On the front line: Health professionals and system preparedness for Zika virus in Peru

Abstract Objectives To analyze the initial healthcare response to the Zika virus in Piura, Peru, and assess the perceptions of midwives and nurses regarding their role in prevention of Zika virus and management of congenital Zika syndrome (CZS). Methods This ethnographic study used a rapid qualitative assessment design. Data were collected through a focus group with midwives and in‐depth interviews with midwives (n=11) and nurses (n=5). Results The focus of the early Zika virus response in Piura was on pregnant women and vector control. Midwives received some training on Zika‐related care during the early response. Nurses did not receive any Zika‐specific training. Neither nurses nor midwives were trained in neonatal CZS surveillance. Midwives were clear about the value and feasibility of incorporating Zika virus surveillance in their daily work, however nurses were not. They referred to lack of training and appropriate tools as limitations. Confusion about Zika virus and CZS symptomatology and effects persisted in both groups. Concerns about their own personal risk influenced the ways they engaged with Zika virus prevention in the community. Conclusion Long‐term management of endemic Zika virus in Piura will require the engagement of both nurses and midwives as primary care providers.


| INTRODUCTION
Health providers are the first line of care and surveillance for the long-term management of any endemic disease, yet little is known about how they process and view their roles in cases of emergent and complex diseases. Zika virus is one such disease. Primarily transmitted through the bite of an infected Aedes aegypti mosquito, it is also transmitted through sexual intercourse, from mother to child, and through blood transfusion. Zika virus infection during pregnancy has been linked to stillbirths and to congenital Zika syndrome (CZS) characterized by varied levels of cognitive impairment, microcephaly, and a host of vision, hearing, and motor impairments. Additionally, Zika virus infection in nonpregnant women has been linked to an increased incidence of Guillain-Barré syndrome. 1 The characteristics of Zika virus infection present new challenges for the public health response and long-term prevention. A large number of asymptomatic cases, 2 the persistence of the virus in semen, 3,4 and the preference of the mosquito for living inside homes mean a higher risk of infection in low-income urban communities and especially among vulnerable women and children. 5 Engaging health professionals effectively in a coordinated response is key to managing Zika virus in its new endemic form.
Between 2016 and 2017 around 6000 cases of Zika virus were reported in Peru, with higher concentrations in the regions of Loreto, Piura, Ica, and Cajamarca. 6 The number of suspected cases was much higher than the final confirmed cases; however, diagnosis was hampered by the need to send blood samples to Lima for confirmation.
While a cluster of Guillain Barré cases in 2018 was suspected to be linked to Zika virus, there have been no large-scale reports of adverse Zika virus-related effects in Peru. However, since surveillance for microcephaly and other fetal abnormalities is limited, it is likely that cases of CZS are underreported.
In Piura, there were around 300 confirmed cases of Zika virus infection between 2017 and 2018, and 88 so far in 2019. 6 Zika surveillance was affected by flooding due to heavy rains in February 2017, which damaged rural health posts and overwhelmed the emergency response. After the flooding a dengue epidemic swept through the city, infecting at least 44 000 7 and killing 41 people. 8 In the wake of these overlapping emergencies, Zika virus was low on the priorities of the regional surveillance apparatus; thus, it is difficult to ascertain the real effects of the virus. However, it is expected that Zika will follow a pattern of periodic activity similar to dengue, making it a future priority for local public health officials.
The aim of the present study was to analyze the initial healthcare response to the Zika virus epidemic in Piura and assess the perceptions of frontline personnel-midwives and nurses-regarding their role in prevention of Zika virus and management of congenital Zika syndrome (CZS).

| MATERIALS AND METHODS
The data reported here were part of a larger study that included focus groups and interviews with local men and women, and health providers. The present article reports only health provider data collected in two phases: phase 1, August-September 2017; and phase 2, January-February 2018.
This ethnographic study used a rapid qualitative inquiry design, featuring intense but shorter data collection than traditional anthropological ethnography. 9 The study site was chosen in collaboration with the Piura Regional Health Directorate.
The location of the study, the Catacaos Micro-Network, is part of the Lower Piura Health Network. The micro-network is comprised of an urban clinic that includes all primary care, pregnancy, birth care, and surgery facilities, and six dependent health centers, with lower levels of care, located in outlying and/or rural areas. Names of participants and their health centers have been omitted in accordance with Institutional Review Board recommendations.
A summary flowchart of the study is given in Figure 1. The original study design only included midwives as participants. During the focus group phase, it became apparent that nurses were also important to meet the objectives of the study and they were subsequently recruited for in-depth interviews. Data were collected through a focus group with midwives (n=6), and in-depth interviews with midwives (n=11) and nurses (n=5). The focus group was a convenience sample of Catacaos midwives. Participants were invited in person. In-depth interviews were recruited among focus group participants who said they were interested, and by referral, asking each F I G U R E 1 Study data collection and community engagement flowchart. Community Engaged Research: Interviews with Regional Health Director and Chief Epidemiologist to agree on shared interest and define suitable study area.

Choose Research Site:
Catacaos Micro-Network in Lower Piura Mixture of urban and rural health establishments.

Focus group (FG) with Midwives
Main Clinic in Catacaos (convenience sample -6) Focus: Contextual informaƟon of early health response to Zika Used to design interviews.

In-depth interviews with midwives
StarƟng with FG parƟcipants and then by referral sampling (11).
Focus: Congenital Zika syndrome awareness. Experiences of ZIKV prevenƟon. Opinions on limitaƟons and needs. The authors coded independently, and coding structure was discussed and harmonized at two points during the process. Codes were organized into categories in response to research questions.

In-depth interview with Nurses
Major themes in each category were derived from the data and represent shared experiences (Fig. 2). Transcripts were not returned to participants, and no interviews were repeated. All participants were invited to a workshop where they provided feedback on preliminary results.
The study was approved by the Pan-American Health Organization

Ethics Review Board and the Ethics Committee of Peruana Cayetano
Heredia University in Lima, Peru. The protocols and informed consent forms were shared with regional partners and assessed for cultural concordance by local research team members. However, at rural facilities, often one nurse or midwife was in-charge of providing all care to women and children.

| Experiences of initial response to Zika prevention in Piura
The increased risk posed by Zika during pregnancy meant that the initial response involved midwives directly, but not nurses.

| Confusion and misunderstanding about Zika virus transmission and symptoms
Interviewees knew that Zika virus is transmitted by the Aedes aegypti mosquito, like dengue. However, it was frequently confused with dengue. For example, M6 reported: Zika is another type of dengue, more aggressive and what's worse it attacks women in the first trimester and produces microcephaly.

Midwife 6
Similarly, M7 said that the symptoms are:

Midwife 10
In this case and that of Nurse 1, health personnel expressed exasperation and confusion as to their current role in Zika virus care.
The lack of an efficient and comprehensive referral and counterreferral system is perceived as a barrier to effectively engage in Zika-related care.

| Perceptions of personal Zika virus risk is low
One key issue is the relationship of health personnel with the community and their own risk awareness. Most midwives and nurses live near where they work, they are exposed to the same risks and share

Midwife 2
Overall, perceptions of risk among participants demonstrated little concern for Zika virus infection, though general concern for mosquitoborne diseases, and dengue specifically, was very high.

Midwife 7
This midwife echoed the need to protect pregnancies, and interestingly also conflates dengue and Zika.
Midwives recognize the need to extend prevention messages to nonpregnant women. However, many of these initiatives were bundled into the educational outreach related to the dengue outbreak, and they focused mostly on pregnancy care, and vector control: we tell them about the symptoms and such, and to go to emergency care at the health center.

Nurse 3
Nurses see more of a role for themselves in CZS surveillance than in working with pregnant women.

| Perception that Zika virus prevention should include outreach and community work
Both nurses and midwives considered that community work should be a part of future Zika virus prevention. They are already engaged in what they call "extramural" work, especially in rural and peri-urban areas where access to health care is limited:

| Perceived role in congenital Zika syndrome surveillance
As explained in the referral system section, the lack of counter-referral protocols limits the ability of midwives and nurses to follow-up sus- have not been consistently targeted either. This reflects the limitation of the primary health system to reach women of reproductive age before they are pregnant and access prenatal visits. This also reflects the political reticence around openly discussing and fulfilling women's sexual and reproductive rights and needs, including offering culturally acceptable contraception. 21 The siloed organization of health programs and targeted populations has also been challenged during the Zika virus epidemic. The few suspected cases of newborns with anomalies required the exchange of medical information between nurses, midwives, and doctors at the primary care level, prior to their referral for confirmation, and between different levels of care, which was cumbersome, slow, and inconsistent. Lack of a counter-referral system in place poses a challenge to children's health and CZS surveillance at primary care level. 22 The occurrence of CZS also revealed the conditions and limitations in which standard child growth and development services are provided, and the challenges they present to identify and care for the wider spec- The present study is novel in its approach to trying to understand how health providers view themselves within Zika virus related care.
However, it has some intrinsic limitations. Firstly, flooding and a dengue epidemic disrupted data collection and moved all local resources toward dengue prevention. This may have influenced the conflation between dengue and Zika virus symptoms in the area. Additionally, this was a qualitative, small sample study that may inform future research and questions but is not meant to be generalized.
In conclusion, in an endemic scenario of long-term surveillance and management of Zika virus and CZS, diagnostic tools, training, and surveillance resources need to be increased to be effective. A cost-effective and culturally competent approach should likely include a broader focus on family planning and child development surveillance. This supposes cross-disciplinary collaboration between nurses and midwives. The present findings are important for health policymakers who have to design training, surveillance, and referral processes at regional and local levels.

AUTHORS CONTRIBUTIONS
RI and LG conceived and designed the study, conducted data analysis, and drafted the manuscript. RI conducted the interviews and fieldwork. All authors contributed to the revision of the article and approved the final version of the manuscript.