Behavioral and structural barriers to human post-exposure prophylaxis and other preventive practices during a canine rabies epidemic

2.1. Ethics statement

This study was approved by the Institutional Review Boards from Universidad Peruana Cayetano Heredia (approval identification number: 65369), Tulane University (#14-606720), and University of Pennsylvania (#823736).

2.2. Study settings and study population

This study was conducted in the Mariano Melgar (MM) district of Arequipa. The city of Arequipa—the second largest in Peru—has a population of 969,000 and is situated ∼2300 meters above sea level (29). MM has 14,500 households and like many districts in Arequipa, has experienced heterogeneous growth and urban development, related to significant unplanned immigration to this city over several decades. Half of the rabid dogs detected in Arequipa have been detected in MM.

Participants in our study came from both urban and peri-urban areas of MM. Urban neighborhoods were founded several decades (in some cases, centuries) ago, and tend to have wealthier and more highly-educated residents compared to those in the peri-urban areas. Homes tend to be more secure due to better construction materials in the urban areas compared to those in the peri-urban areas. Peri-urban neighborhoods, founded just 10-50 years ago during a time of mass rural to urban migration in Peru, tend to have housing made from cheap and often temporary materials, fewer community resources, and more security problems (2).

2.3. Study Design

Data for this qualitative study were obtained through focus group discussions (FGD). The research team developed and applied a focus group guide aimed at examining the following topics: 1) knowledge of rabies virus transmission in dogs and humans, 2) knowledge of signs of canine rabies, 3) knowledge of and practices associated to management of dog bite wounds, and 4) knowledge about and awareness of post-exposure prophylaxis (PEP), and the importance of its completion. Additional topics in the interview guide included dog vaccination and dog ownership; these were analyzed separately and results described elsewhere (2).

Three Peruvian facilitators conducted the FGD: VPS, a social scientist, PhD in public health and experienced focus group facilitator; JBr, BA in psychology and research assistant; and RCN, infectious disease scientist, veterinarian and PhD in epidemiology who leads the study in Arequipa. Another team member from Arequipa took extensive notes. All FGD were carried out in January 2016. After each focus group, a 15-minute conversation was held with the participants to answer questions and eliminate misconceptions about rabies. If any of the FGD participants reported taking actions that would warrant medical care we would refer to the health center for same-day medical advice.

2.4. Sampling strategy

Participants for the eight FGD were recruited from urban and peri-urban neighborhoods of MM through purposive sampling. Participants in households close to health facilities and the district municipal office building are often more exposed to rabies vaccination campaign messages or general health promotion information. Therefore, participants were recruited from households at least 6 blocks away from these sites by pre-selecting blocks on a map.

2.5. Recruitment approach

Recruitment was done by visiting every third house door-to-door in the pre-selected blocks the day before the FGD to recruit 2 or 3 participants per block willing to participate and who met the inclusion criteria. The total number of desired participants per FGD (8 to 12) was reached applying the same procedure in adjacent blocks, although given that the turnout might be low, up to 15 participants were invited. While picking up participants by hired van 15 minutes prior to the FGD, the field team reviewed the informed consent forms with participants to ensure their interest in participating. Participants were picked up from their households with hired transportation 10 to 15 minutes before the FGD and were compensated for their transport home. We recruited a diverse sample (by gender, adult age group, and dog ownership status), ensuring diversity of perceptions and experiences. Prior to starting the FGD, the research team once again reviewed the informed consent with the participants and obtained written consent from each to participate for audio recording.

2.6. Data management and analysis

Digital audio recordings and written notes from the eight FGDs were transcribed verbatim. An inductive coding approach was applied to identify topics that emerged from the data (30). Codes were generated and organized based on emergent and a priori themes of interest related to behavioral and structural barriers to PEP and other prevention strategies. Coding was done by three authors (JBr, CAN, KBM) using ATLAS.ti (31) software. Inconsistent coding was flagged, discussed, and resolved, returning to transcripts when necessary. The analysis of the principal topics of interest was stratified by urban and peri-urban communities, but only mentioned in text when we found differences. All de-identified data are available at https://github.com/chirimacha/Rabies_PEP_Arequipa_qualitative.

3.1. Knowledge of rabies virus transmission in dogs and humans

When asked how dogs become infected with rabies, “dog bites” were identified by most in both the urban and peri-urban areas. However, participants also reported that rabies could be lingering in the environment, and that certain conditions—specifically heat, overexposure to sun, and lack of water for the dog—could lead to dogs acquiring rabies:

“It is my understanding that rabies occurs because the dog is too much in the sun, and that happens more when the dog is on the street, less when it’s at home.” (Female, peri-urban area)

Participants in both areas also mentioned that flies or rats can infect dogs, possibly by direct contact with an open wound. Parasites, poor diet, the stress and frustration of being tied up, and isolation or lack of contact with people were also cited as risk factors for rabies:

“The dog wants to talk but it can’t, so there’s impotence; I imagine it’s stressed. Poor dog. And that’s why they produce rabies.” (Female, urban area)

Although most participants correctly recognized bites from dogs as the main source of transmission to humans, none mentioned dog scratches as possible transmission mechanisms. One urban resident specifically expressed doubts about saliva playing a role in transmission.

FGD participants also differentiated risk of rabies transmission by the dog’s ownership status (owned vs. stray). There was a common perception that stray dogs are at particular risk of carrying rabies because they are not vaccinated, enter easily into fights, are dirty, and often carry other diseases such as scabies. This led to speculation about whether it is necessary to vaccinate dogs that are not let out of the home:

“… if my dog is in the house all day, how will it have rabies if it doesn’t get together with stray dogs? My dog is not from the street, it’s only in the house, it’s domestic. So… maybe I won’t give [worrying about whether her dog may get rabies] much importance.”

(Female, peri-urban area).

FGD participants from both areas suggested there should be more information about human rabies on television, including people’s testimonies, to improve dissemination about disease prevention.

3.2. Knowledge of signs and symptoms of canine rabies

Participants in both study areas correctly described the most widely-recognized symptoms of canine rabies: dog aggressiveness, excessive drooling or foam from the dog’s mouth, and a change in the dog’s behavior. Some participants could describe less well-known symptoms: bright, red, wide-open eyes; staggering; disorientation; and hydrophobia or unwillingness to drink water. No participants recounted the paralytic or dumb form of rabies (in which rabid dogs exhibit progressive limb weakness and inability to swallow, with people frequently presuming dogs have something stuck in their mouths). Participants’ knowledge about what a rabid dog looked like seemed to affect their treatment seeking decision-making for a dog bite if the dog seemed well:

“Because I see there’s nothing wrong with my neighbor’s dog, so it is not worth going (to get treated) and because it takes time. What for, if the dog is fine? [If the dog shows rabies signs later], then yes.” (Man, peri-urban area, owner)

Participants could correctly describe that, following infection with the rabies virus, one might not have any symptoms for days or weeks. While almost all participants thought incorrectly that canine rabies can be easily diagnosed with a simple blood test, only two—from the urban area—mentioned that rabies attacks the dog’s nervous system and that it is necessary to analyze the dog’s brain to confirm disease. Urban residents indicated that they might only seek treatment if signs of infection or other symptoms possibly related to a dog bite or to rabies emerged— focusing on bites with significant soft-tissue damage, but not worrying about small skin-breaking bites.

3.3. Knowledge of and barriers to management of dog bite wounds

FGD participants recognized dog bites as a mode of transmission of rabies virus to humans, and could identify the three key preventive steps that are posted in most area health facilities (Figure 2): 1) wash the wound with water and soap; 2) identify the dog; and 3) seek medical care and report the offending dog at the nearest health post/health center. Participants identified health communication material (Figure 3) as the source of this knowledge.

• Download figure
• Open in new tab

• Download figure
• Open in new tab

Figure 1. (New photos of urban and peri-urban areas)

Despite knowing what to do, most respondents reported not taking action themselves or knowing people who had not sought care or reported a dog following a bite. The most common stated reasons for not seeking care were: 1) lack of awareness that a small bite or skin tear require care, 2) poor treatment at the health post or health center, 3) logistical and economic barriers to seeking care, 4) use of alternative treatments, and 5) perception that a “known” or “good” dog is “safe”.

Respondents agreed that few people would seek care for a minor scratch or bite unless there was significant pain, bruising, bleeding, broken tissue, or the wound failed to heal on its own:

“When the wound bleeds too much, when it doesn’t heal and when it’s way too serious - that’s when you go. When it’s a small bite, people leave it; they cure it with alcohol and leave it.” (Woman, peri-urban area)

Respondents did not draw a connection from waiting to see if the wound healed on its own and delayed treatment.

Perceptions about and experiences of poor quality care was another barrier to seeking care following a bite. Most respondents live near a health post (health centers and hospitals are further away), but had more negative opinions about the posts, particularly the limited hours these are open, the poor customer service, and the stock-outs of the rabies vaccine (which our team confirmed at a nearby health facility following the FGD):

“They mistreat us, don’t attend to us… [Those doctors say] ‘oh, there’s no pills here’. From the post they send us down to the health center, and there they ask us ‘Why were you sent here? They have doctors up there, the health post has this and that…’” (Woman, peri-urban area)

Logistical and economic reasons (i.e., distance to health facility, lack of time, and perceived cost) also prevented respondents from seeking care in a health facility after a dog bite, primarily in the peri-urban area. Participants noted that going to the health post involves hours or even a whole working day, which was not considered feasible, particularly when a dog bite is considered treatable at home:

“It’s a waste of time going to the health post when we can treat it, based on our experience, with soap, hydrogen peroxide…. and it’s cured!” (Woman, urban area).

Some male participants explained that they could not waste work time seeking medical care in the health posts (this comment from a peri-urban male elicited several nods from other men in the group):

“I wouldn’t go because I have debts, I have to work and the post is often full so there’s no time, you waste time—one, two, three hours—hours that you lose of work, so I prefer not to go. You don’t really get a sense of being at risk or not.”

Participants revealed a preference for alternative treatments for dog bites that also reduced likelihood of going to the health post for care. In urban areas people mentioned other remedies such as soap, hydrogen peroxide, penicillin or Mentholatum. People in the peri-urban areas were more likely to mention alternatives associated with local popular healing practices, such as aloe vera and herbs, and even burnt dog hair:

Woman: When I was little a dog bit me on the leg and made a deep wound. The owner cut the dog’s hair, burnt it and put it on the sore with a patch. It scarred well.

Man: That’s a grandmother’s secret. (Peri-urban area)

Older adults were reported as being more likely to use local healing practices—ranging from Mentholatum to herbs to hair ashes—and more reticent to seek care at a health facility than younger people.

Finally, there was also the perception—in both urban and peri-urban areas—that seeking care for a dog bite depended on the specific dog. When probed during the FGDs, peri-urban residents were more likely than urban residents to report having ever been bitten and having been bitten in the past year (some reported three or four bites in the past year); they were also more likely to report not seeking care following a bite. If the biting dog was their own animal or if the owner was known to care for the dog, participants who had been bitten were less likely to seek care. If they did not know the dog’s owner, they perceived a higher chance of acquiring diseases, including rabies. The dog’s behavior also influenced their assessment of rabies risk and subsequent response to a dog bite. If the biting dog was “fierce” (“perro bravo”) or showed aggressive behavior, even if the wound was small, and the dog had an owner, they were more likely to go to the health post than if it was a less aggressive dog. A woman in the peri-urban FGD mentioned that her once-gentle dog suddenly became aggressive, but she did not relate the behavior change to rabies, nor the need to seek treatment for the scratches:

Woman: … and (the dog) started biting my husband. One day she was in my room acting crazy, started to bite and bark (…) I was watching TV, I hadn’t done anything to her, she was irritated (…) and she started biting at me “woof woof woof”. She bit me and I hit her head with my TV thinking she would run away but she didn’t and kept biting me. [It occurred] a month ago I think. My husband had to kill her because she got like that.

Facilitator: Where did she bite you?

Woman: On the leg, but it was just a scratch, she didn’t do anything to me.

Facilitator: Was there any blood?

Woman: Yes (…)

Facilitator: Did you go to the health post?

Woman: No, I washed it with boiled water and soap.*

3.4. Knowledge about and practices associated with post-exposure prophylaxis (PEP) and the importance of its completion

When asked about barriers to accepting PEP or completing PEP once started, participants identified six reasons: 1) rabies vaccines are too painful, 2) people thinking that multiple vaccines could be harmful, 3) people feeling that one or two vaccines are enough, 4) people not wanting to complete the series of vaccines because they are not treated well at health facilities or by health personnel, 5) people not having appropriate information about the importance of completing PEP, and 6) lack of time to continue with the PEP regime. Participants from the urban areas emphasized the first, second, and third responses, while those from peri-urban areas were more likely to mention the fourth, fifth, and sixth reasons.

Participants reported that the rabies vaccine has a particular reputation for being painful. FGD participants in the urban area specifically (and a few in the peri-urban area) conjectured that people might avoid the rabies vaccine for that reason. Poor skills in administering the vaccine was raised as a cause of pain. Estimates of the number of doses of the vaccine required for prophylaxis varied widely, with some older participants mentioning 20 or more shots (as was the case in the 1970s).

Belief that the full series (5 shots) could be harmful was cited as a reason to not adhere the complete series. Multiple respondents noted that “the whole dosage” could have adverse effects, and that other people say that receiving too many vaccines is dangerous:

“I have heard that if you get all the vaccine doses, it harms people…That is why they don’t want to be given the vaccines; other people tell them not to [get the full series] because it’s harmful.” (Woman, urban area).

Participants reported others might stop PEP early if there is no evidence of symptoms and the wound is healed (from start to end of vaccination, it would be 4 weeks). They compared completing PEP to completing antibiotics when one is ill: many stop taking antibiotics when they feel better. One woman in the peri-urban area got all her vaccines when she was bitten, but admitted that if her leg had not looked so swollen, she would have stopped going after the first few vaccines:

“I suppose the person tells the doctor ‘But why should I have five vaccines if I’ve already had three? Besides, I don’t feel anything, it doesn’t hurt, doesn’t itch, doesn’t sting, nothing, I don’t have anything, or don’t have any discomfort’.”

Only peri-urban participants mentioned the quality of services at health facilities as a barrier to continuing PEP. Perceptions of mistreatment surfaced numerous times:

Woman1: …what am I going to go for? To be treated like that? I’d better not go, I’ll just take care of myself.

Woman2: The same with me, they told me ‘no, no, no, you go somewhere else to get attended’. (Peri-urban area)

Suspicions about vaccine quality were also raised, but respondents could not offer further details.

In both areas, participants perceived a lack of information about the risks of discontinuing the PEP regimen, which in turn reduces motivation to make time for dog-bite related medical treatment; participants also referenced people’s ignorance or carelessness in dealing with important issues.

Man: [They don’t continue their vaccines] because people work, they have their families. Facilitator: Even though they can die?

Woman: There are very ignorant people who don’t even care what happens to their children and the rest of people, they just don’t care. (Peri-urban area)

In both areas, lack of time was cited as a likely reason for not seeking to complete one’s vaccination series, specifically referring to people not wanting or not being able to take time off of work. In some parts of Arequipa, this argument might be unfounded since health workers actually go to the homes of people receiving PEP vaccination. Some women in the peri-urban area also mentioned that women might not go due to child care needs. Moreover, because the health posts that primarily service peri-urban communities may not offer vaccines, time spent seeking care at a health center is generally higher for those in peri-urban areas and participants complained about this.