Factors Associated with Health Care–Associated 2009 Influenza A (H1N1) Virus Infection among Thai Health Care Workers

To the Editor—Unvaccinated health care workers (HCWs) are at risk for occupational exposure and infection with influenza and can serve as vectors for health care–associated (HA) influenza transmission to hospitalized patients [1, 2]. The World Health Organization reported 35,446 laboratory-confirmed cases of pandemic 2009 influenza A (H1N1) in Thailand by the end of March 2010 [3]. The vaccine for 2009 H1N1 influenza virus was not available in Thailand until 10 February 2010, and exposure of HCWs to patients infected with this virus increased as the pandemic evolved. We conducted a case-control study to identify factors associated with HA 2009 H1N1 influenza infection among Thai HCWs in a 500-bed tertiary care medical center during the period from 1 May 2009 through 15 January 2010. Data collected included age, sex, occupation, hospital units, unit hand hygiene compliance rates (before and after patient contact) during the month when index HCWs had confirmed 2009 H1N1 influenza infection, and exposure to patients with suspected or confirmed 2009 H1N1 influenza regardless of whether a surgical mask was used. Case patients were all HCWs with laboratory-confirmed 2009 H1N1 influenza. For each case, 2 matched controls were randomly selected among HCWs who did not have influenza during the period 2 weeks before and after the case patient's confirmed infection date and who were assigned to care for patients who were suspected of having 2009 H1N1 influenza. HCWs identified as having influenza-like illness underwent nasopharyngeal swab specimen collection for influenza testing by means of a rapid influenza diagnostic test (SD Bioline Influenza Antigen A/B; MT Promedt Consulting) and reverse-transcription polymerase chain reaction [4]. Influenza-like illness was defined according to criteria set by the Centers for Diseases Control and Prevention [5]. A suspected case of HA influenza was defined as influenza-like illness if onset of illness occurred within 4 days of exposure to a HCW or patient with confirmed 2009 H1N1 influenza. A confirmed case of HA influenza was defined as a suspected case that was laboratory confirmed as influenza and did not result from community influenza exposure, as measured according to the presence of ill household members. Frontline HCWs were defined as staff who worked in the emergency department, triage unit, or outpatient clinic and screened patients with suspected 2009 H1N1 influenza. In this hospital, all HCWs with suspected or confirmed 2009 H1N1 influenza have been required to report to the hospital's occupational health unit since May 2009, and the infection control policy was for HCWs to wear surgical masks while caring for patients with suspected 2009 H1N1 influenza, whereas the use of N95 masks was restricted to aerosol-generating procedures. A hospital-wide hand hygiene program with monthly monitoring and feedback has been sustained since July 2006 [6]. During the 7.5-month study period, 69 HCWs had confirmed 2009 H1N1 influenza, 51 (74%) of whom met the criteria for HA 2009 H1N1 influenza. Eighteen HCWs with 2009 H1N1 influenza were excluded because of potential community influenza exposure (they had ill household members). All 51 HCWs consented to study enrollment; the median age of the study group was 36 years (range, 24–54 years), and 24 (47%) of 51 HCWs reported unprotected exposure to patients with confirmed 2009 H1N1 influenza (Table 1). Nurses were the most frequently exposed (26 [51%] of 51 exposed HCWs), and most exposures occurred on medical units (30 [59%] of 51 exposed HCWs). According to multivariate analysis, unprotected exposure to patients with 2009 H1N1 influenza (adjusted odds ratio [aOR], 2.41 [95% confidence interval {CI}, 1.19–16.71]) was the sole factor associated with HA 2009 H1N1 influenza infection in the case group, whereas protective factors for the control group included assignment to units with >75% adherence to hand hygiene practices (aOR, 0.45 [95% CI, 0.27–0.89]), use of a surgical mask while caring for a patient with suspected 2009 H1N1 influenza (aOR, 0.79 [95% CI, 0.41–0.99]), and assignment as a frontline HCW who cared for patients with suspected or confirmed 2009 H1N1 influenza (aOR, 0.83 [95% CI, 0.46–0.98]). Notably, exposure to patients who did not meet criteria for influenza-like illness was associated with a high proportion of unprotected HCW exposures (11 [46%] of 24). Demographic and Clinical Characteristics of Health Care Workers (HCWs) with and without Confirmed Health Care–Associated 2009 Influenza A (H1N1) Notably, 24 (47%) of 51 cases of HA influenza among HCWs in this study group involved unprotected exposures by HCWs—a behavioral risk associated with HA 2009 H1N1 influenza infection. Unprotected exposures were more extensive among the HCWs who provided initial care to patients who did not have classic influenza-like illness, likely as a result of delays in initiation of standard infection control measures. In addition, a high rate of compliance with hand hygiene practices (175%) was independently associated with reduced transmission of 2009 H1N1 influenza, and the protective effect of surgical masks is consistent with existing literature for 2009 H1N1 influenza [7, 8???]. Last, factors associated with unprotected exposure and reduced risk of infection among frontline HCWs will require further exploration in other study populations to provide additional evidence for optimized infection control practices and minimization of viral infection. Financial support. Thammasat University Infectious Diseases and Infection Control Research Fund. Potential conflicts of interest. L.M.M. is a consultant to WWEpidemiology at GlaxoSmith-Kline, Inc. A.A.: no conflicts.