Comparison of Laboratory-Confirmed Influenza and Noninfluenza Acute Respiratory Illness in Healthcare Personnel during the 2010–2011 Influenza Season Author(s): Emily Henkle, PhD, MPH; Stephanie A. Irving, MHS; Allison L. Naleway, PhD; Manjusha J. Gaglani, MD; Sarah Ball, ScD, MPH; Sarah Spencer, PhD; Sam Peasah, PhD; Mark G. Thompson, PhD Source: Infection Control and Hospital Epidemiology, Vol. 35, No. 5 (May 2014), pp. 538-546 Published by: The University of Chicago Press on behalf of The Society for Healthcare Epidemiology of America

Stable URL: http://www.jstor.org/stable/10.1086/675832 . Accessed: 03/06/2014 08:02 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

.

The University of Chicago Press and The Society for Healthcare Epidemiology of America are collaborating with JSTOR to digitize, preserve and extend access to Infection Control and Hospital Epidemiology.

http://www.jstor.org

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

infection control and hospital epidemiology

may 2014, vol. 35, no. 5

original article

Comparison of Laboratory-Confirmed Influenza and Noninfluenza Acute Respiratory Illness in Healthcare Personnel during the 2010–2011 Influenza Season Emily Henkle, PhD, MPH;1 Stephanie A. Irving, MHS;1 Allison L. Naleway, PhD;1 Manjusha J. Gaglani, MD;2 Sarah Ball, ScD, MPH;3 Sarah Spencer, PhD;4 Sam Peasah, PhD;4 Mark G. Thompson, PhD4

objective. Compare the severity of illnesses associated with influenza and noninfluenza acute respiratory illness (ARI) in healthcare personnel (HCP). design.

Prospective observational cohort.

participants.

HCP at 2 healthcare organizations with direct patient contact were enrolled prior to the 2010–2011 influenza season.

methods. HCP who were fewer than 8 days from the start of fever/feverishness/chills and cough were eligible for real-time reversetranscription polymerase chain reaction influenza virus testing of respiratory specimen. Illness severity was assessed by the sum of selfrated severity (0, absent; 3, severe) of 12 illness symptoms, subjective health (0, best health; 9, worst health), activities of daily living impairment (0, able to perform; 9, unable to perform), missed work, and duration of illness. results. Of 1,701 HCP enrolled, 267 were tested for influenza, and 58 (22%) of these tested positive. Influenza compared with noninfluenza illnesses was associated with higher summed 12-symptom severity score (mean [standard deviation], 17.9 [5.4] vs 14.6 [4.8]; P ! .001), worse subjective health (4.5 [1.8] vs 4.0 [1.8]; P !.05), greater impairment of activities of daily living (4.9 [2.5] vs 3.8 [2.5]; P ! .01), and more missed work (12.1 [10.5] vs 7.8 [10.5] hours; P ! .01). Differences in symptom severity, activities of daily living, and missed work remained significant after adjusting for illness and participant characteristics. conclusions. Influenza had a greater negative impact on HCP than noninfluenza ARIs, indicated by higher symptom severity scores, less ability to perform activities of daily living, and more missed work. These results highlight the importance of efforts to prevent influenza infection in HCP. Infect Control Hosp Epidemiol 2014;35(5):538-546

Influenza virus infection causes substantial morbidity, mortality, and healthcare burden in the United States each year.1 Influenza illness is of particular concern to healthcare personnel (HCP) who have direct patient contact and are at increased risk for infection.2 HCP absenteeism has a significant economic impact, and presenteeism—or working while ill—may result in transmission of influenza virus to vulnerable patients.3,4 Seasonal influenza vaccine is the primary public health tool for preventing influenza and is recommended for HCP, though increasing vaccine uptake in this population has been challenging5 partly because of the persistent misconception among HCP that influenza is not a serious illness among otherwise healthy working adults.6,7 We enrolled HCP in a prospective active surveillance study of respiratory illness during the 2010–2011 influenza season, with the following 3 objectives. First, we compared the severity

of illnesses associated with laboratory-confirmed influenza and noninfluenza acute respiratory illness (ARI) in regard to selfrated symptom severity, overall subjective illness severity, impairment to functioning, missed work, and duration of illness. Although a previous community-based study reported a higher prevalence of respiratory and systemic symptoms and higher subjective severity of symptoms among patients with influenza compared with noninfluenza ARI,8 we extend this research to ARI among HCP. Second, we explored differences between medically attended influenza and non–medically attended influenza illness episodes. Although the majority of influenza virus infections never lead to medical visits,9 the extent to which medically attended and non–medically attended illnesses differ in severity is not well described. Finally, for exploratory purposes, we examined whether vaccination with seasonal influenza vaccine modified illness severity.

Affiliations: 1. Center for Health Research, Kaiser Permanente Northwest, Portland, Oregon; 2. Scott and White Healthcare, Temple, Texas; 3. Abt Associates, Cambridge, Massachusetts; 4. Influenza Division, Centers for Disease Control and Prevention, Atlanta, Georgia. Received September 6, 2013; accepted December 13, 2013; electronically published March 6, 2014. 䉷 2014 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2014/3505-0011$15.00. DOI: 10.1086/675832

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

influenza in hcp

539

table 1. Characteristics of Healthcare Personnel Cohort and Swabbed Participants during the 2010–2011 Influenza Season Entire cohort (N p 1,701) No. Site Scott White Healthcare Kaiser Permanente Northwest Age, years 18–34 35–49 ≥50 Sex Female Male Race/ethnicity White, non-Hispanic White, Hispanic Black Asian Other Education, mean (SD), years Occupation Physician Nurse Technicians and medical assistants Other Influenza risk factors Household with children aged !12 years Current smoker Confirmed high-risk medical condition Body mass index, kg/m2 Normal or lower, !25 Overweight, 25–29.9 Obese, ≥30 Missing Subjective health status Excellent Very good Good Fair or poor 2010–2011 influenza vaccination status Vaccinated Unvaccinated note.

%

Swabbed participants (N p 267) No.

%

1,063 638

62 38

174 93

65 35

545 650 506

32 38 30

85 104 78

32 39 29

1,354 347

80 20

231 36

87 13

1,264 77 149 79 132 16.1

74 5 9 5 8 2.5

183 12 28 14 30 16.0

69 4 10 5 11 2.3

230 576 751 144

14 34 44 8

31 81 131 24

12 30 49 9

655 107 364

39 6 21

105 18 77

39 7 29

578 536 516 71

34 32 30 4

86 76 96 9

32 28 36 3

571 810 301 19

34 48 18 1

80 124 59 4

30 46 22 1

1,307 394

77 23

221 46

83 17

Data are no. (%), unless otherwise indicated. SD, standard deviation.

methods Recruitment and Eligibility As described in full elsewhere,10 participants were enrolled in a prospective observational study of respiratory illness in HCP between September and December 2010 at Scott and White Healthcare in Temple and Round Rock, Texas, and Kaiser Permanente Northwest in the Portland, Oregon, metropolitan area. HCP were eligible if they met the following criteria: (1) aged 18–65 years, (2) working full time (32 hours or more

per week), (3) employed by and receiving medical care from their respective healthcare system for 12 months or longer, and (4) providing direct patient care, defined as “regular, close, face-to-face, or hands-on contact with patients as part of a typical work shift, including regular contact within 3 feet of patients for 5 minutes or more.”11 We conducted recruitment through e-mail invitations, fliers, and/or Internet and radio announcements before the 2010–2011 influenza season began. Employees were offered vaccinations through employee health campaigns from September to November 2010.

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

540

infection control and hospital epidemiology

may 2014, vol. 35, no. 5

table 2. Sociodemographic Characteristics Associated with Influenza Real-Time Reverse-Transcription Polymerase Chain Reaction Results in Healthcare Personnel with Acute Respiratory Illness during the 2010–2011 Influenza Season

Age, mean (SD), years Sex Male Female Race/ethnicity White, non-Hispanic All others Education, mean (SD), years Occupation Physician Nurse Technicians and medical assistants Other Influenza risk factors Household with children aged !12 years Current smoker Confirmed high-risk medical condition Body mass index, mean (SD), kg/m2 Subjective health status Excellent Very good Good Fair or poor 2010–2011 influenza vaccination status Vaccinated Unvaccinated

Influenza positive (N p 58)

Influenza negative (N p 209)

No.

%

No.

%

P

40.6

11.1

42.3

11.4

.40

4 54

7 93

32 177

15 85

.10

37 21 15.4

64 36 2.0

146 63 16.2

70 30 2.4

.38

4 19 30 5

7 33 52 9

27 62 101 19

13 30 48 9

.64

30 5 15 28.0

52 9 26 5.5

75 13 62 29.1

36 6 30 7.2

.03* .52 .57 .44

18 26 13 1

31 45 22 2

62 98 46 3

30 47 22 1

.99

44 14

76 24

176 33

84 16

.14

.01*

note. Data are no. (%), unless otherwise indicated. P values for categorical variables calculated using x2 or Fisher exact tests; continuous variable comparisons calculated using nonparametric 1-way ANOVA. SD, standard deviation. * P ! .05.

Participants received small incentives in the form of cash (Scott and White Healthcare) or gift cards (Kaiser Permanente Northwest) after completing specific study events, including the surveys and illness surveillance. The study was approved by the institutional review boards at both sites. Participant Characteristics Participants completed surveys over the Internet or on paper at enrollment (fall 2010) and following the influenza season (May/June 2011). We gathered date of birth, sex, highest education level achieved, occupation, and smoking status (current, prior, never) from these surveys. We calculated age as of September 1, 2010, and body mass index from either selfreported height and weight or electronic health record (EHR) data, if available. We collected self-rated health status on a 5-point scale12 at enrollment and collapsed these scores into fair/poor, good, very good, and excellent. From the EHR, we extracted a complete history of diagnosis and procedure codes, going back to January 1, 2005.

We used International Classification of Diseases, Ninth Revision (ICD-9) codes to identify participants at high risk for influenza complications, such as those with asthma or diabetes.13 We considered participants to be confirmed high risk if they had at least 2 high-risk ICD-9 codes at different visits or across disease categories over the 5-year period. We obtained EHRconfirmed influenza vaccination records for the 2010–2011 influenza season. HCP were considered vaccinated if the vaccination was documented in the EHR or self-reported on the end-of-season survey. Respiratory Illness Surveillance We prospectively followed the HCP cohort through the 2010– 2011 influenza season. The influenza season, on the basis of predefined criteria referencing local influenza-like illness surveillance and identification of laboratory-confirmed influenza cases from clinical or state laboratories, began on December 15, 2010, at Scott and White Healthcare and on December 22, 2010, at Kaiser Permanente Northwest and ended at both

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

influenza in hcp

541

table 3. Clinical Characteristics Associated with Influenza-Positive versus -Negative Acute Respiratory Illnesses (ARIs) in Healthcare Personnel during the 2010–2011 Influenza Season Influenza-positive ARIs (N p 58) No. Days from onset to swab collection, mean (SD) Respiratory symptoms present Cough Fever/feverishness Chills Sore throat Nasal congestion Wheezing Ear pain Shortness of breath Nonrespiratory symptoms present Fatigue Headache Muscle aches Nausea Vomiting Diarrhea Subjective symptom severity and impact Severity score, mean (SD)a Subjective health during illness, mean (SD)b ADL impairment rating, mean (SD)c Illness duration, mean (SD), daysd Missed work due to illnesse Hours of worked missed, mean (SD) Medical utilization MAARI-related outpatient visitf MAARI-related hospitalizationf Antibiotic prescriptiong Antiviral prescriptiond

4.3

% 2.3

Influenza-negative ARIs (N p 209) No. 4.7

% 1.9

P .21

58 56 53 42 53 32 27 23

100 97 91 72 91 55 47 40

204 174 165 175 184 73 92 63

98 83 79 84 88 35 44 30

.24 .01* .03* .05 .48 !.01* .73 .17

58 52 49 23 9 15

100 90 84 40 16 26

196 180 162 69 18 54

94 86 78 33 9 26

.05 .48 .25 .35 .12 .98

14.6 4.0 3.8 11.4 106 7.8

4.8 1.8 2.5 5.2 51 10.5

17.9 4.5 4.9 11.2 39 12.1

5.4 1.8 2.5 8.1 67 10.5

17 0 15 6

29 0 26 12

41 1 41 12

20 0 20 7

!.0001*

.04* !.01*

.24 .03* !.01* .14 NA .30 .31

note. Data are no. (%), unless otherwise indicated. P values for categorical variables calculated using x2 or Fisher exact tests; continuous variable comparisons calculated using nonparametric 1-way ANOVA. ADL, activities of daily living; MAARI, medically attended acute respiratory illness; NA, not applicable; SD, standard deviation. a Summed subjective severity of 12 symptoms: cough, fever/feverishness, chills, sore throat, nasal congestion, ear pain, fatigue, headache, muscle aches, nausea, and vomiting. Each symptom rated as absent (0), mild (1), moderate (2), or severe (3). b Participants asked to rate their health during their illness (0, normal health; 9, worse possible health). c Participants asked to rate their ability to perform activities during their illness (0, able to perform usual activities; 9, unable to perform usual activities). d Self-reported. Denominator for antiviral use and illness duration is 52 polymerase chain reaction (PCR)⫹ and 168 PCR⫺; illness duration interviews were not collected from 6 PCR⫹ and 41 PCR⫺ participants. e As of specimen collection; did not capture work missed after collection. f MAARI-related encounters were defined as encounters within 7 days prior to illness onset through 28 days following onset. g Filled antibiotic prescription within 28 days of illness onset. * P ! .05.

sites on April 22, 2011. Participants reported respiratory illnesses by completing a weekly symptom surveillance survey online or by automated telephone or by contacting study personnel directly. In addition, we conducted daily EHR screenings to identify participants with documented medically attended ARI (MAARI) visits and/or triage phone calls that were potentially associated with a respiratory illness. An MAARI visit was defined as an inpatient, outpatient, or emer-

gency department visit associated with ARI ICD-9 diagnosis codes 460–466, pneumonia diagnosis codes 480–486, or influenza diagnosis codes 487–488.1. Participants met the case definition for influenza testing if they reported an ARI defined as fever/feverishness/chills and cough with onset during the prior 7 days. Study personnel collected nasopharyngeal, nasal, and oropharyngeal swabs from participants meeting the case definition. Influenza virus testing was performed by real-time

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

542

infection control and hospital epidemiology

may 2014, vol. 35, no. 5

figure 1. Mean and 95% confidence interval of summed self-rated symptom severity among healthcare personnel by influenza versus noninfluenza illness and whether it was medically attended during the 2010–2011 influenza season. Summed severity of 12 symptoms: cough, fever/feverishness, chills, sore throat, nasal congestion, ear pain, fatigue, headache, muscle aches, nausea, and vomiting. Each symptom rated as absent (0), mild (1), moderate (2), or severe (3). MAARI, medically attended acute respiratory illness.

reverse-transcription polymerase chain reaction (rRT-PCR) at the Marshfield Clinic Research Foundation Core Laboratory; a positive test from any of the 3 specimen types was considered a positive influenza result. We administered an in-person interview of illness episode characteristics at the time of specimen collection. We assessed overall activities of daily living impairment on a scale of 0– 9, with 0 being able to perform most activities and 9 being unable to perform most activities. Subjective health during illness was collected on the same scale, with 0 being normal health and 9 being the worst possible health. Participants were asked if they had missed work during the current illness. We measured the severity of 14 symptoms, including respiratory (cough, nasal congestion, wheezing, sore throat, ear pain, and shortness of breath) and other symptoms (fever/feverishness, chills, fatigue, vomiting, headache, muscle aches, nausea, and diarrhea). A symptom severity score was calculated for each swabbed illness by summing self-rated scores for 12 symptoms, using the following scale: 0, absent; 1, mild; 2, moderate; 3, severe. For consistency with the symptom severity reported by Belongia et al,8 shortness of breath and diarrhea were excluded from the calculation. We assessed the date of symptom resolution and antiviral medication use during an Internet-based survey administered 10–14 days after specimen collection. An illness episode was defined as the period between reported onset and resolution of symptoms; participants could have multiple episodes over the study period. A new episode was defined as initiation of symptoms at least 14 days after the resolution of symptoms from the previous illness. An MAARI visit occurring anytime from 7 days prior to symptom

onset to 28 days after symptom onset was considered to be associated with an illness episode. Statistical Methods We compared participant characteristics using the Pearson x2 or Fisher exact test for categorical outcomes or nonparametric Wilcoxon Mann-Whitney U test for continuous variables. The primary outcome was rRT-PCR-confirmed influenza, with additional stratification for episodes with and without associated MAARI visits. The Wilcoxon rank sum test for trend was used to compare symptom severity across combinations of rRT-PCR results and MAARI presence/absence. For all analyses, we included a single episode per participant: either the influenza-positive illness episode or the first influenza-negative illness for those without a positive result. Multivariate linear regression was used to examine the association between influenza and the severity indicators. Predetermined participant and illness characteristics included in the multivariate analysis were age, sex, enrollment site, 2010–2011 vaccination status, presence of a high-risk health condition, days from illness onset to specimen collection, and whether the illness was medically attended; other variables were added if significantly associated (P ! .05) with influenza positivity in bivariate analyses.

results We enrolled a total of 1,834 HCP (20% of eligible HCP at Kaiser Permanente Northwest and 40% at Scott and White Healthcare) and collected both enrollment and end-of-season surveys from 1,701 HCP (93%), for whom the full set of

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

influenza in hcp

543

table 4. Linear Regression Predicting Illness Severity Indicators with Unstandardized b and Standard Errors (SEs) for 267 Febrile Acute Respiratory Illnesses among Healthcare Personnel Symptom severitya

Participant characteristics Site (KPNW) Sex, female Age, years Child in household Education, years Chronic illness Illness characteristics Days since illness onset Medically attended illness Influenza positive Received influenza vaccine

Subjective healthb

Perform ADLc

Hours missed workd

b

SE

b

SE

b

SE

b

SE

⫺1.33 2.37 ⫺0.02 ⫺0.07 ⫺0.20 ⫺0.55

0.66* 0.94** 0.03 0.68 0.14 0.62

0.02 0.34 0.00 0.00 0.04 ⫺0.16

0.25 0.35 0.01 0.26 0.05 0.23

0.67 0.49 ⫺0.01 ⫺0.07 ⫺0.06 ⫺0.22

0.33* 0.46 0.02 0.33 0.07 0.33

1.04 0.37 0.01 ⫺2.41 ⫺0.27 ⫺0.26

1.41 2.01 0.06 1.44 0.30 1.43

⫺0.01 1.35 2.68 ⫺0.66

0.16 0.74 0.74*** 0.81

0.08 0.16 0.52 ⫺0.67

0.06 0.28 0.28 0.30*

0.03 0.94 0.95 ⫺0.68

0.08 0.36** 0.37** 0.40

0.98 6.66 4.22 ⫺2.54

0.33** 1.56*** 1.57** 1.72

note. ADL, activities of daily living; KPNW, Kaiser Permanente Northwest. a Sum of ratings (0, absent; 1, mild; 2, moderate; 3, severe) for cough, fever/feverishness, chills, sore throat, nasal congestion, ear pain, fatigue, headache, muscle aches, nausea, and vomiting. b Participant’s rating of current health while ill (0, normal health; 9, worse possible health). c Participant’s rating of ability to perform activities during their illness (0, able to perform usual activities; 9, unable to perform usual activities). d Hours missed from work as a result of current illness. * P ! .05. ** P ! .01. *** P ! .001.

characteristics was available for analysis (Table 1). We identified 288 ARIs meeting our case definition with a respiratory specimen collected. After limiting analyses to 1 illness episode per participant, we included 267 illnesses: 58 (22%) influenza cases and 209 (78%) influenza-negative ARIs. Swabbed participants (Table 1) were evenly distributed across the 3 age groups and were predominantly female (87%) and white nonHispanic (69%). Overall, most (76%) reported that they were in very good or excellent health. Approximately 29% had a diagnosed high-risk condition, and 221 (83%) were vaccinated with the 2010–2011 seasonal influenza vaccine. Of the 58 influenza cases, 53 (91%) had influenza A virus and 5 (9%) had influenza B virus detected. Among the 53 influenza A viruses, 37 (70%) were A(H3N2), 15 (28%) were A(H1N1)pdm09, and 1 (2%) was an unsubtypeable influenza A. Because the symptom severity score and other severity indicators did not significantly vary by influenza A or B type (data not shown), results were reported for influenza types combined. Participant characteristics (Table 2) were similar for those with influenza versus noninfluenza illnesses, with 2 exceptions. More influenza-positive participants lived in a household with children under 12 than influenza-negative participants (52% vs 36%; P p .03), and influenza-positive subjects were slightly less educated (mean [standard deviation (SD)], 15.4 [2.0] years) than influenza-negative subjects (16.2 [2.4] years; P p .01). Influenza and noninfluenza respiratory illnesses are com-

pared in Table 3. Days between illness onset and swab collection were similar for influenza versus noninfluenza illnesses (4.3 vs 4.7 days). At the time of specimen collection, participants with influenza were significantly more likely to have fever/feverishness (97% vs 83% with influenza-negative illnesses; P p .01), chills (91% vs 79%; P p .03), and wheezing (55% vs 35%; P ! .01). Influenza illnesses compared with noninfluenza illnesses were associated with higher summed 12-symptom severity score (mean [SD], 17.9 [5.4] vs 14.6 [4.8]; P ! .0001), worse subjective health during the illness (4.5 [1.8] vs 4.0 [1.8]; P p .04), greater disruption to activities of daily living (4.9 [2.5] vs 3.8 [2.5]; P ! .01), greater likelihood of missing work (67% vs 51%; P p .03), and more total hours of missed work (12.1 [10.5] vs 7.8 [10.5]; P ! .01). There were no significant differences between influenza and noninfluenza illnesses in the proportion with associated outpatient visits (29% vs 20%), influenza antiviral use (12% vs 7%), or duration of illness (mean, 11 days for both; Table 3). The summed severity scores increased depending on whether the episodes were influenza positive and medically attended (Figure 1; P ! .001 for trend). A similar trend was noted for hours of missed work, with non-MAARI noninfluenza illness associated with the fewest hours of missed work (mean [SD], 6.3 [8.9]) and MAARI influenza associated with the most missed work (17.8 [12.3]; P ! .001). In multivariate models that adjusted for correlates of influenza (education and child in household) and other participant and illness characteristics, influenza continued to be

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

544

infection control and hospital epidemiology

may 2014, vol. 35, no. 5

table 5. Mean (Standard Deviation [SD]) of Illness Severity Indicators for Vaccinated versus Unvaccinated Healthcare Personnel for Influenza and Noninfluenza Acute Respiratory Illnesses (ARIs) Influenza-positive ARIs

Influenza-negative ARIs

Vaccinated (N p 44) Unvaccinated (N p 14) Group Symptom severitya Subjective healthb Perform ADLsc Missed work, hoursd Duration of illness, days

17.6 4.4 4.7 10.9 11.2

A (5.5) (1.9) (2.6) (11.3) (8.7)

18.6 4.9 5.6 15.9 11.0

B (5.3) (1.4) (2.0) (13.9) (6.3)

P .56 .32 .27 .18 .94

Vaccinated (N p 176) Unvaccinated (N p 33) 14.4 3.8 3.6 7.1 11.3

C (4.8) (1.8) (2.4) (10.2) (4.9)

15.5 4.6 4.8 11.2 11.9

D (5.1) (1.7) (2.5) (11.5) (6.6)

P .24 .02 .01 .04 .61

note. Data are mean (SD), unless otherwise indicated. P value is associated with Student t test comparing means of vaccinated versus unvaccinated participants within influenza versus noninfluenza illness groups. ADL, activities of daily living. a Sum of ratings (0, absent; 1, mild; 2, moderate; 3, severe) for cough, fever/feverishness, chills, sore throat, nasal congestion, ear pain, fatigue, headache, muscle aches, nausea, and vomiting. b Participant’s rating of current health while ill (0, normal health; 9, worse possible health). c Participant’s rating of ability to perform activities during their illness (0, able to perform usual activities; 9, unable to perform usual activities). d Hours missed from work as a result of current illness.

significantly associated with higher symptom severity, impairment in activities of daily living, and missed work (Table 4). Medically attended illnesses were also associated with activities of daily living impairment and missed work in the multivariate models. Receipt of seasonal influenza vaccination was high among both influenza-positive and influenza-negative participants (76% and 85%, respectively). Vaccinated and unvaccinated individuals with influenza illnesses did not significantly differ on any severity indicators, though the direction of the effect was consistently toward lower severity among vaccinees (Table 5). Vaccinated individuals with influenza-negative ARIs reported better subjective health during their ARI, less impairment to activities of daily living, and missed fewer hours of work than unvaccinated individuals with influenza-negative ARIs (Table 5).

discussion Influenza illness was associated with higher self-rated symptom severity, greater disruption to activities of daily living, and more missed work than influenza-negative ARIs in this population of HCP. We observed these differences even after adjusting for sociodemographic, health, and illness characteristics. These findings add to a growing literature that illness due to influenza virus infection is moderately more severe than illnesses due to other respiratory pathogens8,14 and reinforce the importance of influenza prevention and control efforts among HCP.5 Overall, our findings are consistent with prior studies of influenza illness. We observed more fever/feverishness, chills, and wheezing among the influenza illnesses, similar to the Belongia et al8 study comparing A(H1N1)pdm09 virus infections with noninfluenza respiratory illness in adults recruited in a community-based setting. We found a mean 12symptom severity score of 18 (out of 36) for influenza

illnesses and 15 for noninfluenza illnesses, which was also similar to the Belongia et al8 findings using the same scale. In our HCP cohort, 29% of PCR-confirmed influenza episodes were associated with an outpatient visit, similar to previous estimates of 31% among healthy adults.3 One of the major economic impacts of influenza is lost productivity from missed work days.3 Our findings suggest that among HCP, these losses may be higher than previously estimated. Molinari et al,3 for example, assumed that medically attended influenza illness was associated with a single 8-hour day of lost productivity per outpatient visit, while non–medically attended influenza was associated with onehalf day of lost productivity. However, in our study, HCP with influenza illness missed an average of 12 hours of work, and those with medically attended influenza missed an average of 18 hours of work, suggesting that losses of productivity may be 1.5–2 times what was previously assumed. This may be due in part to the implementation of stricter workplace policies around HCP working when ill. Illnesses of this severity among HCP also have downstream healthcare consequences, such as staffing shortages and delays in patient care.5 Of concern was the fact that one-third of HCP who work full time and have direct patient contact did not miss work during their influenza illness, despite the fact that the average influenza illness lasted 11 days. Although our study was not designed to examine or quantify secondary exposure of influenza to patients, our findings support previous observations that HCP often continue to work despite being ill with influenza and may serve as a potential source of exposure to patients and coworkers.5 It is important to maintain a workplace atmosphere where ill HCP are not penalized but encouraged to stay home until symptoms have improved. Current workplace policies at the study sites require employees to stay home for 24 hours after fever subsides and decreasing

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

influenza in hcp

cough. After returning, HCP are instructed to follow respiratory etiquette (covering cough or wearing a surgical mask if coughing persists) and be vigilant about hand hygiene. All of these potential impacts of influenza highlight the importance of influenza prevention and control measures among HCP.5 We observed a high vaccination rate as a result of strong institutional support for employee influenza vaccination programs that provided vaccines free of charge; emailed and printed media messages; and utilized peer vaccinators, walk-in flu shot clinics, and roving nurses with vaccine carts. Outpatients with influenza-like illnesses are encouraged to wear standard surgical masks, and droplet precautions are taken with inpatients. Nonetheless, our findings illustrate that some vaccinated HCP will become infected with influenza, despite these efforts. Antiviral medications, which may reduce complications of influenza infection and shorten the duration of symptoms,15-17 were reported to be prescribed infrequently (12%) for influenza illnesses in our cohort. Taken together, our findings highlight why ill HCP should follow workplace recommendations and stay home to avoid transmission of influenza to patients and coworkers.18 Further research is needed to examine whether vaccination may modify influenza illness severity. We did not observe statistically significant differences in influenza illness severity between vaccinated and unvaccinated individuals. Although we observed trends in this direction, our findings suggest that such differences may be small to moderate in magnitude and will require larger samples to detect. The significantly lower illness severity we observed among vaccinated individuals with noninfluenza illness suggest that future investigations will need to account for differences between vaccinated and unvaccinated individuals that may further complicate these associations. Strengths of the current study include its focus on HCP at medical centers in 2 geographic regions and its use of medical records to document vaccination status and chronic medical conditions. We also used active surveillance techniques, including weekly Internet symptom surveys, calls from staff if surveys were missed, and daily scanning of medical visits for MAARI diagnoses. By linking our surveillance to participants’ health records, we were also able to compare illness episodes with and without associated medical visits. Our study also has several limitations. We enrolled only 20%–40% of eligible HCP. The vaccination rate of participants was higher than that reported by employee health at Scott and White Healthcare (71%) and Kaiser Permanente Northwest (64%), though we targeted employees with direct contact with patients who typically have higher vaccination rates. By focusing on relatively healthy, working adult HCP, we may have limited variability in illness impact and thus may have underestimated differences between influenza and noninfluenza illness for outcomes such as duration of illness. We examined a relatively small number of influenza cases during a single influenza season, using screening criteria that required febrile illness with cough, which limited our ability

545

to evaluate differences between influenza (sub)types, examine nonspecific illness presentations, or examine more serious illness outcomes, such as hospitalization. Although we captured and quantified missed work, we did not measure patient exposure among those who did not miss work during their illness episodes. We also did not ask about the duration of fever, which is explicitly mentioned in the Centers for Disease Control and Prevention guidelines for returning to work (24 hours after fever subsides).18 In conclusion, HCP reported that influenza illnesses were more severe and disruptive than noninfluenza illnesses. The higher than expected hours of missed work due to influenza and the implications that 1-in-3 HCP continued to work while ill reinforce the importance of multifaceted influenza prevention, therapy, and control measures in healthcare settings.

acknowledgments Financial support. This work was supported by the Centers for Disease Control and Prevention (contract 200-2010-F-33396 to Abt Associates). The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention, Abt Associates, Kaiser Permanente Center for Health Research, or Scott and White Healthcare. Potential conflicts of interest. A.L.N. reports that she has received research funding from GlaxoSmithKline. M.J.G. reports that she has received research funding from Novartis and MedImmune. All other authors report no conflicts of interest relevant to this article. All authors submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest, and the conflicts that the editors consider relevant to this article are disclosed here. Address correspondence to Allison L. Naleway, PhD, Center for Health Research, Kaiser Permanente Northwest, 3800 North Interstate Avenue, Portland, OR 97227 ([email protected]).

references 1. Sullivan KM, Monto AS, Longini IM Jr. Estimates of the US health impact of influenza. Am J Public Health 1993;83(12): 1712–1716. 2. Kuster SP, Shah PS, Coleman BL, et al. Incidence of influenza in healthy adults and healthcare workers: a systematic review and meta-analysis. PLoS ONE 2011;6(10):e26239. 3. Molinari NA, Ortega-Sanchez IR, Messonnier ML, et al. The annual impact of seasonal influenza in the US: measuring disease burden and costs. Vaccine 2007;25(27):5086–5096. 4. Pearson ML, Bridges CB, Harper SA. Influenza vaccination of health-care personnel: recommendations of the Healthcare Infection Control Practices Advisory Committee (HICPAC) and the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2006;55(RR-2):1–16. 5. Poland GA, Tosh P, Jacobson RM. Requiring influenza vaccination for health care workers: seven truths we must accept. Vaccine 2005;23(17/18):2251–2255. 6. Hollmeyer HG, Hayden F, Poland G, Buchholz U. Influenza vaccination of health care workers in hospitals: a review of studies on attitudes and predictors. Vaccine 2009;27(30):3935–3944. 7. Hofmann F, Ferracin C, Marsh G, Dumas R. Influenza vacci-

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

546

8.

9.

10.

11.

12.

13.

infection control and hospital epidemiology

may 2014, vol. 35, no. 5

nation of healthcare workers: a literature review of attitudes and beliefs. Infection 2006;34(3):142–147. Belongia EA, Irving SA, Waring SC, et al. Clinical characteristics and 30-day outcomes for influenza A 2009 (H1N1), 2008–2009 (H1N1), and 2007–2008 (H3N2) infections. JAMA 2010; 304(10):1091–1098. Harper SA, Bradley JS, Englund JA, et al. Seasonal influenza in adults and children—diagnosis, treatment, chemoprophylaxis, and institutional outbreak management: clinical practice guidelines of the Infectious Diseases Society of America. Clin Infect Dis 2009;48(8):1003–1032. Thompson MG, Gaglani MJ, Naleway A, et al. The expected emotional benefits of influenza vaccination strongly affect preseason intentions and subsequent vaccination among healthcare personnel. Vaccine 2012;30(24):3557–3565. Melia M, O’Neill S, Calderon S, et al. Development of a flexible, computerized database to prioritize, record, and report influenza vaccination rates for healthcare personnel. Infect Control Hosp Epidemiol 2009;30(4):361–369. Singh-Manoux A, Martikainen P, Ferrie J, Zins M, Marmot M, Goldberg M. What does self rated health measure? results from the British Whitehall II and French Gazel cohort studies. J Epidemiol Community Health 2006;60(4):364–372. Fiore AE, Uyeki TM, Broder K, et al. Prevention and control

14.

15.

16.

17.

18.

of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP), 2010. MMWR Recomm Rep 2010;59(RR-8):1–62. Irving SA, Patel DC, Kieke BA, et al. Comparison of clinical features and outcomes of medically attended influenza A and influenza B in a defined population over four seasons: 2004– 2005 through 2007–2008. Influenza Other Respir Viruses 2012; 6(1):37–43. Hernan MA, Lipsitch M. Oseltamivir and risk of lower respiratory tract complications in patients with flu symptoms: a meta-analysis of eleven randomized clinical trials. Clin Infect Dis 2011;53(3):277–279. Hsu J, Santesso N, Mustafa R, et al. Antivirals for treatment of influenza: a systematic review and meta-analysis of observational studies. Ann Intern Med 2012;156(7):512–524. Jefferson T, Jones MA, Doshi P, et al. Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children. Cochrane Database Syst Rev 2012;1:CD008965. Centers for Disease Control and Prevention (CDC). Prevention Strategies for Seasonal Influenza in Healthcare Settings: Guidelines and Recommendations. Atlanta: CDC, 2013. http://www.cdc.gov /flu/professionals/infectioncontrol/healthcaresettings.htm. Accessed June 11, 2013.

This content downloaded from 85.64.25.112 on Tue, 3 Jun 2014 08:02:10 AM All use subject to JSTOR Terms and Conditions

Comparison of laboratory-confirmed influenza and noninfluenza acute respiratory illness in healthcare personnel during the 2010-2011 influenza season.

Compare the severity of illnesses associated with influenza and noninfluenza acute respiratory illness (ARI) in healthcare personnel (HCP)...
257KB Sizes 0 Downloads 3 Views