Overall, 91.8% (617/672) of households that expressed interest in participation met eligibility criteria; 82.5% (509/617) were reachable and completed the initial home visit. The mean duration of the households in the study was 55.5 weeks with a 13.0% (66/509) dropout rate. The 509 participating households included a total of 2,788 members (). The majority of household members were Latino (96.2%, 2,682/2,788), born outside the U.S. (54.0%, 1,500/2,776), and 18 years of age or older (52.7%, 1,456/2,763). Most (90.4%, 2,448/2,708) reported having no chronic respiratory disease. The mean household size was 4.5 people per one-bedroom apartment. Households randomized to the Education (n=174), Hand Sanitizer (n=169), or Hand Sanitizer and Face Mask (n=166) group were comparable in terms of mean household size (five members, range: 3–14), gender, ages, and proportion born outside the U.S.
Number of households contacted, recruited, and randomized in a clinical trial of non-pharmaceutical interventions for URIs, New York, November 2006 to July 2008
About one-third of working individuals were either homemakers or unemployed, 28.3% were employed in service industries (e.g., food preparation, health care, education, and childcare), and 38.3% were employed in other settings (e.g., construction, transportation, or sales). Overall, 44.6% of the children attended a public or private school, 31.5% were routinely cared for at home, and others were in some form of daycare. The Education group included significantly more household members of Latino descent (98.1%) than the other two groups (94.2% for the Hand Sanitizer group and 96.4% for the Hand Sanitizer and Face Mask group) (p<0.005), as well as members who had not completed high school (54.6% vs. 44.7% in the Hand Sanitizer group and 39.8% in the Hand Sanitizer and Face Mask group (p<0.005). Significantly more individuals spent at least 40 hours per week outside the home in the Hand Sanitizer and Face Mask group (42.6%) as compared with the Education (33.8%) or Hand Sanitizer (32.3%) groups (p<0.005) ().
Summary of characteristics of household member participants in a study of non-pharmaceutical interventions for URIs, by intervention group, New York, November 2006 to July 2008
Incidence of URIs, ILIs, and confirmed influenza
A total of 5,034 URI symptoms were reported, most commonly rhinorrhea or cough. About 83.3% (424/509) of households had at least one member with one or more symptoms, but 48.6% (1,355/2,788) of members had no reported symptoms. Households in the Hand Sanitizer group included significantly more members without any reported symptoms (57.6% as compared with 49.4% in the Education group and 38.7% in the Hand Sanitizer and Face Mask group, p<0.01). summarizes unadjusted rates of household members with no reported symptoms; URI, ILI, and influenza; change scores on the KAP survey; and compliance with symptom reporting (≥75% of required time) by intervention group.
Outcomes of household members in a study of non-pharmaceutical interventions for URIs, New York, November 2006 to July 2008
Households reported 669 episodes of ILI (0–5 per individual). Of the 234 deep nasal swabs obtained, 33.3% (n=78) tested positive for influenza; 43.6% (n=34) were influenza A and 56.4% (n=44) were influenza B. Among the 66.7% who tested negative for influenza, 30.8% (48/156) tested positive for other viruses: seven for respiratory syncytial virus, nine for parainfluenza, 11 for enterovirus, 10 for rhinovirus, six for adenovirus, and five for metapneumovirus. Swabs were not obtained from the remaining 435 reported ILI episodes for the following reasons: 72.0% (n=313) did not meet the CDC definition of an ILI and were therefore included in the URI symptom count, 21.4% of episodes (n=93) were reported after 48 hours of ILI onset or the participant refused to be swabbed, and the research staff were unable to reach the participant in 6.7% of episodes (n=29).
Based on the Poisson GEE analysis, people born in the U.S. had approximately 1.5 more URI episodes than those born outside the U.S. (mean of 2.3 and 1.4 episodes per person, respectively, p=0.004), younger age was significantly associated with higher rates (p<0.001, data not shown), and those with chronic respiratory illness had about 1.4 times more URI episodes than those without respiratory problems (2.5 and 1.8 episodes per person, respectively, p=0.009). Men were significantly less likely to have both URIs and ILIs than women. The odds of getting influenza were 5.16 times higher for college graduates than for those with less than a high school education, even when adjusting for the number of hours each week spent outside the home. The odds of getting influenza were 2.56 times higher for homemakers and those who were unemployed vs. those working in other professions. However, there were no significant differences in rates of URI, ILI, or influenza by intervention group ().
Regression coefficients and p-values for outcomes from GEE logistic models (for influenza) and GEE Poisson models (for ILI and URIs) in a study of non-pharmaceutical interventions for URIs, New York, November 2006 to July 2008
Secondary attack rates
We used a total of 3,463 episodes of URI, ILI, or influenza with complete data for this analysis, yielding a total of 1,751 secondary cases. presents the means and standard deviations of the secondary attack rate in each of the three intervention groups. The mean secondary attack rates were 0.137 for the Education group, 0.144 for the Hand Sanitizer group, and 0.124 for the Hand Sanitizer and Face Mask group. There were 323 episodes of ILI and influenza, resulting in 29 secondary cases. The mean secondary attack rates for the Education, Hand Sanitizer, and Hand Sanitizer and Face Mask groups were 0.023, 0.020, and 0.018, respectively.
Secondary attack rates of influenza, ILI, and URIs among participants in a study of non-pharmaceutical interventions for URIs, by intervention group (unadjusted), New York, November 2006 to July 2008
Regarding URI, ILI, and influenza episodes, there was a significant decrease in secondary attack rates in the Hand Sanitizer and Face Mask group when compared with the Education group. Regarding the other significant explanatory variables, crowding had a negative association with the secondary attack rate (p<0.0001) (i.e., more crowded households had less transmission). To confirm that the significant effect of crowding was not a result of confounding, we ran the logistic regression model on the data from all URI, ILI, and influenza episodes with all the covariates listed in the Methods section. We found that the effect of crowding was still significant (p<0.0001), with an odds ratio of 0.80 (95% confidence interval 0.72, 0.89). In addition, secondary attack rates were significantly lower when the index case was 0–5 years of age and significantly higher when the index case was 6–17 years of age when compared with adult index cases. For the ILI and influenza episodes, secondary attack rates were again significantly lower when the index case was 0–5 years of age as compared with episodes in which the index case was at least 18 years of age ().
Final logistic regression models summarizing significant predictors of secondary attack rates for influenza, ILI, and URIs among participants in a study of non-pharmaceutical interventions for URIs, New York, November 2006 to July 2008
Adherence to study protocols
The Hand Sanitizer and Face Mask group was more compliant with weekly reporting of symptoms (80.7%) as compared with the Hand Sanitizer (75.7%) and Education (65.5%) groups (p=0.005) (). Members of the Hand Sanitizer group used a mean of 12.1 ounces/month and the Hand Sanitizer and Face Mask group used a mean of 11.6 ounces/month (including counted empty bottles and self-reports) (p=0.36). At the exit survey, 44.2% (65/147) of households from the Education group reported using hand sanitizer occasionally at some point during the study, and 56.9% of these (37/65) reported using hand sanitizer one to two times within the previous 24 hours. Compliance with mask use was poor. Although households were instructed about mask use, and reinforcement by the RA and Project Manager occurred frequently, only half (22/44) of the households with an ILI reported using masks within 48 hours of episode onset. Those who used masks at all reported a mean of only two masks/day/ILI episode (range: 0–9).
Respondent knowledge of prevention and treatment strategies
A total of 441 households completed both a baseline and exit interview. The KAP scores at baseline were generally low (the mean in each of the three intervention groups was slightly more than 50% correct out of 10 items), but improved significantly in all groups by the end of the study (5.12 to 5.75 in the Education group, 5.48 to 7.24 in the Hand Sanitizer group, and 5.11 to 6.40 in the Hand Sanitizer and Face Mask group). The change in KAP scores was significantly greater for the Hand Sanitizer group as compared with the other two groups (p<0.0001). There were no significant interaction effects among covariates such as education, occupation, and group on KAP scores.
Influenza vaccination rates
There was an increase between the baseline and exit interview in all three groups that reported ≥50% of members receiving influenza vaccine (pre- vs. post-intervention for each group: 21.1% and 40.8% in the Education group, 19.0% and 57.1% in the Hand Sanitizer group, and 22.4% and 43.5% in the Hand Sanitizer and Face Mask group (p<0.001). Additionally, those in the Hand Sanitizer group reported a significantly greater increase than the other two groups, controlling for baseline rates (p=0.002).