Health Care Worker Intake of Oseltamivir as Post Exposure Prophylaxis (PEP) During a Nursing Home Influenza Outbreak: Predictors of PEP Willingness

Elderly residents of nursing homes are among the most susceptible population and, as such, vulnerable to influenza-related morbidity and mortality. This is due to their age, the presence of multiple comorbidities, and a weakened immune system [1,2]. Influenza outbreaks in nursing homes can therefore have substantial consequences. Health care workers (HCWs) can transmit influenza into these health care settings; a significant number of HCWs (estimated at 10% to 30%) are infected with influenza each winter [3] and most of them continue to work, despite infection (approximately 50% to 80% of those infected) [4]. Thus, prevention of influenza transmission in nursing homes may lead to major health gains, for both residents and staff.

Post Exposure Prophylaxis (PEP) with oseltamivir has been effective in limiting transmission among healthy household contacts of index cases in seasonal influenza outbreaks [5] and could be a potentially effective adjunct to vaccination [6,7]. The Dutch Association of Nursing Home Physicians recommends that HCWs get influenza vaccination and when there is an outbreak to take PEP with oseltamivir because of their contact with those at risk of complications from influenza. In this way, they may protect themselves and the residents against influenza and its complications [8]. Notwithstanding this recommendation of the Dutch Association of Nursing Home Physicians, there is debate as to what extent oseltamivir can be effective in preventing influenza transmission in nursing homes. Therefore, a nationwide Randomized Controlled Trial (RCT) in Dutch nursing homes was started in 2009 to study the effectiveness of PEP with oseltamivir in preventing influenza transmission in nursing homes, the PEpPiE (Post-Exposure Prophylaxis in Evaluation) study. If PEP with oseltamiviris proven effective in this setting, then information on how to implement this strategy in nursing homes is needed.

For successful implementation of PEP with oseltamivir in preventing influenza transmission, it is important to understand factors determining the willingness of HCWs to take PEP with oseltamivir. Although there have been several studies that examined the willingness of nursing home HCWs to accept a vaccine during an influenza outbreak to protect the residents [9,10], there are no comparable studies on PEP with oseltamivir.

This study examined the factors that predicted willingness to take PEP with oseltamivir among HCWs by exploring which demographic, social, and/or organizational factors might be associated with PEP intake in HCWs in the Netherlands. The results of this study provide insight to potentially effective implementation of PEP with oseltamivir among HCWs in Dutch nursing homes.

Study population

We conducted the study among HCWs in nursing homes. Health care workers who worked >1 hour per week in a nursing home were asked to participate in the study.

Study design and setting

From July to September of 2010, we approached physicians and location managers of 29 nursing homes that were recruited from the PEpPiE study by phone and e-mail to participate in this study. Of the 29 nursing homes, nine (31%) agreed to participate in the study. Hereafter, we scheduled an appointment and asked the location managers to announce the study to their HCWs. A researcher was present during the data collection from 7.00-18.00 hours to include HCWs from different shifts. Questionnaires were distributed to each department and collected at the end of the day when possible from boxes or from the individuals themselves. All questionnaires were completed anonymously and administered on the same day in each nursing home. The Ethics committee of University Medical Center Utrecht, The Netherlands, gave ethical approval to conduct this study, as a supporting side study of the trial (NL27938.041.09). No subjectidentifiable results were generated.

Outcome measure

The primary outcome was the willingness of HCWs to take PEP with oseltamivir during an influenza outbreak. The question was formulated as ‘Would you be willing to take PEP with oseltamivir during an influenza outbreak if this is offered to you according to nursing home policy?’ The possible answers were ‘yes’ (positively intended) or ‘no’ (negatively intended).

Intention to take PEP with oseltamivir

We used an anonymous, self-administered questionnaire with 46 questions to assess the factors of the HCW’s willingness to take PEP with oseltamivir during an influenza outbreak. Questions were based on a review of the literature [11-18] and three questionnaires previously developed by the Julius Centre [9]. The questionnaire consisted of items related to demographic, social, and organizational information and consisted mainly of closed questions with a five-point Likert scale: (1) strongly agree, (2) agree, (3) neutral, (4) disagree and (5) strongly disagree.

Demographical factors

There were 11 questions examining demographic factors. These included gender, age, educational level, profession, working hours per week, number of years working in health care, type of shifts worked, whether the HCW shared a home with children, and whether she was pregnant. Nursing home data collected consisted of number of beds per nursing home and occurrence of an influenza outbreak in the last three years. Also, type of ward(s) (somatic, psycho geriatric, revalidation, combined) a HCW worked at was asked and recorded.

Social factors

There were 31 questions examining social factors. Questions on social factors were based on the ‘Health Belief Model’ [19] and the ‘Behavioural Intention Model’ [20]. The Health Belief Model assesses the following five domains: perceived susceptibility, perceived severity, perceived benefits, perceived barriers and cues to action. The questionnaire was complemented with the two domains of the Behavioural Intention Model: attitude (most of these were questions with an ethical perspective) and social influences. The questions on social factors consisted of statements about influenza and oseltamivir, influences of the HCW’s social environment, influence of media attention, and questions about the Dutch guideline for prevention and control of nursing home influenza outbreaks.

Organizational factors

Organizational factors were assessed through four questions about the current situation concerning organization of information on PEP with oseltamivir (information received or not, route of information, and whether information has been sufficient or not). Finally, we asked the HCWs their opinion of various routes of receiving information (e.g., information through friends, manager, physician, colleagues, etc.).

Statistical analysis

Continuous factors (age, years working in health care, working hours per week) were dichotomized with the median as cut-off point. Answers on the five-point Likert scale were divided into three groups:

(1) Agree [defined as agree/strongly agree], (2) neutral, and (3) disagree [defined as disagree/strongly disagree]. Answers of questions from the same factors of the Health Belief Model and the Behavioral Intention Model were summed up to present the total score of that factor. Pearson chi-square test was used to compare differences in proportions among categorical variables. To identify which factors among HCWs were associated with willingness to take PEP, univariate associations of potential factors were assessed. Next, all factors with a P value less than 0.1 in the univariate analysis were used to determine multivariate associations. The final multivariate logistic regression model was obtained using backward selection of variables with P>0.05. The Area under the Receiver Operating Characteristic Curve (AUC) with its corresponding 95% confidence interval (95% CI) was calculated as a measure of discriminative value of the final model. Data were analyzed using SPSS for Windows (Version 19) and SAS for Windows (Version 9.3).

Study population

In total, 524 of 652 (80%) distributed questionnaires were returned. Of all respondents, 61% indicated a willingness to take PEP with oseltamivir during an influenza outbreak if this was offered according to nursing home policy. The nine participating nursing homes had an average bed size of 143 beds per nursing home and consisted mainly of different wards, such as somatic, psycho geriatric and revalidation wards. The non-responding nursing homes had an average bed size of 87 beds per nursing home and consisted mainly of psycho geriatric wards.

Table 1 shows baseline characteristics. The median age of the HCWs was 42 years (IQR, 28-51 years) and 90% of the respondents were female. Four percent of HCWs were nursing home physicians, 8% were nurse, 52% were nursing assistant and 8% were paramedics. The remaining 28% reported a variety of other professions including, for example, 36 domestic workers (7%), 20 feeding assistants (4%), and 14 temporary workers (3%).

Table 1: Baseline characteristics of the study subjects.

Factors univariately associated with willingness to take PEP with oseltamivir during an influenza outbreak

Demographic factors: Table 2 shows the results from the univariate analysis for association of the demographical factors with willingness to take PEP. Male HCWs were more willing to take PEP compared to female HCWs (75% vs. 60% respectively, OR 2.0). The majority of physicians, 90%, indicated that they were willing to take PEP, versus 57% of the nurse assistants, 43% of the paramedics, and 42% of the nurses.

Table 2: Univariate analysis: demographic factors associated with willingness to take PEP among HCWs.

Social factors: Social factors univariately associated with willingness to take PEP with oseltamivirare shown in Table 3. Perceived benefits and a positive attitude towards taking PEP with oseltamivir were strongly associated with taking PEP. The perception that ‘PEP with oseltamivir can reduce work pressure’ was the most strongly associated factor with an intention to take PEP. In addition, the perception that ‘HCWs should take PEP with oseltamivir because of their duty to protect against influenza infection’ was an important factor for taking PEP with oseltamivir. Among this group, 91% of the respondents were willing to take PEP. Other important factors were ‘Agreeing with the contents of the guideline’, ‘PEP with oseltamivir among HCWs reduced the risk to infect residents’, ‘HCWs should take PEP with oseltamivir to ensure continuity of care’ and that it is ‘important not to infect residents’.

Table 3: Univariate analysis: social factors associated with willingness to take PEP among HCWs.

When asked about perceived susceptibility to influenza infection, 39% of the HCWs thought that they were at ‘high risk’ of influenza infection whereas 28% of HCWs thought that they had ‘no risk’, and 34% of the respondents were neutral. Willingness to take PEP was highest when HCWs thought that they had a high risk for influenza infection (74%) compared with the neutral (55%) and the no-risk (51%) group. The most prevalent barrier for not intending to use PEP was the perceived futility in view of a constant flow of visitors.

Organizational factors: Organizational factors univariately associated with willingness to take PEP with oseltamivirare reported in Table 4. Most respondents agreed with the statement that it was important to receive information about taking PEP with oseltamivir from the nursing home, either by letter (93%), from a nursing home physician (63%), through an information meeting (46%) and by Internet (46%). Almost two-thirds of these respondents were willing to take PEP during an influenza outbreak. Respondents who were neutral or thought that it was not important to receive information were less willing to take PEP with oseltamivir.

Table 4: Univariate analysis: organizational factors associated with willingness to take PEP among HCWs.

Factors multivariately associated with willingess to take PEP with oseltamivir during an influenza outbreak

The multivariate analysis resulted in a final seven-item logistic regression model, with three demographic and four social factors that were independently associated with willingness to take PEP during an influenza outbreak (Table 5). The strongest independent factors of willingness were perceived benefits (agree vs. disagree, OR: 4.4) and agreeing with the Dutch guideline for prevention and control of nursing home influenza outbreaks (OR: 5.0).

Table 5: Multivariate logistic regression analysis: factors associated with willingness to take PEP with oseltamivir among HCWs.

HCWs were more likely to indicate willingness to take PEP if they were less than 42 years of age (OR: 2.3), had no children living at home (OR: 1.6), or recorded their occupation as physician (physician vs. nurse/nurse assistant, OR: 11.3). The social factors ‘people close to me think it’s important to get PEP with oseltamivir during an influenza outbreak’(agree vs. disagree, OR: 3.1) and ‘finding it important to do what people close to me think is important’ (agree vs. disagree, OR: 3.0) were also significantly associated with willingness to take PEP. We found no independent association between organizational factors and willingness to take PEP. The AUC for the final model including all seven factors was 0.85 (95% CI, 0.81-0.89).

This study examined the factors that predicted willingness to take PEP with oseltamivir among HCWs in nursing homes. We found a positive inclination toward acceptance of 61% regarding PEP with oseltamivir among HCWs, and furthermore, seven independent factors could explain 85% of the variation in willingness. The strongest factors were agreeing with the Dutch guideline for prevention and control of nursing home influenza outbreaks and perceived benefits for personnel, namely reducing work pressure, reducing the risk to infect residents and continuity of care. HCWs were also more likely to take PEP if they were of young age and had no children living at home. Similarly, HCWs considered their social environment as important factors in the decision to take PEP with oseltamivir.

Reported willingness to take PEP varied considerably according to the individual’s knowledge/awareness and their job classification. Physicians were significantly much more likely than the nursing staff to have a positive intention. This may be attributed to greater knowledge about PEP among the physicians, or may also reflect their stronger beliefs in the efficacy and safety and a perception of greater personal susceptibility to influenza [21,22].

Many HCWs had no strong opinion about PEP with oseltamivir, supported by the relatively high percentage of neutral responses to the questions. This may be related to the fact that many HCWs may not know much about oseltamivir to prevent influenza. This is also confirmed by the fact that many HCWs reported no knowledge of the advice and contents of the Dutch guideline for prevention and control of nursing home influenza outbreaks. It also may indicate that HCWs have doubts about the effectiveness of the antiviral drug. Respondents were much more likely to be willing to take PEP when they accepted that oseltamivircan be effective in preventing influenza cases and that it can reduce work pressure. Helping HCWs to understand various protocols (in this case for influenza outbreak) might increase their acceptance of its effectiveness and thereby, their adherence. Providing enhanced education, staff communication and support could well improve HCW adherence to PEP [23].

This study has several strengths. To our knowledge, this is the first study to examine the willingness to take PEP with oseltamivir among HCWs. An extensive questionnaire based on the literature and previous similar researches was used. This questionnaire addressed demographic, social, and organizational factors. As such, it is unlikely we have missed major other predictors of the willingness to take PEP.

There are also some limitations that have to be acknowledged. The first limitation of this study was the representativeness of the study sample. Only 31% of invited nursing homes participated in the study. Non-responding nursing homes tended to be smaller than responding nursing homes. Yet, when we compare the percentage of female HCWs in our study with national data, we notice that this is similar to all Dutch nursing homes [24]. As a second limitation, although the questionnaire was anonymous, there is a possibility that HCWs may have given what they thought were socially desirable answers to questions. It is also important to recognize that willingness is not necessarily the same as actual behaviour.

We identified one previous study that examined adherence behaviour in HCWs who received PEP with oseltamivir after influenza exposure [25]. In this study, adherence to PEP was higher if HCWs were aware of personal risk for influenza infection and believed that PEP with oseltamivir was effective in preventing influenza transmissions. Those participants who experienced no side effects or knew someone else in their environs taking oseltamivir were more likely to take PEP with oseltamivir. This study indicates that improving knowledge and awareness by providing formal counselling may increase the willingness to take PEP with oseltamivir among HCWs, as we have discussed earlier [25,26].

Although vaccination is the mainstay of influenza control and prevention, the average vaccination rate among HCWs in Dutch nursing homes is low (20%) [6,7,27]. Therefore, the use of (a combination of vaccination and) PEP with oseltamivir during an influenza outbreak may play an important role in preventing influenza transmission in nursing homes [7]. By providing social support and HCW education with reliable information about the perceived benefits, efficacy, and safety could raise awareness and improve knowledge of appropriate influenza outbreak interventions and confidence in antivirals among the nursing staff.

The authors would like to thank the nursing homes for participating, as well as the staff from every department and wards that cooperated in this study.