Background
Within the next two decades, the number of households led by someone aged seventy-five years of age or older will double to 28.2 million [
1]. Notwithstanding the fact that the majority of these individuals present with multiple co-morbidities and decreased activities of daily living [
2], these individuals want to age in their homes and communities [
3,
4]. Further, as in-home medical equipment evolves, there is a growing percentage of the medically vulnerable population who can receive care at home rather than in a healthcare facility. For example, more than 2.5 million Medicare beneficiaries rely on electricity-dependent medical equipment, such as ventilators, to live independently in their homes [
5]. Concomitantly, there is an increasing frequency of large-scale natural disasters [
6]. These disasters disproportionally affect both the medically complex [
7] and the older old [
8,
9]. Consequently, it is incumbent on communities to mitigate the impact of disasters on community-dwelling, medically vulnerable older adults.
Home health agencies (HHAs) provide supportive healthcare services to community-dwelling, medically vulnerable older adults. Medicare certified HHAs provide skilled, multi-disciplinary healthcare in the home, and the number of HHAs across the United States is growing [
10]. Accordingly, the new Centers for Medicare and Medicaid Services (CMS) Emergency Preparedness Rule included HHAs among the 17 provider and supplier types that are required to meet national emergency preparedness standards for both natural and manmade disasters [
11]. Yet, there is limited understanding of the actual efforts required for these programs to implement a preparedness, response, and recovery program in an actual disaster event.
Home Based Primary Care is a subset of home health care. The CMS Independence at Home (IAH) demonstration has supported a growing interest in understanding the benefits of expanding this model of care [
12‐
14]. The Veterans Health Administration (VA) Home Based Primary Care (HBPC) Program, which has a similar model to the IAH demonstration [
14,
15], and has been in practice for more than 40 years, serves a population with a mean age of 76.5 years. Like a traditional HHA, VA HBPC provides interdisciplinary care in the home to select veterans who present with complex chronic disease [
16]. To understand the degree of support required for home-based care programs to support their staff and patients through a natural disaster, we conducted a mixed methods study of the VA’s HBPC programs impacted by the Fall 2017 Hurricanes (Hurricane Harvey (August 2017), Hurricanes Irma and Maria (September 2017).
Methods
This study was part of a larger study that aimed to understand the role of VA HBPC during the Fall 2017 Hurricane season [
17]. To obtain a multifaceted understanding of support provided by HBPC programs to support staff and their patients through a natural disaster, we applied a convergent mixed methods design, whereby we incorporated independently conducted qualitative and quantitative analyses. Using a mixed-methods design allowed the exploration of both qualitative insights of program staff directly involved in the hurricane preparedness and response, as well as a quantitative illustration of the timeline of activities around the hurricanes at the same sites. Phase One of this study used qualitative interviews to examine the experiences of nine VA HBPC programs in their responses to Hurricanes Harvey, Irma, and Maria (Fall 2017 Hurricanes). Phase Two of this study used the VA’s Corporate Data Warehouse (CDW) to quantitatively examine HBPC data for these same nine sites.
Qualitative data
Study design
The first part of the study used qualitative interviews with key stakeholders from HBPC programs located in regions impacted by Hurricanes Harvey, Irma, and Maria (Fall 2017 Hurricanes) in order to explore their experiences as part of their program’s Hurricane response.
Setting
We studied VHA HBPC programs impacted by the Fall 2017 Hurricanes. The HBPC program is composed of an interdisciplinary team. The clinical staff are generally composed of nurse practitioners, registered nurses, social workers, occupational therapists, and dieticians, as well as physicians, physical therapists, and psychologists. Each VA Medical Center (VAMC) has an HBPC program. The HBPC programs can also be located in an affiliated Community Based Outpatient Clinic (CBOC), where the HBPC clinical staff are based at the CBOC, but they share a Program Director with the main VAMC. At eight of the nine sites the main VA Medical Center (VAMC) was impacted by a hurricane. At three sites, an affiliated CBOC was more seriously impacted than the primary facility, while at the other six sites, both the main VAMC and multiple affiliated CBOCs were impacted.
Sample
Nine VA HBPC programs were included from Texas, Florida and Puerto Rico. We used a purposive sampling approach, with the HBPC Program Director as the first point of contact at each site. These individuals were then asked to identify additional HBPC team members who were on staff at the time of the Hurricanes. Study staff then contacted these additional team members by email and phone. All respondents were reminded that study participation was voluntary. We conducted 34 interviews. Due to minimal impact by a hurricane, it was determined that saturation was reached at one site after speaking to a single respondent. All other sites had three to five respondents.
Design and data collection methods
We developed semi-structured interview guides (see Supplementary file
1) that queried respondents on three key themes: hurricane preparedness protocols and activities, hurricane response activities, with a focus on continuity of care, and facilitators and barriers to disaster response. Respondents were asked to describe their role in the HBPC program and any past experiences with disasters. Participants were interviewed separately by phone for 30 to 60 min, and interviews were audiotaped and transcribed. At one site, a focus group with clinical staff was conducted due to scheduling conflicts. In this case, all participants agreed to the joint sharing of experiences.
Analysis plan
All interview data were uploaded into Atlas.ti (v.7) for analysis. The initial code list was composed of an a priori code list, which was then revised an expanded, based on in-vivo coding methods [
18]. All interviews were coded independently by authors XX and XX., All disagreements were resolved by consensus.
Quantitative data
Using data from the VA CDW, a national repository of clinical and administrative data from VA medical facilities, information about each clinical visit was extracted for the three most impacted VAMCs: Houston, Tampa, and San Juan.
The initial study cohort for the Houston VAMC included VA healthcare-users who had accessed the VAMC at least once in the 24 months prior to Hurricane Harvey. Similarly, for the other two VAMCs (Tampa and San Juan), the initial study cohort included VA-users who had accessed their respective VAMC at least once in the 24 months prior to Hurricanes Irma and Maria.
For the Houston and Tampa VAMCs, the total number of daily HBPC services and telephone HBPC visits were examined 14-days before and 14-days after each hurricane. Given that Puerto Rico experienced extensive damages from Irma and Maria, for the San Juan VAMC, the total number of daily HBPC services and telephone HBPC visits were examined 30-days before and 30-days after each hurricane. The percentage of HBPC patients with one, two, three, and four or more HBPC visits were also examined two-weeks before, one-week before, one-week after and two-weeks after each hurricane.
Discussion
From 2015 to 2030, the percent of individuals age 65 and older worldwide will increase from 9 to 12% [
19]. In the United States alone, the number of people 85 years and older is expected to nearly double from 2016 to 2035 [
20]. With an increased ability to provide health care to individuals in their home, more older individuals are now choosing to age in their homes. And although there is an increasing evidence base on the value of providing healthcare to patients in their homes [
21,
22], there is limited research about how to support these individuals during a disaster. The role of home-based health care programs as critical components in supporting the older old during disasters is becoming clearer [
23‐
25]. However, few studies have examined the activities required by these programs to support a real disaster [
26]. Accordingly, we sought to explore the activities conducted in advance of, and in response to, a natural disaster (in this case, hurricanes) and determine the resources required to support such activities. We found that VA HBPC program staff undertook extensive preparedness actions and quickly resumed delivering patient care even in the most impacted regions.
The disaster resilience of place (DROP) model proposes that the total disaster impact on a community is defined through a combination of antecedent conditions [
27]. SL Cutter, L Barnes, M Berry, C Burton, E Evans, E Tate and J Webb [
26] enumerate several ways in which a community can respond to a disaster event, including predetermined evacuation plans, creation of shelters, information dissemination, and emergency response plans. The overall local impact of a disaster can be moderated by the ability of the community to absorb event impacts using predetermined coping responses [
27].
The multiple preparedness phases implemented by the VA’s HBPC programs demonstrate the efforts required to invest in the types of antecedent conditions that enhance a community’s absorptive capacity. Prior research has shown that the older old and their caregivers are often isolated from their communities [
28,
29] and unprepared for disasters [
8]. By talking with patients about emergency plans, providing information and support for shelter registration, and heightening medication management during the hurricane season, HBPC program staff bridge this essential gap between patients and their communities. This early groundwork allows the HBPC programs to call each patient and support execution of the pre-established emergency plan. Without the initial groundwork, the management of an entire census of patients would be much more challenging.
Moreover, these three phases of preparedness set up the possibility of a rapid response if required. As shown by SA Bell, J Horowitz and T Iwashyna [
24], the response period was quite challenging for most HHAs after Hurricane Harvey. In contrast, we found minimal delays in following up with patients after the hurricane events. In cases where patients had evacuated, staff reported on the status of their patients’ homes, and even in Puerto Rico, the hardest hit region, our data show program staff beginning to conduct well-visits within 3 days of landfall. These findings exemplify the vital role of the HBPC program in bolstering community resilience.
Past disasters have demonstrated that the majority of medical surge arises from individuals with one or more chronic conditions or dependency on electric-dependent equipment [
7,
9,
30‐
32]. Understanding ways to reduce this medical surge in the days after a disaster is an essential piece to bolstering a region’s disaster resilience [
33]. A well-visit conducted by skilled clinicians who are familiar with their patients’ care, can provide essential support to medically vulnerable individuals and potentially mitigate some of this surge. Indeed, our study showed that those patients who prior to the hurricane were only being seen once a week, did not require additional visits. In addition, both the qualitative and quantitative data demonstrate that the temporary reduction in care for higher frequency patients did not result in any critical disturbance to patients’ medical status in the immediate post-hurricane period. These findings highlight the significance of intense preparedness and a structured response.
As our cities grow, and the share of residents aged 60 and older steadily increases, more resources need to be dedicated to understanding how to support active aging and aging in place. Specifically, communities need to expand their conversation about how to better support active aging by optimizing opportunities for health and security [
34,
35]. One way in which disasters exacerbate existing deficiencies is by stressing existing infrastructure. The vulnerability of older adults after disasters is in part related to a lack of existing social support before a disaster event. As has been noted in our results, home-based care programs can bolster socially isolated, homebound individuals support structures, yet they cannot do it alone. As has been noted in the literature, although the larger field of HHAs have an important role to play in community disaster response [
23], in practice, few agencies are part of local healthcare coalitions, a key center of local disaster response [
24]. Understanding HHA preparedness and response activities within the context of an actual disaster can help healthcare coalitions understand the need to and potential benefits of incorporating HHAs into such coalitions and their response strategies, and by extension provide increased attention and support to our growing population of older adults.
Limitations
This study has limitations. First, the data collection and design of the qualitative and quantitative components were conducted independently. Future studies should incorporate a more integrated mixed methods approach where one method (qualitative or quantitative) guides the other. Second, the findings from the interviews cannot be generalized to all VA HBPC programs. Third, quantitative data were reported by staff via electronic health records. Future studies should incorporate data from disaster-impacted patients. Finally, the VA HBPC program is a singular example of home-based care programs and has the support of the larger VA healthcare system during emergencies. Nevertheless, VA HBPC programs’ activities are examples of best practices. Future studies should examine whether (a) programs located in disaster-prone regions are more prepared than others, (b) how to ensure adequate preparedness in regions that are less frequently impacted by disasters, and (c) how non-VA home-based care programs support their patients and staff during emergencies.
Conclusions
Historically, caring for the older old has been a challenge in the aftermath of disasters due to the limited presence of pre-established coping mechanisms. And as our population ages, and aging in place becomes more common, more attention needs to be placed on how to support our older old, particularly those with medical vulnerabilities, around disasters. Home based care programs such as HHAs and the VA’s HBPC program, which already have strong relationships with their patients and help connect homebound, medically complex, older old to the greater healthcare community, can play an essential role in establishing and fulfilling these coping mechanisms. Engaging with these programs both pre- and post-disasters is central to bolstering community resilience for these at-risk populations.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.