant to this analysis (Figure 1). After eliminating 8,711 undeliverable or bounced emails, we calculated an overall response rate of 8.1% and a completed survey response rate of 5.8%.
Participation by Healthcare Providers Identified Using NYC Department of Health Provider Data Warehouse Contact Information in Survey Regarding Public Health Emergency Communications, New York City, 2016
Note: The New York City Department of Health and MentalHygiene Provider DataWarehouse aggregates healthcare provider information fromstate and local sources and is intended to include all healthcare providers in New York City.
Slightly more respondents were female, compared to male (50.9% vs 43.5%), and most respondents were white (58.4%) (Table 1). Among respondents, the most common provider type reported was physician (78.7%). The most commonly reported principal practice type was a hospital-based practice (42.0%). Private practice (25.8%), group practice (13.6%), and community-based facility (18.0%) were also reported. For fewer than 1% of respondents, the principal practice type could not be determined and was classified as “some other type of facility.”
Demographics of Eligible Respondents to a Survey Regarding Public Health Emergency Communications Distributed Electronically to Email Addresses Contained in the NYC DOHMH Provider Data Warehouse, New York City, 2016 a
Eligible respondents were those that reported they were a healthcare provider that practiced in New York City during 2016.
m/dk/r defined as missing, don't know, and refused to answer.
Provider type was determined based on the response to the question, “What category best describes your profession or occupation at [your primary practice site] during 2016?” Providers were reclassified as follows: physician (MD or DO); nurse professional (includes registered nurse, licensed practical nurse, certified nursing assistant, certified nurse midwife, other midwife, nurse practitioner, or self-described as a nurse professional, including nurse manager); or other health professional (includes administrator, clinical laboratory personnel, dentist, other dental professional, health educator, counselor, pharmacist, physician assistant, psychologist [PhD, PsyD, EdD], other mental health provider, social worker [LCSW, MSW, DSW, LMSW, etc], veterinarian).
Practice setting was determined based on the response to the question, “What best describes your principal practice in 2016?” Settings were categorized as follows: solo or private practice; group practice; hospital-affiliated practice (includes inpatient and outpatient); community practice (includes community health centers, clinics, urgent care, residential sites [nursing home, assisted living, hospice, long-term care], and other community-based facilities); and other (includes self-reported facility types that could not be reclassified into one of the other categories).
Among physicians (n = 1,139, 78.7%), the most commonly reported specialty was internal medicine (30.2%), and the next most commonly reported specialty was pediatrics (14.2%). Obstetrics/gynecology and family medicine were each reported by nearly 7% of physicians. Other specialties were reported less frequently (Table 2).
Clinical Specialties Reported Among Physician Respondents to a Survey Regarding Public Health Emergency Communications Distributed Electronically to Email Addresses Contained in the NYC DOHMH Provider Data Warehouse, New York City, 2016
Defined as respondents who selected the following options to the question, “In 2016, was your clinical specialty any of the following?”: allergy/immunology, anesthesiology, colorectal surgery, dermatology, genetics, neurology, nuclear medicine, occupational medicine, ophthalmology, orthopedic surgery, pathology, physical medicine, plastic surgery, preventive medicine, psychiatry, radiology, surgery (general or subspecialty), cardiac and thoracic surgery, urology, or self-described as some other specialty not listed.
Respondents were asked about their use of specific NYC DOHMH information sources and non–NYC DOHMH information sources to keep themselves, their colleagues, and their patients informed about Zika during 2016. Reported use of these sources by provider type and principal practice type are displayed in Table 3 (NYC DOHMH information sources) and Table 4 (non–NYC DOHMH sources). Overall, the NYC DOHMH Health Alert Network was the most commonly reported local health department source, reported by 73.0% of all respondents. More than 60% of respondents reported using the NYC DOHMH eCity Health Information (eCHI), even though the NYC DOHMH never released information about Zika in the eCHI. An additional 22.4% of all respondents reported they did not use any local health department source of information about Zika (Table 3). Among non-DOHMH sources of information, the CDC was most frequently cited by respondents (63.7%), while 10.6% did not use any of the alternative sources of information about Zika (Table 4).
Information Sources from NYC DOHMH Used to Stay Informed About Zika, Among Eligible Respondents, New York City, 2016
Alternative Information Sources Used by Providers to Stay Informed About Zika, New York City, 2016
Patterns of reported use of NYC DOHMH information sources appeared to vary by provider type (Table 3), as did reported use of non–NYC DOHMH information sources (Table 4). Among physicians, the NYC Health Alert Network was the most commonly cited local health department source of information (75.6%), and 59.6% of physicians reported that they used the NYC DOHMH website. Nearly 20% of physician respondents did not report use of any NYC DOHMH source of information about Zika. When reporting their use of non-DOHMH information sources, physician respondents were most likely to report that they relied on the CDC (66.9%), followed by the New York State Department of Health (46.9%), and medical journals, online, or point-of care resources (38.9%).
Among the small number of nursing professional respondents, the NYC Health Alert Network was also the most commonly cited NYC DOHMH information source (81.7%), followed by the NYC DOHMH website (77.4%) (Table 3). Among nursing professionals, 15.0% of respondents reported that they did not use any of the NYC DOHMH information sources to stay informed about Zika. Nursing professionals commonly reported using non–NYC DOHMH information sources to stay informed about Zika, most frequently citing the CDC (69.9%) and the New York State Department of Health (65.6%), while only 23.6% of nursing professional respondents reported that they used medical journals, online, or point-of-care resources (Table 4). Just over half of health professional respondents in the “other” category reported use of NYC DOHMH information sources to stay informed about Zika, with only 55.5% reporting that they used the NYC Health Alert Network (Table 3). Similarly, just under half (43.7%) of other health professionals reported using the CDC as an information source (Table 4).
Reported use of information sources also appeared to vary by principal practice setting, both for NYC DOHMH information sources (Table 3) and non–NYC DOHMH information sources (Table 4). A higher proportion of healthcare providers whose principal practice setting was private practice reported not using any NYC DOHMH source of information about Zika (28.6%), in comparison with other practice settings (group practice, 21.8%; hospital practice, 22.5%; community-based facility, 16.1%; and other type of facility, 12.5%). Use of the NYC DOHMH Health Alert Network was also least frequently cited by private practitioners (68.4%) compared with other practice settings (group practice, 71.6%; hospital practice, 72.9%; community-based facility, 80.5%; and other facility, 87.5%). Use of non–NYC DOHMH information sources also varied by principal practice type. For non–NYC DOHMH information sources, respondents working primarily in private practice settings reported use of the information source less frequently than did respondents working in hospitals settings. For example, the most commonly cited non–NYC DOHMH information source for all practice settings was the CDC. However, only 48.5% of respondents in private practice reported use of this source, compared to 69.9% of respondents practicing in hospital settings (Table 4). Among the non–NYC DOHMH information sources, only publicly available websites and general media were more frequently reported among respondents in private practice (37.8%) compared to those in hospital-affiliated practices (34.4%).
Respondent primary preferences for communication of public health alerts and guidance, by provider type and principal practice type, showed a clear and strong preference for email communications (Table 5). Overall, 76.6% of respondents selected email as their first preference for communications. The next most frequently selected communications method as the first preference was hard copy, ranked first by only 8.2% of respondents.
Preferred Method for Public Health Communications and Guidance, by Provider Type and Principal Practice Type among Eligible Respondents, New York City, 2016
Discussion
Our experience designing and conducting this survey of healthcare providers during and following a public health emergency highlights the challenges inherent in maintaining accurate, timely, comprehensive contact information for the purposes of outreach to individual providers during public health emergencies. However, our approach to the survey, using the best available email contact list for healthcare providers in New York City, did allow us to explore the mechanisms by which local public health authorities and others communicate timely provider information during public health emergencies. There are very few reports that have attempted to elucidate the reach of electronic public health messaging, though health alert systems are in common use by public health departments. Better understanding the reported use of various information sources among those providers that responded to our survey can provide important insights into how providers respond to the ways in which local, state, and federal public health authorities disseminate information. Our results also revealed the breadth of audience for public health messaging to healthcare providers. These findings can help to guide future investment in public health emergency communications to healthcare providers.
A wide range of provider types and practice settings were represented among our respondents. However, most respondents, nearly 80%, were physicians. This probably reflects the various sources of information included in the Provider Data Warehouse rather than the true universe of healthcare providers in New York City. Both ComDir and PNDS, major sources of information for the data warehouse, are likely biased toward prescribing providers, because those providers must use these systems to obtain New York State prescription pads and use other state systems. While other providers who are licensed in New York State, such as registered nurses, may be included in ComDir if provided a health commerce system role through an organizational coordinator, it is likely that those nonprescribing providers are not well represented in the system.
Further, even though DOHMH attempted to target providers who work with women of childbearing age, many types of healthcare provider respondents reported using public health information sources in response to emergence of Zika as a health threat. We also observed a large number of medical specialties among the physicians that participated in the survey, including those specialties targeted by NYC DOHMH outreach, in part through collaborations with local chapters of professional societies (obstetrics and gynecology, family medicine, pediatrics, internal medicine, infectious disease, and emergency medicine), as well as a wide variety of other specialties, including surgery, radiology, psychiatry, and others. Obstetricians and gynecologists represented 6.9% of physician respondents to our survey. National data from the American Medical Association Physician Masterfile (December 2013) indicate that obstetricians and gynecologists make up about 5% of all US physicians. 6 These national data also indicate that around 6.7% of US physicians are pediatricians, suggesting that pediatricians may also be overrepresented among our respondents (14.7% of respondents to our survey were pediatricians). However, given that we have limited information about the true number of physicians in each specialty in NYC, it is unclear if this finding is related to bias in our sampling frame, response bias, or actual differences between physician specialties in the use of public health information.
Most respondents, regardless of provider type or practice setting, reported that they used NYC DOHMH email health alerts during the local response to Zika as an emerging health threat, and that they prefer this method of communication. Though limited by the fact that this email survey may have biased the finding toward those who prefer email, this reinforces previous reports that providers prefer email communications and rely on public health authorities for information about emerging threats. A 2011 review of available literature on public health communications for providers identified 25 different communications systems described in the literature but found no methodologically rigorous evaluations of the effectiveness of those communications systems. 7 In 2016, a randomized, controlled trial comparing public health messaging to healthcare providers via email, text message, and fax suggested a slightly higher recall of messages in the email group. 8 Overall, healthcare providers enrolled in that study reported a preference for email as the mode of alert delivery. 9 Our findings support the use of email for public health emergency risk communications, since many providers reported using it and preferred email as a method of communication. We did not assess whether email was more effective than other communication methods in improving healthcare knowledge or practices related to Zika.
Examining differences in the reported use of public health and non–public health information sources by provider type has implications for the design and distribution of future public health messaging. Public health practitioners should be aware of the wide range of health professionals that make up the audience for risk communications. For example, among our respondents, nurses more frequently reported using nearly all NYC DOHMH sources, and more frequently reported using information from CDC and New York State Department of Health, in comparison to physicians. However, these findings would need to be validated in further studies, since nursing professionals made up less than 10% of our respondents. In addition, our data source for provider contact information is likely biased toward nursing leadership or those particularly interested in public health communications. In our study, nearly 15% of respondents were other health professionals. This very heterogeneous group less frequently reported use of most public health information sources. It is unclear if these providers were not seeking information about Zika because they felt it was not relevant to their practice, or because much of the healthcare provider messaging developed by public health agencies was clearly directed to physicians of particular specialties and nursing professionals. In future public health emergencies, developing messaging targeted to other health professionals, if relevant to their areas of practice, could be an important communication strategy.
The finding that more than 60% of respondents reported using the NYC DOHMH electronic City Health Information (eCHI) during the Zika response is curious. An eCHI dedicated to the topic of Zika virus infections was never released during the response. The eCHI is a clinical bulletin produced by DOHMH about 10 times per year and offers timely, evidence-based information of relevance to practicing clinicians. Typically, the timeline for producing an eCHI bulletin requires many months; for this reason, it is generally not used for emergency purposes. However, given the apparent popularity of the eCHI format among our respondents, it is worthwhile to consider producing similar clinically focused brief, evidence-based bulletins relevant to emerging threats.
Healthcare organizations, such as hospital systems, federally qualified healthcare networks, and healthcare associations, among others, may be important partners for public health in disseminating emergency messages to their staff, including healthcare providers. Some healthcare organizations in New York City have described incorporating the local health department alerts (NYC DOHMH HAN) into their electronic health record systems as a way to improve the implementation of new guidelines into clinical practice. 10 In our analysis, healthcare providers in private practice more frequently reported that they did not use any information source to stay informed about Zika. Meanwhile, respondents who practiced in hospital-based and community-based facilities more frequently reported that they used information from the NYC Health Alert Network, the CDC, and the New York State Department of Health. This highlights the difficulty of reaching individual providers in the community with public health emergency messaging, since our findings suggest that those providers less connected to healthcare systems are using public health information sources less frequently.
Our results also indicate that healthcare providers use multiple sources of information during health emergencies. Each different communication source can be an opportunity for contradictory messages, if public health agencies at the local, state, and federal levels do not coordinate risk communication messaging and strategies. Over a third of respondents, regardless of provider type or practice setting, reported using publicly available sources of information. These sources were reported as frequently as were medical journals and online and point-of-care sources, and more frequently than information distributed through professional societies, yet public health officials have limited control over the information being distributed through the general media.
This study has a number of limitations. We used NYC's Provider Data Warehouse as the source for our email distribution of this survey. Email addresses once included in the warehouse are never deleted. While this ensures the widest possible reach for public health communications, it is likely that many of the more than 44,000 email addresses included in our sample are no longer in use, or belong to providers who may have retired or moved outside of NYC, making NYC communications less relevant to them. While we were able to eliminate duplicate email addresses, we were not able to exclude the possibility that some providers might have different, valid emails from different sources included in the Provider Data Warehouse. We observed that nearly 20% of the emails we sent a survey link to were undeliverable. In fact, some email addresses appeared to be attached to organizations or other entities, rather than individual people, and it is unclear if survey links sent to those addresses actually reached anyone, let alone a healthcare provider.
A small proportion of the total number of individual survey links were completed, yielding a low response rate. Using a more accurate source of healthcare provider contact information could therefore have resulted in a much higher response rate, and the quality and comprehensiveness of the email data sources could have systematically biased the results presented. Given that part of the DOHMH response to Zika included collaboration with local chapters of professional societies representing obstetricians and gynecologists, family practice physicians, and pediatricians, it is possible that distribution of the survey or reminders through those mechanisms could have resulted in a higher response rate, at least among those specialties. However, since societies are unable to share contact information for the members with external entities, this would have reduced our ability to monitor the number of email addresses that the survey was sent to.
Nevertheless, it is likely that those who completed the survey are those most engaged with local public health or those particularly interested in Zika virus disease. Our results indicate a bias toward prescribing providers, such as physicians, and it appears that healthcare providers who do not need to prescribe and do not frequently use the data systems that feed the Provider Data Warehouse are poorly represented among our respondents. Therefore, our findings may not be generalizable to a larger population of NYC healthcare providers. However, identifying communications preferences and use patterns among this group is still valuable for informing future response to public health emergencies.
In this broadly distributed survey of NYC healthcare providers, we identified multiple sources of information that healthcare providers rely on during public health emergencies. Although DOHMH sources were perceived as useful, other sources were also cited, highlighting the importance of aligning public health messages for providers across public health agencies. Further research should examine how provider knowledge, attitudes, and behaviors are influenced by public health messaging related to emerging public health threats and how this differs by the information source and medium of dissemination. Effective strategies to identify and engage healthcare providers who are practicing independently, rather than in community-based or hospital-affiliated settings, are needed to improve communications with all healthcare providers during public health emergencies. Public health authorities, recognizing the importance of timely and accurate risk communications to providers at the time of emergencies, should invest in thoughtful development of provider databases that could allow for maintenance of accurate contact information and targeted messaging.