Immunization barriers and facilitators

Barriers to and facilitators of immunization exist on both the supply-side (which includes logistics, such as cold chain management) and the demand-side (which includes access to and affordability of vaccines and psychosocial factors like fear and acceptance) [8]. On the demand-side, several health service-related behaviors of parents and caregivers are associated with greater immunization uptake, including retaining the child’s immunization card [9–11], obtaining services at privately-owned facilities [10], and giving birth in an institutional setting [9, 12]. Institutional delivery is particularly correlated with immunization coverage, as health workers can more readily administer the first scheduled vaccine (BCG) immediately after birth than in a home setting [9, 12]. Additionally, giving birth in a medical facility may offer parents an additional opportunity to learn about the benefits of vaccines.

Awareness of the vaccine schedule is a crucial aspect of parent knowledge [10, 11, 13]. Nepal requires seven separate visits over 15 months to complete the full schedule [12]. As such, parents who are less familiar with this schedule are at risk for forgetting or delaying vaccine appointments [9].

Immunization may be hindered by several factors, including parental fear and health system barriers. Several studies suggest that parental fear could be contributing to low immunization rates by creating poor perceptions of vaccines [11–15]. Both in the US and globally, parents have reported considerable distrust in vaccines, which has created a pattern of vaccine hesitancy [16, 17]. Some studies have found that parents may view their children as being “too ill” to receive a vaccine [12, 14] or fear that the side effects will outweigh potential benefits [11, 13, 15]. Indeed, some vaccine-preventable diseases have re-emerged in both developed and developing countries due to parental fear.

On the health system side, factors such as procurement and supply of vaccines, as well as provider competency and communication, have been found to influence vaccine coverage. For example, Rainey et al. [18] revealed in their systematic review that the most common reason associated with under-vaccination is the immunization system. More specifically, issues of inadequate access to services, long distance to immunization centers, poor health worker knowledge and attendance, and vaccine availability are strongly linked with under-vaccination in low- and middle-income countries [18]. Favin et al. [15] also reported a concerning pattern–that mothers are often treated in a disrespectful and, in some cases, abusive manner by health workers at immunization appointments–which discouraged them from future visits and left them with unanswered questions about their children’s health.

Study objectives

This study presents findings from the formative assessment of the Rejoice Architecture Project (see Paul et al. [19] for the study protocol), a social norms-based intervention aiming to improve child vaccination in the Makwanpur District in Nepal. To do so, the project addresses the physical, structural, and social barriers within health facilities at the nexus of both demand- and supply-sides; namely, between maternal caregivers and health workers. In this study, we envisioned the healthcare clinic as the site that straddles both the demand- and supply-side issues that could result in reduced visits and, by extension, reductions in immunizations. In particular, we sought to understand the demand- and supply-side barriers to and facilitators of immunization in Makwanpur District, Nepal.

Study setting

Administratively, Nepal is divided into seven provinces, each of which is further divided into districts (77 in total), then municipalities (called a “Palika” in the local vernacular, which is also the term we use in this paper) and, finally, wards. This study was conducted in three Palikas (Thaha, Kailash, and Bakaiya) in the Makwanpur District. Makwanpur has a population of approximately 460,000, of which 4% are between 0 to 23 months of age and 22% are married women of reproductive age [20]. Makwanpur is considered a Category 1 District in terms of immunization access (DPT-HepB-Hib1 coverage) and utilization (DPT-HepB-Hib1 vs DP-HepB-Hib3 dropout), meaning it has approximately 80% coverage and less than 10% dropout [21]. As of 2020, the percentage of children fully immunized as per NIP schedule in Makwanpur was 68%, comparable to the national rate of 65% [3].

Makwanpur was selected for this study for its representation of Nepal’s ethnic and geographic diversity and large size. For example, Makwanpur represents 78 of Nepal’s ethnic groups, inclusive of various cultures and languages [22]. In addition, the three selected Palikas were chosen to represent Nepal’s three regions of the mountains (Thaha Municipality), hills (Bakaiya Rural Municipality) and terai (lowlands; Kailash Rural Municipality). Lastly, Makwanpur has a land area of approximately 2,426km², about 1.6% of Nepal’s total land area [23].

Participants

Seventy-six individuals participated in this study across the catchment areas (1-kilometer radius) of five health facilities. All participants were 18 years or older and were either permanent residents of the health facility’s catchment area or worked for, oversaw, or volunteered at the health facility. Ten in-depth interviews were conducted with five fathers and five grandmothers of children under the age of 2 (see Table 1). Twelve key informant interviews were conducted with four female community health volunteers (FCHVs); four health workers, the majority of which were in a leadership role (i.e., Health In-Charge, Immunization Program Manager, etc.); and four representatives of the local government (i.e., Ward Chair, Chair of the Health Facility Operation Management Committee, etc.; see Table 1). For mothers, who are the target population of the Rejoice Architecture Project, we were particularly interested in understanding the ways in which they communicate with each other about immunizations; and so, focus groups were conducted. Fifty-four mothers with children under the age of 2 participated in eight focus group discussions that ranged in size from five to nine participants (see Table 1). For two of the eight focus groups, only individuals from marginalized castes were sampled to gain a more comprehensive sense of the community’s wide array of experiences.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Description of the data collection modalities.

https://doi.org/10.1371/journal.pone.0261905.t001

According to Nepal’s National Immunization Schedule [7], all vaccines are recommended to be administered by 15 months. Therefore, caregiving to a child under 2 years was selected as inclusion criterion for in-depth interviews and focus groups to include families at all stages of the immunization schedule. Participants were excluded if a family member had previously participated in one of the interviews or focus groups.

Recruitment and consent

In Nepal, FCHVs are the frontline health workers at the community level. Two FCHVs from each ward assisted data collectors in recruiting potential participants for in-depth interviews and focus groups. Health facilities maintain an updated list of mothers with newborn children in the community. Using these lists, FCHVs identified mothers, fathers, and grandmothers of children under the age of 2 for data collectors to recruit in-person at their homes. Data collectors used convenience sampling for in-depth interviews and focus groups, using a quota system to guide their efforts. Key informants were purposively selected for having substantial involvement in immunization planning and procedures and were identified through discussions with community leaders. Key informants were recruited at their place of work by data collectors.

Informed consent was obtained from all participants immediately before data collection. Data collectors provided each participant with an information sheet and consent form, read the forms aloud to the participant, received written consent to participate, and provided the participant with an unsigned copy of the consent form. In-depth interviews with fathers and grandmothers took place in a private area in or nearby the home. Interviews with key informants were conducted in the informant’s office or an empty room at their place of work. Mothers participating in focus group discussions were provided a date, time, and location for the discussion after they were recruited, often in a community center or central outdoor area.

Data collection procedure

Data were collected in-person from January to March of 2020. Interviews and focus groups were facilitated by male and female data collectors overseen by the second, third, and fourth authors. All data collectors completed an in-person training led by the Principal Investigator which covered techniques for facilitating interviews and focus groups, probing, and conducting ethical research. Before administering the interview or focus group, data collectors identified themselves as researchers from Kathmandu interested in learning more about child immunization in Makwanpur District. To build rapport with participants and facilitate a comfortable space for sharing, data collectors began each session with two broad questions about the participant or their community. Interviews and focus groups lasted approximately 60–90 minutes and were audio-recorded. During all interviews and focus group discussions, two data collectors were in attendance; one to facilitate the session, the other to take detailed notes. The data collector that was not facilitating the conversation recorded field notes to document their perceptions of the participants’ attitudes and willingness to engage, as well as the surroundings, interruptions, or other notable characteristics of the session.

In one case, a key informant did not consent to audio-recording, and so, field notes were used to document this participant’s responses. During two focus group discussions, one mother in each group left the discussion early to tend to her crying child. Aside from these instances, no participants dropped out or refused to participate. No repeat sessions were conducted.

Audio-recordings of interviews and focus group discussions were translated from Nepali to English, and transcribed verbatim by native Nepali-speaking team members (see Paul et al. [19] for data management procedures). Through this process, conducted in phases as the data were collected, the researchers observed data saturation by the time data were collected in the fifth catchment area. Although data collection was initially planned for a sixth catchment area, no additional data were collected. Additionally, participants were not asked to comment on or correct transcripts, as native Nepali study team members with extensive experience working with communities provided context and clarity.

Instruments

Specific topics covered in each interview and focus group varied based on the instrument in use and the flow of the conversation. Each instrument was designed to assess topics of immunization facilitators, barriers, and social norms (see Paul [24] for the instruments used). In addition, the in-depth interview guide covered items related to the family’s experiences with vaccination and family involvement in child health and decision-making. The key informant interview guide included items related to immunization coverage in the ward, typical communication between caregivers and health care providers, the physical environment of the local health facility, and opinions of their workload. For the focus group guide, items regarding characteristics of an ideal health facility, a description of their local health facility, and typical communication between caregivers and health care providers were also included. All instruments were designed collaboratively by study team members from both the US and Nepal to assure cultural appropriateness. Instruments were translated from English to Nepali, pre-tested in a non-study site in Makwanpur, and revised for comprehension and cultural congruence.

Data analysis

Four study team members analyzed the data using an iterative, grounded theory approach informed by Glaser [25] to uncover emergent themes. Each researcher held, or was working toward, a relevant graduate-level degree and had experience leading or analyzing qualitative research related to maternal health, infant and child health, health services, or other health topics in developing settings.

The researchers began by independently reading through the transcripts to gain familiarity with the data, then met to identify prominent patterns and develop the initial codebook. Researcher triangulation was then used to ensure the credibility of analyses. Two researchers were assigned to each transcript for coding. Researcher assignments rotated on a weekly schedule to ensure researchers were working within different pairs throughout the analysis process. Each researcher then coded their assigned transcripts independently, using NVivo (released in March 2020). Research pairs met at least once per week to compare codes, discuss discrepancies, and reach a consensus for each transcript. The full analysis team also met weekly to revise the codebook. After all transcripts were coded, the team reviewed previously coded transcripts to align with the final codebook.

Following coding, prominent themes were identified by comparing codes across the various types of interviews; running word queries; and creating visual conceptual maps to link patterns. Researchers thematized both the demand-side (mothers, fathers, grandmothers) and supply-side (FCHVs, health workers, Palika representatives) facilitators and barriers of immunization as well as the connection between the two.

Demand-side facilitators of and barriers to immunization

Supply-side facilitators of and barriers to immunization

Lack of mutual trust

Throughout the data, themes of mistrust and lack of mutual respect between health workers and clients were pervasive and appeared to partially explain the context of existing service gaps, including those related to immunization uptake and continuation. There appeared to be a bi-directional relationship in terms of trust; one direction in which, in the eyes of the clients, mistrust is generated by health workers towards them, and the other in which, in the eyes of the health workers, mistrust is generated by clients, particularly those from marginalized and disadvantaged groups. Lack of mutual trust was presented through four key illustrations, including 1) micro-aggressions, 2) power imbalance, 3) poor quality of care, and 4) client non-compliance and provider bias.

Positive service bias

Perhaps a key finding from this research is our observation of how certain segments of the population are being left behind because of the pressures exerted in the system to reach 100 percent compliance. Ironically, the government’s priority for full vaccination has pushed some clients out of the health service delivery chain for two primary reasons. First, various researchers have noted that clients who miss vaccinations are likely to be members of the most vulnerable and marginalized groups [27–29]. They face significant distance and travel challenges and often cannot allocate time for travel, mostly because of financial constraints and the need to be continuously engaged in income-generating tasks. Members of these groups are likely to miss vaccines, despite their better knowledge and desires. Once they miss a few vaccines, they become fearful to visit the health facility to catch up with missed regimens due to fears of how they will be treated by the providers.

The second part of this equation pertains to our earlier point about providers’ zeal for full vaccination coverage that often drives them to treat non-complying caregivers with disrespect and even contempt. Indeed, we saw this play out in how providers discussed those who did not comply. This kind of behavior was also prominently noticed by caregivers, who related to us the maltreatment they received from providers.

This fear of being ostracized, and being implicitly labeled as an irresponsible parent, made it more likely that they would discontinue their vaccinations altogether. This is what we term a positive service bias, the idea that providers, acting with good intentions to maximize vaccination rates in their communities, inadvertently drive caregivers away when caregivers have been occasionally non-compliant. Positive service bias may be conducted through social interactions (i.e., ignoring or speaking rudely) or service removal (i.e., threatening citizenship or school admission), which may further discourage caregivers from seeking health care from the health facilities.

Recommendations

The intervention implications of our findings from the demand side–that immunization attitudes are already substantive but that several barriers threaten the sustainability of immunization behaviors–is two-fold. First, immunization interventions should focus not on persuasion (to vaccinate), as rates are already high and attitudes favorable, but rather on the facilitation of people’s desires to immunize their children. Instead of trying to convince people to vaccinate their children, efforts should address barriers so that people can translate their intentions into behaviors. Some of those barriers are physical and structural, while others are interactional, pertaining to how providers and caregivers interact with each other.

Second, health communication messages should focus on addressing normative misperceptions about the prevalence of vaccination rates, by highlighting the fact that full immunization is the norm, not the exception, in most places. Indeed, given the high immunization rates, this type of messaging could be based solely on making people aware that most of their neighbors vaccinate their children and that those who do not do so constitute only a minority. These efforts could also model the ways in which caregivers have overcome specific barriers to vaccination and provide support systems based on caregivers assisting each other–by, for example, pooling childcare responsibilities so that mothers with multiple children are able to travel with the infant needing vaccinations.

Our findings from the supply side–that health workers can be highly motivated to provide quality services but require adequate support from Palikas to do so–speaks to the need to provide appropriately tailored training for providers. More frequent training could help providers understand the plights faced by caregivers that might account for why they have missed vaccinations, and also assist providers in adopting a compassionate, rather than a punitive, approach. The health system also needs to address this positivity bias, and one way of doing so may be to allow for, indeed incentivize, “catch-up protocols,” procedures put into place specifically for helping close the gap on prior vaccines that may have been missed.

Lastly, we point to the need to focus on building trust in the health system. The public health sector runs not on a market-based system, as all services are provided free of charge, but on trust and mutual respect. Caregivers are more likely to use vaccination services if they feel respected by providers and the overall health system and if they can trust the quality of care they receive. Nepal is relatively close to achieving full vaccination coverage, and it appears that this last stretch requires work on both the demand- and supply-sides. On the demand-side, greater trust is required on the part of marginalized groups that have been left out in other major health initiatives. But such trust can come only if these groups believe that the supply-side system is invested in and acting on their interest, an indication of which may be a feeling of being treated with respect when they visit health clinics.

Limitations

Because this study was conducted in one district of Nepal, the experiences of health workers and community members we report may vary from those in other parts of the country. Our study district, Makwanpur, has a higher immunization rate (68%) than the national average (65%), which may explain the positive social norms and high demand related to immunization depicted in this study [3]. Additionally, Makwanpur is geographically more accessible from Kathmandu, with more health facilities and immunization centers than other more remote districts. Hence, people’s access to care and enhanced transportation facilities (at least compared to many remote districts in Nepal) render our findings a bit less generalizable and somewhat privileged. Another limitation is that we did not perform member-checks in that the transcripts were not reviewed by participants. This oversight could introduce some error in our interpretations, but we do note that interviews were captured by both recordings and simultaneous notetaking by a second researcher during the interview.

We also note that participants in the study were limited to those who lived within a one-kilometer radius from the clinic. This means that study findings may not generalize to others at greater distances (and thus facing additional barriers).