Self-collected nasal swabs to detect infection and colonization: a useful tool for population-based epidemiological studies?

Perspective Self-collected nasal swabs to detect infection and colonization: a useful tool for population-based epidemiological studies? M.K. Akmatov a, *, F. Pessler a,b a Department of Infection Genetics, Project Group Epidemiology, Helmholtz Centre for Infection Research, Inhoffenstraße 7, D-38124 Braunschweig, Germany b University Children’s Hospital, Division of Rheumatology and Immunology, Technical University Dresden, Dresden, Germany 1. Introduction Why are there so few large population-based epidemiological studies on risk factors for infectious diseases? Indeed, infectious diseases have been all but excluded from the large populationbased cohort studies that are currently in various stages of evolution worldwide. This is the more surprising since environmental, lifestyle, and host genetic factors clearly contribute to acquisition, transmission, clinical course, and outcome of essentially all infectious diseases known to date. Host genetic factors in particular may play important roles in many acute infectious diseases, as exemplified by influenza 1,2 and norovirus infection. 3 This paucity of population-based approaches to infectious diseases/infections may be explained, at least in part, by methodological limitations. In contrast to the common chronic diseases that are usually featured in population-based studies (e.g., cardiovascular, metabolic, neoplastic, or neurodegenerative diseases), infectious diseases often follow an acute course and the pathogen may only be detectable during a narrow time-window. For instance, maximal shedding of influenza virus occurs in the first 3 days of infection in humans, and the rate of viral detection by PCR declines rapidly after 5 days. 4 Moreover, infectious diseases may follow a mild or even asymptomatic course, and some individuals may be colonized with microbial pathogens (e.g., Staphylococcus aureus) without displaying any symptoms. Thus, healthcare utilization by these individuals is low and their identification in the context of population-based studies is difficult. Consequently, research on infectious diseases is often hospital-based, resulting in an oversampling of diseases characterized by comparatively high morbidity and healthcare utilization. Therefore, different methods of sample collection are needed to identify asymptomatic carriers or individuals with acute infections not severe enough to necessitate a visit to a medical care provider. As exemplified by studies on influenza infection, a trained member of the field team may collect the diagnostic specimen by performing a nasal swab on the study participant during a house call. 5 However, for this approach to function, the subjects need to notify the study centre promptly and reliably at the onset of symptoms. An additional drawback is that the cost of this approach skyrockets when large numbers of incident cases need to be identified, e.g., for a genome-wide association study (GWAS). An attractive alternative approach would be to ask the participants to perform the swabs themselves (‘self-swabbing’). There is ample evidence from various clinical scenarios demonstrating the feasibility of self-swabbing among specific risk groups and patient populations. 6–10 Self-swabbing has several advantages over the collection of samples by study personnel. First, incident cases of acute infections that feature a short duration or mild symptoms can be identified more easily. Thus, this method may improve identification of incident cases and thereby reduce disease misclassification. Second, a larger number of study subjects can be included in a given study due to the lower cost and simpler logistics. For International Journal of Infectious Diseases 15 (2011) e589–e593 A R T I C L E I N F O Article history: Received 17 November 2010 Received in revised form 1 March 2011 Accepted 22 April 2011 Corresponding Editor: J. Peter Donnelly, Nijmegen, the Netherlands. Keywords: Self-collected nasal swabs Bacterial and viral infections Epidemiological studies S U M M A R Y Population-based epidemiological studies on infectious diseases are limited by methodological problems that may not be encountered in other fields of epidemiology. The acute or asymptomatic nature of many infections hinders a timely diagnosis by trained personnel in a study centre, indicating the need for new collection methods of biological specimens. One alternative approach is to have the participants collect the specimens themselves, for instance nasal swabs for the detection of bacterial or viral pathogens. Although self-collection is widely accepted in clinical studies of specific populations (e.g., self-collection of vaginal swabs by young women to diagnose sexually transmitted infections), it has not been employed much in population-based studies. Here, we review recent experience with selfcollection of nasal swabs for the detection of microorganisms and discuss future prospects and applications for this technique. ß 2011 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved. * Corresponding author. Tel.: +49 0 531 6181 1112; fax: +49 0 531 6181 1199. E-mail address: [email protected] (M.K. Akmatov). Contents lists available at ScienceDirect International Journal of Infectious Diseases jou r nal h o mep ag e: w ww .elsevier .co m /loc ate/ijid 1201-9712/$36.00 – see front matter ß 2011 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijid.2011.04.009 example, for a planned GWAS on genetic factors of influenza infection (detected in nasal swabs) in an urban/suburban population, we compared the cost of swabbing by study personnel to the cost of self-swabbing. In order to detect an odds ratio of 1.5 (assuming 10% clinical attack rate, 80% power, and 5  10 8 type 1 error accounting for multiple hypothesis testing), 1000 cases would be needed. Expecting an infection rate of approx. 25%, a discovery sample of n = 4000 participants would be needed to yield 1000 cases. Assuming an average of two acute respiratory infections (ARI) per individual per year, 11 approx. 8000 home visits at a cost of at least 20 euros/visit (personal communication, Thomas Behrens, Bremen Institute for Prevention Research and Social Medicine; based on current cost for personnel time and travel in Germany) would be required, costing upward of 160 000 euros. Including participants from rural or remote areas would result in even higher costs due to higher travel expenses. On the other hand, the cost of self-collection of swabs would be approx. 32 000 euros, mostly for packaging and postage. Here, we review the literature on self-collection of nasal swabs for the detection of microorganisms and discuss future prospects and applications of this promising new tool for infectious disease field research. 2. State of the art 2.1. Sample collection in an unsupervised setting A summary of all studies that have used self-collected nasal swabs is presented in Table 1. 4,12–20 Only two studies were truly population-based and featured self-collection outside of a study centre. 4,19 van Cleef et al. examined the prevalence of livestockassociated methicillin-resistant Staphylococcus aureus (MRSA) in the Netherlands. 19 In this study, swabbing supplies were sent by mail to individuals who had agreed to participate in the study (n = 644). Approximately 90% returned the swabs, indicating that self-collection may be highly feasible in the general population. In another population-based study of respiratory infections, parents were asked to collect nasal and pharyngeal swabs from their children. 4 The main outcome in this study was the proportion of swabs that were positive for any one of seven respiratory pathogens. By comparing results for (1) nasal and throat swabs and (2) swabs collected by parents who worked in healthcare vs. parents who did not, it was found that nasal swabs were more often positive than pharyngeal swabs, but that there was no difference in the proportion of positive swabs when comparing Table 1 Summary of published studies using self-collected nasal swabs Author/year Study design Study population Total sample size Outcome Self-swabbing Most relevant results Gilbert et al. 2007 18 Cross-sectional (patient-based) Adults (marginalized urban population, e.g., drug users) 271 MRSA colonization Self-collected nasal swabs MSSA was detected in 95 of 271 (36%) swabs, any MRSA strain in 13 (4.8%), and MRSA USA300 in 10 (3.7%) Cooper et al. 2008 16 Cross-sectional (patient-based) Adults 296 Influenza virus detection Self-collected nasal swabs 142 of 296 (48%) swabs were returned by mail No difficulties in self-swabbing reported Mean time between the phone call and self-swabbing 7.4 days No differences in mean times between positive and negative samples Lambert et al. 2008 4 Cohort study (community-based) Children 234 Detection of respiratory viruses Parent-collected nasal swabs Higher positivity rates of nasal samples for any virus (78%) compared to samples from throat (71%) No differences in the positivity rates between parents who did and did not work in healthcare Time from symptom onset to selfswabbing of less than 5 days irrespective of time from selfswabbing to laboratory analysis did not influence the positivity rate Time from symptom onset to selfswabbing of more than 6 days combined with time from selfswabbing to laboratory analysis of 2 and more days decreased the positivity rate High acceptance (e.g., 87% of participants agreed to participate in further studies) Difficulties with throat swab (58%) Lu et al. 2008 15 Convenience sample of laboratory staff Adults 4 Rhinovirus Serial self-collected nasal swabs Samples were used to develop a real-time PCR assay for detection of human rhinoviruses Elliot et al. 2009 13 Cross-sectional (patient-based) Adults 3129 Influenza surveillance Self-collected nasal swabs 1783 swabs of 3129 (57%) were returned by mail 1076 swabs of 1346 (80%) were received within 7 days of symptom onset M.K. Akmatov, F. Pessler / International Journal of Infectious Diseases 15 (2011) e589–e593 e590 parental profession (healthcare vs. non-healthcare workers). These results are a first indication that – at least in this setting of parental swabbing – lay persons can collect swabs as efficiently as healthcare workers. The viral detection rate remained constant as long as the time that elapsed between the onset of illness and collection of the specimen was 5 days or less. However, it decreased when the time from illness onset to laboratory analysis exceeded 8 days, underscoring the importance of timely return of self-collected swabs to the study centre. Self-sampling has also been found to improve symptomatic surveillance by providing additional laboratory-based information. 13,16 For example, during the 2004–2005 influenza seasons, the UK National Health Service established a telephone hotline (‘NHS Direct’). 16 Callers who voiced a concern of suffering from an influenza-like illness were sent swabbing supplies by mail and were asked to self-collect a nasal swab and mail it to a central reference laboratory. Most study participants performed the swab and returned it within the requested time: the time intervals between the phone call and self-swabbing and between the phone call and laboratory analysis were 4.1 and 7.4 days, respectively. The majority of participants reported no difficulties with selfsampling. 2.2. Sample collection in a supervised setting In a recent patient-based study of influenza infection in children, researchers compared the efficiency of nasal swabs collected by pediatricians and parents. 14 Sensitivity (89%) and specificity (98%) of parent-collected swabs (compared to pediatrician-collected swabs) for the detection of influenza virus by realtime PCR were high, and the mean viral copy number per positive swab did not differ between parentand pediatrician-collected samples. The parents reported that the children were more satisfied when nasal swabs were taken by the parents than by the pediatricians. 14 In a very recent study, a newly developed flocked nasal swab was evaluated for the detection of respiratory infections. 17 Initially, symptom-free participants were asked to collect swabs from one nostril. At the same time, swabs were collected by study personnel with a rayon nasopharyngeal swab (gold standard) from the other nostril. The self-collected nasal swabs yielded better results than the gold standard (based on a higher epithelial cell count and detection of higher levels of human b -actin gene coding sequences, which were used as proxy endpoints). Furthermore, respiratory viral infections were detected in symptomatic subjects; among the 108 symptomatic subjects any one of several ARI was detected in 42 self-collected swabs (39%). However, this endpoint was not compared with the gold standard method. The majority of subjects (87% of 55) had no or minor difficulties with self-swabbing. Sixty-five percent of subjects reported no or only minor discomfort from self-swabbing, 31% reported moderate discomfort, and 4% severe discomfort. At first glance, nasal swabbing does not appear to be a technically demanding procedure, but specific aspects such as the applied pressure and the number of revolutions during swabbing, the depth of insertion toward the turbinates, or even which nostril to probe, are all parameters that could potentially affect the efficiency of detection. Hence, it must be noted that the study on parent-collected swabs is the only one where the detection rate of self-collected nasal swabs was compared directly to samples obtained by trained personnel. 14 However, parents collected the swabs in the presence of study personnel and may have felt more confident, resulting in better swabbing. In the above-mentioned study by Smieja et al., 17 the number of nasal epithelial cells and the Table 1 (Continued ) Author/year Study design Study population Total sample size Outcome Self-swabbing Most relevant results Loeb et al. 2009 12 Randomized trial (convenience sample of hospital staff) Adults 446 Influenza virus transmission Self-collected nasal swabs Comparison of a surgical mask (group 1) and an N95 respirator (group 2) to prevent transmission of influenza A or B virus Influenza virus was detected in 2.8% of the 212 swabs from group 1 vs. 1.9% of the 210 swabs from group 2 Other respiratory viruses were detected in 9.4% (group 1) and 10.5% (group 2) of the swabs Esposito et al. 2010 14 Cohort study (patient-based) Children 203 Influenza virus detection Parent-collected nasal swabs High sensitivity (89%) and specificity (98%) of the selfcollected swabs (gold standard: pediatrician-collected swabs) Higher satisfaction of children with parentthan pediatrician-collected swabs No difference in virus detection rate between parentand pediatriciancollected swabs Smieja et al. 2010 17 Cross-sectional study (convenience sample of hospital staff and visitors) Adults 250 Respiratory viral infections Self-collected nasal swabs 108 participants of 250 (44%) selfswabbed within 4 days of symptom onset and returned the swabs within 5 days Van Cleef et al. 2010 19 Cross-sectional study (community-based) Adults 644 MRSA colonization Self-collected nasal swabs 583 swabs of 644 (91%) were returned by mail Ammons et al. 2010 20 Cross-sectional study (convenience sample of student population) Young adults 375 a MRSA colonization Self-collected nasal swabs Staphylococcus aureus was detected in 7 of 375 swabs (1.9%); 6 of which were MRSA MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus. a Number of swabs. M.K. Akmatov, F. Pessler / International Journal of Infectious Diseases 15 (2011) e589–e593 e591 presence of DNA sequences encoding the human b -actin gene were used to quantify host tissue in which viruses or other pathogens might be detected. However, the viral detection rate was not compared between self-collected and personnel-collected swabs. Thus, the diagnostic equivalence of self-collected swabs by adults has not been established formally. We are currently conducting such a study, in which self-collected and trained personnel-collected nasal swabs are being compared in terms of their accuracy in the detection of viral respiratory pathogens. In two studies, nasal swabs were self-collected by laboratory 15 and hospital staff. 12 In these selected populations, self-swabbing should be feasible since understanding of the swabbing procedure and compliance with its technical aspects should be good. The above-mentioned studies indicate that self-swabbing was highly acceptable among the studied populations. It also appears to be a feasible method for the diagnosis of various microorganisms, as long as the time intervals from onset of symptoms to sample collection and then to arrival in the laboratory, are monitored carefully. Further validation studies are needed to investigate the efficacy of self-collected nasal swabs to detect respiratory pathogens, particularly by comparing self-swabbing with the respective gold standard method, e.g. staff-collected nasal swabs. 3. Challenges The above-mentioned studies show that self-swabbing may be a viable and cost-effective alternative method for the purpose of collecting nasal swabs for epidemiological studies. However, there are some challenges that have to be considered prior to using this method. First, an inadequate sampling technique will undoubtedly compromise the value of any laboratory analysis. Even though it appears obvious that study participants should be invited once to a study centre where study personnel can explain and demonstrate the correct self-swabbing technique, it has not yet been determined whether this is really necessary. In countries with wellfunctioning mail delivery systems, swabbing supplies have also been provided by mail. 13,16,19 If this approach is to be used, participants should be provided with detailed printed instructions on how to obtain the sample, including simple visual material. Second, to achieve timely and reliable sample collection (which is particularly important in the detection of ARI), participants should receive reminders on a regular basis. Various reminder systems can be used, such as telephone calls, e-mail, or SMS (short message service; text). Third, logistic problems can arise if self-swabbing is to take place outside of a study centre (e.g., at home) and the swabs are to be returned by the participants in person or by mail. In the abovementioned studies, the swabs were returned by mail, with return rates varying between 48% 16 and 91%. 19 The results of laboratory analyses will depend on the time elapsed between self-swabbing and laboratory analysis; an early delivery of swabs may increase the sensitivity and specificity, but this will vary with the organism to be detected. As mentioned above, when a panel of pathogenic viruses, mostly consisting of RNA viruses, was used, the detection rate for any virus decreased if more than 8 days elapsed between symptom onset and laboratory analysis. 4 Conceivably, this was due to nucleic acid degradation during transport. Interestingly, in a study of self-collected vaginal swabs (in which e-mail was used successfully as a reminder tool), no effect of transport time on DNA quantity was detected, perhaps due to the inherent higher stability of DNA compared with RNA. It therefore appears important to evaluate nucleic acid preservatives (e.g., RNAlater 1 , RNAprotect 1 , and others) as alternate collection and transport media for scenarios in which RNA stability would be a concern, and culture (requiring viable organisms) is not planned. Use of such media would be particularly important in studies in which gene expression patterns of entire microbial populations (microbiomes) are to be determined. 4. Conclusions and prospects for the future Self-collection of nasal swabs may be a cost-efficient method for the detection of microorganisms in population-based epidemiological studies. By using this method one may increase the size of the study population, thus facilitating investigations on complex associations between infectious diseases/infections and various risk factors. In the small number of studies in which self-swabbing was employed, it was found to be safe and was met with a high degree of acceptance and satisfaction on the part of the participants. However, it has not been established whether the detection rates for various microorganisms from self-collected swabs and from swabs obtained by trained personnel are comparable. Thus, further validation studies are needed before the use of self-swabbing can be advocated without reservations. Acknowledgements We thank Dr. Thomas Ja ¨nisch (Heidelberg University Hospital), Prof. Lothar Kreienbrock (University of Veterinary Medicine Hannover) and Prof. Klaus Schughart (Helmholtz Centre for Infection Research, Braunschweig) for critical readings of the manuscript and helpful discussions. This work was supported by intramural funding from the Helmholtz Association (Program Infection and Immunity). Conflict of interest: No conflict of interest. References 1. Albright FS, Orlando P, Pavia AT, Jackson GG, Cannon Albright LA. Evidence for a heritable predisposition to death due to influenza. J Infect Dis 2008;197:18–24. 2. Trammell RA, Toth LA. Genetic susceptibility and resistance to influenza infection and disease in humans and mice. Expert Rev Mol Diagn 2008;8:515–29. 3. Hill AV. Aspects of genetic susceptibility to human infectious diseases. Annu Rev Genet 2006;40:469–86. 4. Lambert SB, Allen KM, Nolan TM. Parent-collected respiratory specimens—a novel method for respiratory virus and vaccine efficacy research. Vaccine 2008;26:1826–31. 5. Cowling BJ, Chan KH, Fang VJ, Lau LL, So HC, Fung RO, et al. Comparative epidemiology of pandemic and seasonal influenza A in households. N Engl J Med 2010;362:2175–84. 6. Alexander S, Ison C, Parry J, Llewellyn C, Wayal S, Richardson D, et al. Self-taken pharyngeal and rectal swabs are appropriate for the detection of Chlamydia trachomatis and Neisseria gonorrhoeae in asymptomatic men who have sex with men. Sex Transm Infect 2008;84:488–92. 7. Garland SM, Tabrizi SN. Diagnosis of sexually transmitted infections (STI) using self-collected non-invasive specimens. Sex Health 2004;1:121–6. 8. van der Helm JJ, Hoebe CJ, van Rooijen MS, Brouwers EE, Fennema HS, Thiesbrummel HF, et al. High performance and acceptability of self-collected rectal swabs for diagnosis of Chlamydia trachomatis and Neisseria gonorrhoeae in men who have sex with men and women. Sex Transm Dis 2009;36:493–7. 9. Holland-Hall CM, Wiesenfeld HC, Murray PJ. Self-collected vaginal swabs for the detection of multiple sexually transmitted infections in adolescent girls. J Pediatr Adolesc Gynecol 2002;15:307–13. 10. Winer RL, Feng Q, Hughes JP, Yu M, Kiviat NB, O’Reilly S, et al. Concordance of self-collected and clinician-collected swab samples for detecting human papillomavirus DNA in women 18 to 32 years of age. Sex Transm Dis 2007;34: 371–7. 11. Monto AS. Epidemiology of viral respiratory infections. Am J Med 2002; 112(Suppl 6A):4S–12S. 12. Loeb M, Dafoe N, Mahony J, John M, Sarabia A, Glavin V, et al. Surgical mask vs N95 respirator for preventing influenza among health care workers: a randomized trial. JAMA 2009;302:1865–71. 13. Elliot AJ, Powers C, Thornton A, Obi C, Hill C, Simms I, et al. Monitoring the emergence of community transmission of influenza A/H1N1 2009 in England: a cross sectional opportunistic survey of self sampled telephone callers to NHS Direct. BMJ 2009;339:b3403. 14. Esposito S, Molteni CG, Daleno C, Valzano A, Tagliabue C, Galeone C, et al. Collection by trained pediatricians or parents of mid-turbinate nasal flocked swabs for the detection of influenza viruses in childhood. Virol J 2010;7:85. M.K. Akmatov, F. Pessler / International Journal of Infectious Diseases 15 (2011) e589–e593 e592 15. Lu X, Holloway B, Dare RK, Kuypers J, Yagi S, Williams JV, et al. Real-time reverse transcription-PCR assay for comprehensive detection of human rhinoviruses. J Clin Microbiol 2008;46:533–9. 16. Cooper DL, Smith GE, Chinemana F, Joseph C, Loveridge P, Sebastionpillai P, et al. Linking syndromic surveillance with virological self-sampling. Epidemiol Infect 2008;136:222–4. 17. Smieja M, Castriciano S, Carruthers S, So G, Chong S, Luinstra K, et al. Development and evaluation of a flocked nasal mid-turbinate swab for self-collected respiratory virus diagnostic testing. J Clin Microbiol 2010;48:3340–2. 18. Gilbert M, Macdonald J, Louie M, Gregson D, Zhang K, Elsayed S, et al. Prevalence of USA300 colonization or infection and associated variables during an outbreak of community-associated methicillin-resistant Staphylococcus aureus in a marginalized urban population. Can J Infect Dis Med Microbiol 2007;18: 357–62. 19. van Cleef BA, Verkade EJ, Wulf MW, Buiting AG, Voss A, Huijsdens XW, et al. Prevalence of livestock-associated MRSA in communities with high pigdensities in The Netherlands. PLoS One 2010;5:e9385. 20. Ammons DR, Puttagunta R, Granados JC, de la Garza G, Eyambe GS, Rampersad J. An exploratory study of methicillin-resistant Staphylococcus aureus and SCCmec elements obtained from a community setting along the Texas border with Mexico. Curr Microbiol 2010;60:321–6. M.K. Akmatov, F. Pessler / International Journal of Infectious Diseases 15 (2011) e589–e593 e593