Study team
EXECUTIVE SUMMARY

Pregnant women and their infants are at high risk for complications associated with influenza infection.  The World Health Organization recommends maternal influenza vaccination to protect both mothers and infants but there are reports that this recommendation has been met with mixed acceptance from pregnant women and their healthcare providers. We aim to describe the knowledge, attitudes and practices of pregnant women and healthcare workers regarding maternal influenza vaccination in South Africa. 

INTRODUCTION

Project Summary
Maternal influenza vaccination is recommended to protect pregnant women and their infants from influenza associated complications. Both pregnant women and their healthcare providers play a crucial role in ensuring maternal influenza vaccination during the season. We aim to describe the knowledge, attitudes and practices regarding influenza vaccination in pregnant women and healthcare workers in antenatal clinics in South Africa.

Background
During the 2009 influenza A (H1N1) pandemic, pregnant women were found to be at increased risk for influenza-associated complications, including hospitalization and death [1,2]. Influenza vaccination is the most effective strategy for preventing influenza infection. Recent studies have shown that maternal influenza vaccination is safe and has protective benefit, reducing severe complications in both mothers and infants in the first 6 months of life [3]. 

In South Africa, the Department of Health recommends influenza vaccination for pregnant women regardless of stage of pregnancy [4]. Despite this priority indication, few doses of influenza vaccine are available for use in pregnant women. For example, in 2013 in the Gauteng province, 31,000 doses of influenza vaccine were available for use in a target population of approximately 130,000 pregnant women [5].  Underutilization is another barrier to influenza vaccination in pregnant women. A recent cross sectional survey among health care workers in South Africa showed low vaccine coverage among providers despite appropriate knowledge about influenza (J. Duque, personal communication, December 17, 2014). We now propose to describe motivators and barriers to maternal influenza vaccination as reported by South African pregnant women and their health care providers. 

To gather evidence on clinical outcomes among women who receive influenza vaccination, an influenza vaccine effectiveness study in pregnant women and their newborns is planned in South Africa starting in 2015. As a corollary to this study, we intend to characterize the knowledge, attitudes and practices regarding maternal influenza vaccination among pregnant women and healthcare workers at the study sites.

OBJECTIVES

Primary objective
Describe the knowledge, attitudes and practices related to maternal influenza vaccination among pregnant women and healthcare workers in antenatal clinics in South Africa.

Secondary objectives

• Describe barriers to influenza vaccine acceptance and receipt among pregnant women and health care workers caring for pregnant women.
• Identify factors related to vaccination decisions among pregnant women.
• Measure provider knowledge of influenza vaccine recommendations.

METHODS

Location and timing of study
We will interview pregnant women and health care workers, using a standardized questionnaire, at antenatal clinics in the Gauteng and Western Cape Provinces, Cities of Cape Town and Johannesburg, South Africa (Appendix A). Pregnant women will be interviewed in early June after the seasonal influenza vaccination campaign is completed.  A convenience sample of health care workers and women presenting to antenatal clinics will be approached sequentially and asked to enroll until the sample size is reached. 

Inclusion criteria:

• Pregnant women, 18 years of age or older, who presented for medical care at one of the designated antenatal clinics in Gauteng and Western Cape provinces in June-October 2015(see Appendix A) 
• Healthcare workers (i.e. family physician, obstetricians, advanced midwife or nurse) in clinical practice at the designated antenatal clinics who care for pregnant women

Exclusion criteria

• Pregnant women younger than 18 years old
• Pregnant women seeking medical or antenatal care solely outside the designated 38 clinics
• Refusal to participate in the survey
• Inability to provide informed consent for participation
• Health care workers employed solely at antenatal clinics outside of Gauteng and Western Cape provinces

Sample size
We plan to enroll at least 854 pregnant women in the study. This sample size will provide 80% power to detect a 10% difference in survey responses between vaccinated and unvaccinated women (assuming 50% of surveyed women are vaccinated, a 10% non-response rate and alpha = 0.05). We will enroll, if possible, a total of 930 health care workers.  This sample size will provide 80% power to detect a 10% difference in survey responses between health care workers recommending or not recommending influenza vaccination (assuming 50% of surveyed health care workers provide maternal influenza vaccination, a 20% non-response rate and alpha = 0.05).

Survey tool and variables of interest
The study will consist of two cross-sectional interview-administered surveys, one designed for pregnant women (Appendix B) and the other for healthcare providers (Appendix C). In general, the survey questions will address the following: socioeconomic background, prior influenza vaccination history, knowledge of influenza infection and vaccine and current vaccination status. For pregnant women, the survey will include questions regarding beliefs about influenza vaccine safety, risks and benefits for mothers and infants. For health care workers, additional questions will characterize their knowledge of influenza vaccine guidelines, attitudes towards vaccination and practices related to recommending influenza vaccine in pregnant women. The co-primary outcome variables are 1) receipt of influenza vaccination among pregnant women and 2) influenza vaccine recommendation among health care workers. 

Data analysis
Descriptive statistics will be conducted using SAS v9.3 software to perform univariate, bivariate and multivariate logistic regression analyses. We will identify variables associated with the decision to receive or prescribe influenza vaccines, calculating Chi-square p-values and creating a logistic regression model to determine the strength of association (if any) between variables. 

Dissemination and publication of results
Study results will be compiled and distributed to key stakeholders: CDC South Africa, National Institute for Communicable Diseases and the South Africa Department of Health. An abstract and manuscript will be prepared for a peer-reviewed scientific journal.

ETHICAL CONSIDERATIONS

This observational research study poses no anticipated risks to subjects. Completion of an approximately 15 to 20 minute self-administered survey or in-person interview is the only requirement for participation. Participants will be given a written informed consent form describing the risks and benefits of study participation (Appendices D and E). Their participation is completely voluntary and collected study findings will be reported anonymously. We will apply for ethics approval from the Human Research Ethics Committee (Medical), University of the Witwatersrand, Johannesburg and University of Cape Town. Additional approval, if needed, will be requested from provincial Health Departments, collaborating institutions and/or local universities. 

Study team