Symptoms of respiratory illness in wind players and their relationship to instrument hygiene Authors: Gerard Breaden Madden, Steffen A Herff, Eckart Altenmüller, Maria Schuppert, Michael Pfeifer, Hans-Christian Jabusch Published: 2026 ABSTRACT AbstractPlaying a wind instrument has been proposed as a potential trigger of respiratory illness if harmful microorganisms residing in the instrument are repeatedly inhaled. Case reports suggest that respiratory symptoms may improve when affected individuals avoid playing their instruments and engage in appropriate hygiene measures. In this epidemiological study, an online questionnaire assessed the twelve-month prevalence of self-reported respiratory symptoms and diseases in wind players and non-wind controls recruited from German orchestras and other professional music institutions. Associations were examined between symptoms/diseases, instrument contact, hygiene behaviours, and hygiene-specific education and attitudes. Data were analysed using frequentist inferential statistics and Bayesian mixed-effects models, with Evidence Ratios (ER) 19 interpreted as strong evidence. Wind (n = 245; 117 woodwind, 128 brass) and non-wind controls (string, n = 43) did not differ in instrument contact time or smoking habits. We observed strong evidence that asthma (ER = 47.58) and diagnosed respiratory disease (ER = 22.10) were reported more frequently by wind players compared to non-wind, but no differences were seen for other symptoms such as coughing or shortness of breath. In terms of hygiene, most wind players (woodwind 86%, brass 94%) reported never disinfecting their instrument, and the frequency of both cleaning and drying were inconsistent. Contrary to our predictions, more frequent disinfection predicted a higher likelihood of reporting respiratory symptoms (ER = 432.24), possibly reflecting reactionary behaviour following symptom onset. Hygiene-specific education was inconsistent, with most wind players (85%) reporting having received no instruction about disinfection. Those who received hygiene education were more likely to consider hygiene as an essential part of their musical activity (all ER > 19). Findings may help to clarify potential relationships between wind instrument playing and respiratory health, although interpretations are drawn cautiously due to the inherent limitations of self-reported data and cross-sectional designs. As respiratory illness can potentially have severe consequences, stronger integration of evidence-based hygiene practices into wind pedagogy is warranted.Keywords: Wind instrument, Respiratory illness, Microorganism, Hygiene, Disinfection, EpidemiologySubject terms: Diseases, Health care, Medical research, Risk factors Introduction Wind instruments (both woodwind and brass) have long been recognised as a potential source of microbial contamination1,2. Repeatedly playing a wind instrument can lead to saliva saturation both inside the instrument and of its constituent parts such as a mouthpiece or reed. These damp and humid conditions may support the development and persistence of bacteria and fungi particularly when appropriate hygiene measures are not implemented3. Wind instrument contamination is believed to be extensive and widespread4,5. For example, an examination of thirteen high school band instruments (six woodwind, seven brass) revealed that all thirteen were heavily contaminated, based on samples extracted from multiple sites within each instrument. In this study, approximately three hundred different bacteria isolates and sixty mould species were identified across all instruments. Many of these were pathogenic, allergenic, or opportunistic6. In a study by Soumagne et al. (2018)7, saliva and blood samples were collected from forty healthy wind instrument players, in addition to culture samples taken from their instruments, mouthpieces, and reeds. The authors observed that 95% of the wind instruments were colonised by fungi, and reported that 80% of these wind players had blood serum precipitins (antibodies) consistent with the antigens of the fungi present in their respective instruments. In other words, although these musicians were healthy, their immune systems had already reacted to the fungi present in their instruments. Musicians’ cleaning habits were assessed via a questionnaire in this same study. Factors associated with contamination included a lack of systematic drying of the instrument after playing and irregularly changing the water in containers used for hydrating reeds before use. As a contaminated wind instrument may transfer potentially-harmful microorganisms to a user via inhalation, playing a wind instrument has been proposed as a possible trigger for respiratory illnesses, including rare and occasionally-severe conditions such as Hypersensitivity Pneumonitis (HP)7,8. HP is an allergic inflammation of the lung tissue, in particular the alveoli and bronchioles9. Inflammation is triggered due to inhalation of organic particles such as mould, to which an individual has become sensitised10. The typical symptoms of HP include progressive coughing during periods without upper respiratory tract infection, wheezing/whistling sounds whilst breathing, and shortness of breath (dyspnea) while at rest. In more severe cases, malaise, fatigue, and fever may occur11,12. HP is rare in the general population13–15. It is most often associated with occupational groups such as bird keepers or farmers who work with organic materials such as animal proteins, hay, and silage (monikers for HP include Bird Fancier’s Lung16,17 and Farmer’s Lung18,19). In musicians specifically, HP appears to be uncommon – only eight case reports have been published since 198820–26. In each of these cases, the affected individuals presented with respiratory symptoms typical of HP and all were heavily involved in playing a wind instrument (bassoon, bagpipes, saxophone, tenor horn, and trombone, respectively). Various fungi, bacteria, and/or atypical mycobacteria were identified in the instruments and were considered likely to be the triggers of the respiratory symptoms in all eight individuals. In the majority of these cases, in addition to medical treatments such as corticosteroids20,21,23,25,26 hygiene measures were implemented21–23, moreover rest24, reduction of instrument contact time22,24 or complete cessation of instrument playing25,26. Following their respective treatments, four of the affected individuals showed improvement and three were symptom-free. Unfortunately however, one individual (a bagpipe player) died following an acute deterioration of HP that was attributed to fungal contamination of his instrument24 An additional health concern adjacent to HP associated with the inhalation of germs (and especially fungi) is asthma. Based on self-report data, the prevalence of asthma among wind players in the USA ranges from 14.8-19.2%27–29. This is higher than that of the general US population (7.2%30), based on data collected between 2000-2003, approximately. In Australia, a similar prevalence (19%) has been observed in tertiary music students, compared to 12% in non-wind instrumentalists31. Interestingly however, various technical assessments of lung function (e.g., spirometry) have shown no consistent findings that wind players might suffer from impaired lung function32–34. In fact, Žuškin et al. (2009)35 identified only one wind player who suggested that asthma might be a direct consequence of their own musical activities. HP suffers from a lack of standardised diagnostic criteria and accurate antibody testing procedures to confirm sensitisation10,36. Furthermore, HP often presents with clinical features which overlap with multiple conditions making it difficult to accurately estimate the prevalence of the associated symptoms12,37,38. Due to the relative scarcity of confirmed and published cases, HP is believed to be under-diagnosed in relevant occupational groups, including wind players20. According to anecdotal observations made by a high school band leader, 50% or more of their students at any given time suffered from respiratory difficulties, requiring therapy6. Although the consequences of respiratory conditions such as HP in wind players can be severe and even fatal24, they can be readily treated. In terms of treatment approaches, respiratory symptoms have been reported to improve when affected individuals cease playing or reduce instrument contact for extended periods22,24–26. However, this approach clashes with the occupational reality of professional wind players and instrument teachers, for whom it is necessary to play regularly4. In circumstances where avoiding one’s instrument is not feasible, effective instrument hygiene is a critical issue. Recommendations intended to reduce or limit contamination include, for example, hand-brushing instruments using commercial detergents and rinsing in sanitiser solutions3, using specific chemical agents such as chloroxylenol39, and immersing mouthpieces in hot water40. Hygiene guidelines have been in circulation for many years, ranging from online resources for music educators and instrument manufacturers41–44, video tutorials45, to detailed instrument-specific instructions for disinfection46. Nonetheless, wind players may lack sufficient knowledge regarding how to properly care for their own or others’ equipment4,24,47 and with respect to the discrete functions of different hygiene activities (i.e., cleaning to remove dirt/debris, drying to remove moisture, and disinfecting to eliminate microorganisms using chemical agents). Complicating the matter is the fact that certain hygiene practices well-established among wind players (e.g., drip-drying) have not been assessed for efficacy24, and commonly recommended hygiene materials such as 70% isopropyl alcohol can damage certain wood or plastic instrument components48. In this exploratory epidemiological study, we investigate (i) the prevalence of respiratory symptoms in a sample of wind players and non-wind player controls, (ii) the relationship between respiratory symptoms, instrument contact, and hygiene behaviours, and (iii) wind players’ hygiene-specific education and attitudes. In light of existing research, we predict that wind players will report more respiratory symptoms compared to non-wind players; greater instrument contact and less-frequent hygiene will be associated with more symptoms; wind players educated about instrument hygiene will engage in more frequent instrument care than those who have not learned about these topics. Methods Participants and procedure Ethical approval for this study was provided by the responsible institutional review board (see the section “Ethics Statement”), and the study adhered to the Declaration of Helsinki. Both wind and non-wind players were eligible to participate, with the latter serving as the control group. Participants were recruited in Germany via an email advertisement distributed to orchestras and governing bodies representing wind players. Inclusion criteria were (a) playing a musical instrument and (b) the ability to take part in the German language. Exclusion criteria included a former, current, or suspected COVID-19 infection. All participants provided informed consent before completing an online questionnaire covering multiple topics: Demographics, musical biography, and musical activities Participants indicated their primary musical instrument and musician status (e.g., member of an orchestra or big band, etc.) and whether their primary instrument was ever played by other users. They reported their age at the time of the study, and the age at which they began playing their primary instrument. From this, the total duration of years playing was calculated. Instrument contact was assessed based on the number of hours per week they typically spent playing music. Participants also indicated how often per year they took breaks of several days or more during which they did not play their instrument (e.g., vacations). Key demographic and musical variables are summarised in Table 1. Table 1Overview of sample demographics and musical biography factors.Wind (N = 245)Non-Wind (N = 43)pSex (female / male / other)95 / 149 / 122 / 21n.sAge (years)52 (21 / 70)51 (23 / 64)n.sAge of commencement (AoC; years)11 (5 / 59)7 (4 / 17)Duration of playing (DoP; years)40 (3 / 61)41 (17 / 55)n.sWeekly playing time (hours)20 (<1 / 40)21 (<1 / 36)n.sProfessional musician (yes / no)185 / 6039 / 4Student musician (yes / no)7 / 2384 / 39n.sOrchestral musician (yes / no)153 / 9235 / 8Big Band musician (yes / no)3 / 2420 / 43n.sPrimary instrument owned (O) or borrowed from elsewhere (B)?(O)235 / (B)10(O)39 / (B)4n.sPrimary instrument ever played by others? (yes / no)6 / 2392 / 41n.sNumber of other users (if yes)2 (1 / 4)1 (1 / 1)n.sAdditional wind instruments (yes / no)145 / 100--Smoking (ever and/or currently; yes / no / missing)78 / 149 / 1813 / 27 / 0n.sDuration of smoking (smokers only; years)20 (1 / 51)13 (1 / 48)n.sNumber of environmental triggers (home/occupational; e.g., mould, sauna)0 (0 / 2)0 (0 / 2)n.sValues indicate frequencies (N), or median (min/max). Analyses presented in this table were conducted within a frequentist inferential framework (Mann Whitney U-test or Chi-square/Fisher’s exact test, 2-sided, uncorrected). p < 0.05; p < 0.001; n.s. non-significant. Overview of sample demographics and musical biography factors. Values indicate frequencies (N), or median (min/max). Analyses presented in this table were conducted within a frequentist inferential framework (Mann Whitney U-test or Chi-square/Fisher’s exact test, 2-sided, uncorrected). p < 0.05; p < 0.001; n.s. non-significant. Health background including twelve-month prevalence of respiratory symptoms Participants indicated whether they had experienced respiratory problems including asthma, or symptoms typically associated with HP, such as coughing, wheezing, and shortness of breath. The duration of each issue (e.g., total number of asthma episodes per year; number of weeks per year coughing) was also reported. Participants also indicated whether they had ever been diagnosed with any cardiovascular or respiratory disease (including chronic bronchitis and pulmonary emphysema). Smoking history including number of years smoking and quantity of tobacco products smoked per week was also requested. The presence of potential triggers of respiratory symptoms in their common living and occupational environments was reported by all participants (e.g., visible mould, indoor saunas, birds, stores of hay/silage). Instrument hygienics Musicians reported how often they cleaned, dried, and disinfected their primary musical instrument via a seven-point Likert scale (where 0 = never, 1 = once per year or less, 2 = once every few months, 3 = once per month or more often, 4 = one per week or more often, 5 = every day, 6 = after every use). As the same mechanical action (e.g., wiping or swabbing) may serve both cleaning and drying purposes, these activities were not operationally defined so that participants could report these behaviours according to their usual understanding within their instrumental context; however, disinfecting was differentiated by asking participants to optionally describe the chemical substances they typically used (e.g., isopropyl alcohol). Equivalent hygiene questions were posed with respect to reeds, mouthpieces, and storage accessories. These data will be reported elsewhere. Hygiene education and attitudes Musicians reported whether they had ever been educated about cleaning, drying, and disinfection. If so, they reported the sources of this information (teachers, self-taught, etc.). They also indicated how essential they believed these activities to be as part of their musical activities, and the overall importance that their music instructor and/or music institution attached to hygiene topics. Statistical approach Frequentist Mann–Whitney U tests were used to compare continuous variables between different groups. Categorical variables are presented as frequencies (N). Associations between categorical variables were conducted using Chi-Square/Fisher’s exact tests. P-values for frequentist statistical significance were not corrected for multiple tests. In addition, we used Bayesian mixed effects models using the brms package49 in R50 to model the collected data. Hypothesis tests on model predictions were conducted using the marginaleffects package51. A Bayesian framework was chosen because it allows probabilistic inference based on posterior distributions and is therefore well suited to datasets with unequal group sizes52. Depending on the measurement level of the target/outcome variable of interest, each model used either a Gaussian, Cumulative, or Bernoulli identity link. All models ran on 4 chains across 4 cores with 10000 iterations and 3000 warmups each, initialised through zeros. All models controlled for sex, age, total number of home and/or occupational environmental risk factors, smoking history (current and/or former smoker, and current cohabitation with a smoker), and the total number of additional wind instruments played. Each model was fitted with a random effect for participant and included a weakly informative prior by default for both population level effects (t(3,0,1)) as well as group level effects (cauchy(0,1))53. An equivalent model approach has been established in various studies on music psychology, perception, and physiology54–58. To investigate the main effects of each model, we report Bayesian hypothesis tests which evaluate the evidence ratio (expressed as Odds() of a given effect to be smaller or larger than zero). We also report the coefficient estimate () and the estimate error within each estimate (EE). Evidence ratios were considered strong evidence for an effect59. Model specifications, analysis code, and supplementary materials are available on OSF (https://osf.io/8fqb4/), including an overview of medications prescribed to manage respiratory issues reported by participants. Results Overview of demographic and musical biography characteristics Most musicians in this sample were male and the majority reported being professional and/or orchestral musicians. The median age was slightly older than fifty years at the time of study. Wind and non-wind players did not differ with respect to sex distribution, overall duration of playing, weekly playing time, or number of home and/or occupational environmental triggers, nor was there any association identified between smoking and status as a wind player. Wind players commenced playing music at an older age compared to non-wind players (this is considered typical within wind instrument pedagogy60). The wind instrument group comprised brass players (n = 128) and woodwind players (n = 117). The majority of wind players reported playing more than one wind instrument (median number of additional instruments = 1; min/max = 1/7). The non-wind group comprised only plucked- or bowed-string instrumentalists. An overview of these data is shown in Table 1. Prevalence of respiratory symptoms among wind and non-wind players Musicians reported the twelve-month prevalence of several respiratory symptoms: experiencing wheezing and/or a whistling sound whilst breathing (“wheezing”), dyspnea without exercising (“shortness of breath”), progressive coughing during periods without upper respiratory tract infection (“coughing”), asthma without exercise (“asthma”), and sickness during periods without a cold or other confirmed illness (“malaise”). They also reported whether they had received a current or former diagnosis of any cardiovascular disease or respiratory disease. The majority of musicians reported not experiencing any of these issues within the preceding twelve months, nor reporting any diagnosed disease. Proportionally however, the prevalence of several was elevated amongst wind players compared to non-wind (including wheezing in wind = 20.4% / non-wind = 11.6%; shortness of breath = 20.4% / 11.6%; asthma = 21.6% / 4%; cardiovascular disease = 12.3% / 4.9%; respiratory disease = 30.3% / 14.3%). An overview is shown in Fig. 1. Prevalence of various respiratory symptoms and diagnosed diseases amongst wind and non-wind players. We deployed several Bayesian mixed effects models to examine the connection between playing a wind instrument and reporting symptoms/diseases. We observed strong evidence (Odds() 19) that reporting both asthma and a diagnosed respiratory disease were increasingly predictive of playing a wind instrument. No such evidence was observed with respect to other symptoms or cardiovascular disease. An overview of these models is shown in Table 2, where values in bold indicate cases where we observed strong evidence for an effect. Table 2Summary of Bayesian mixed effects model: playing a wind instrument and respiratory symptoms/diseases.Symptoms & diseasesWheezingShortness of breathCoughingAsthmaMalaiseCardiovascular diseaseRespiratory diseaseWind player0.390.19-0.170.80.910.640.52EE0.360.320.290.441.30.460.31Odds ()6.332.702.6347.584.1013.1122.10Positive coefficients indicate increased odds of a symptom or disease being reported by wind player rather than a non-wind player. All analyses here and subsequently involving ‘wind players’ refer to woodwind and brass instrumentalists combined.Odds() indicates strong evidence for an effect. Summary of Bayesian mixed effects model: playing a wind instrument and respiratory symptoms/diseases. Positive coefficients indicate increased odds of a symptom or disease being reported by wind player rather than a non-wind player. All analyses here and subsequently involving ‘wind players’ refer to woodwind and brass instrumentalists combined. Odds() indicates strong evidence for an effect. Respiratory symptoms and instrument contact Wind players reported playing music a median of twenty hours per week (IQR = 12/25; min/max = < 1/40). Most (89%) reported taking breaks throughout the year lasting several days or longer during which times they did not play their instrument. The median duration of these breaks was thirty days (IQR = 21/50; min/max = 7/300). Greater instrument contact time did not predict a greater likelihood that wind players reported any respiratory symptoms/diseases (Odds( > 0)). However, we observed strong evidence that a longer duration of breaks per year was increasingly more commonly observed in symptom/disease-free wind players ( -0.31; EE 0.18; Odds( < 0) = 28.72). In an additional model, we focused on the dynamic temporal relationship between symptom severity and instrument contact. We observed strong evidence that reporting break-related improvement-and-worsening of symptoms was increasingly likely to be reported by a wind player than a non-wind player (Odds( < 0) = 99.72). Figure 2 displays the frequencies of break-related changes in symptom severity in wind and non-wind players. The temporal relationship between respiratory symptoms and instrument contact. Hygiene behaviours Wind players indicated how frequently they cleaned, dried, and disinfected their primary musical instrument, assessed via a seven-point Likert scale (where 0 = never, 6 = after every use). These data are displayed in Fig. 3 for woodwind players (n = 117, upper plot) and brass players (n = 128, lower plot) separately. Frequency of various hygiene behaviours among woodwind and brass players. Missing bars indicate a count of 0. Percentages calculated from valid responses within each instrument family. Cleaning, drying, and disinfecting the instrument (i) Cleaning: cleaning the primary musical instrument occurred irregularly. Woodwind players were more active in this regard than brass players, at a descriptive level. Most woodwind players reported cleaning once per week or more often, whereas most brass players cleaned only once every few months. (ii) Drying: the majority of woodwind players dried their instrument after every use, whereas most brass players never dried their instrument. Reporting any other frequency of drying was rare. (iii) Disinfecting: the vast majority of both woodwind and brass players reported that they never disinfected their instrument. Among the minority that did, disinfection was infrequent. Hygiene regularity and respiratory symptoms Two models were deployed here to assess the likelihood of wind players reporting respiratory symptoms based on the frequency of hygiene behaviours. The first model focused specifically on disinfection – the most efficient method by which to reduce the number of germs61,62. This model provided strong evidence that more frequent disinfection predicted a greater likelihood of reporting respiratory symptoms ( 0.23; EE 0.09; Odds( < 0) = 432.24). In the second model, we derived a composite hygiene variable from cleaning and drying behaviours. Although the activities operate through different mechanisms (i.e., cleaning to remove dirt/debris, drying to reduce moisture), both aim to limit the build-up and persistence of microbes inside the instrument. We therefore defined the composite hygiene variable as the maximum reported frequency of cleaning or drying. Higher scores indicate that at least one potentially protective behaviour is present, while lower scores reflect limited hygiene activity in both domains. This approach avoids penalising wind players who may be unable to dry their instrument simply due to design constraints (as is the case with many brass instruments, for example). As with the preceding model, we identified strong evidence that more frequent cleaning and/or drying predicted a greater likelihood of reporting more symptoms ( 0.25; EE 0.1; Odds( < 0) = 136.25). Figure 4 displays the marginal effects of both models, where the upward trajectory of both bands indicates more frequent instrument hygiene. Marginal effects of more frequent hygiene on the predicted probability of reporting respiratory symptoms among wind players. Line colour indicates each hygiene behaviour. Bands represent 95% credible intervals (CI). Values are normalised (M = 0, SD = 1). Knowledge and attitudes toward instrument hygiene 65% of wind players reported having received education about cleaning, 47% about drying, but only 12% reported specific education about disinfection. Figure 5 displays these frequencies. In terms of comprehensive hygiene education, Bayesian conditional probabilities indicated substantial variation in the overlap between these topics. Wind players who received education about either cleaning or drying had an 74% probability of receiving education on the other (range: 63-85%). Education about disinfection was much less consistently involved, with wind players having only a 30% (22–40%) probability of receiving this instruction even when education about both cleaning and drying was received. Amongst those educated on hygiene, these topics were most frequently provided by a music instructor (84%), a fellow musician/bandmate (36%) or via autodidactic methods (24%). Frequency of hygiene education among wind players. On the whole, wind players did not consider hygiene to be a particularly essential aspect of their musical activities. Attitudes towards cleaning, drying, and disinfection also differed significantly from one another ( = 95.59; p < .001), whereby disinfection was considered significantly less essential compared to the others (cleaning; W = 1676.00; p < .001; drying; W = 1174.50; p < .001). Amongst the wind players who reported currently belonging to a musical institution, both the music instructor and the institution at large were rated as placing limited importance on hygiene (see Table 3 for an overview). Table 3Attitudes towards hygiene among wind players, and institutional value given to hygiene topics.Wind playersCleaning is an essential part of my musical activities3 (2 / 4)Drying is an essential part of my musical activities2 (1 / 5)Disinfecting is an essential part of my musical activities1 (1 / 2)Instructor InstitutionImportance of cleaning is emphasised by my...2 (1 / 4)1 (1 / 1)Importance of drying is emphasised by my...1 (1 / 4)1 (1 / 1)Importance of disinfecting is emphasised by my...2 (1 / 4)1 (1 / 1)Ratings were based on a five point Likert scale (1 = strongly disagree, 5 = strongly agree).Values indicate median (IQR). Attitudes towards hygiene among wind players, and institutional value given to hygiene topics. Ratings were based on a five point Likert scale (1 = strongly disagree, 5 = strongly agree). Values indicate median (IQR). In a final set of models, we identified strong evidence that wind players who received hygiene education considered hygiene to be a more essential part of their musical activities. This was the case for cleaning ( 0.69; EE 0.17; Odds( < 0) > 9999), drying ( 1.11; EE 0.14; Odds( < 0) > 9999), and disinfection ( 0.94; EE 0.27; Odds( < 0) = 5599); see Fig 6, left side. Furthermore, receiving hygiene education predicted a higher frequency of both cleaning and/or drying ( 0.52; EE 0.18; Odds( < 0) = 527.3) and disinfection ( 0.55; EE 0.25; Odds( < 0) = 74.68); see Fig 6, right side. Left plot: Attitudes towards hygiene amongst wind players with vs. without hygiene education. Right plot: Frequency of hygiene activity amongst wind players with vs. without hygiene education. Cleaning and drying were merged for the behavioural perspective in the latter model, as outlined in section 3.4.2. Bars represent 95% credible interval (CI). Values are normalised (M = 0, SD = 1). Discussion In this study, we investigated the twelve-month prevalence of self-reported respiratory symptoms and the frequency of diagnosed respiratory diseases in a sample of wind players and non-wind player controls, the relationship between symptoms/diseases and various instrument hygiene behaviours, as well as hygiene-specific education and attitudes. With respect to the predictions outlined in the introduction, we identified the following: (i) strong evidence that both asthma and a diagnosed respiratory disease were reported more often by wind players compared to non-wind; (ii) instrument hygiene behaviours among wind players occurred inconsistently and was at times rare or even absent. Furthermore, engaging in more frequent hygiene was associated with a higher likelihood of reporting respiratory symptoms; (iii) many wind players reported never receiving hygiene-specific education. Receiving such education was associated with a greater likelihood that wind players engaged in more frequent hygiene. These hygiene activities, however – especially disinfection – were rarely seen as essential parts of their musical activities. The extent of respiratory symptoms in wind players Asthma and a diagnosed respiratory disease were particularly elevated among wind players compared to non-wind. Several additional symptoms were also more common, at a descriptive level. Overall, observed prevalences in wind players exceeded those reported for the general population. For example, shortness of breath affects 7–15% of the general adult population (according to a meta-analysis including 20 studies in high-income countries)63 compared to 20.4% of wind players here. The prevalence of asthma is 15–20% among US wind players27–29 compared to 21.6% in our sample—over three times higher than the current 6.2% prevalence among German adults64. The global prevalence of respiratory disease is 6-7%65,66 compared to 30.2% of wind players in the current study. As musicians often devote thousands of hours to playing their instrument, it is possible that the level of respiratory issues among wind players observed here may be related to inhalation of harmful microorganisms residing in contaminated wind instruments. This premise is consistent with (a) prior evidence showing widespread wind instrument contamination6,7, (b) all published case studies on hypersensitivity pneumonitis in wind players20–26, and (c) research on respiratory illnesses in non-musical and occupational settings: for example, agriculture67,68, the manufacturing industry69, or the use of Continuous Positive Airway Pressure (CPAP) therapeutic equipment70. An alternative explanation is that air quality in musical settings may contribute to a higher frequency of respiratory illness. Research on air quality in urban environments suggests rising levels of dust, mould spores, and other particulate matter over time71,72. As many of our sample were professional and orchestral musicians, they likely spend extended periods in shared performance and rehearsal spaces, including concert halls and opera houses. The air quality of such environments depends on ventilation and humidity control73,74. Due to the increased breathing demands of wind playing, which necessitates large volumes of inhaled air and thus potentially greater exposure to airborne particles and microorganisms, poor indoor air quality may pose particular risk to wind players. Beyond ambient air quality, certain wind instrument practices which can occur in close quarters, such as emptying the “spit valve” water keys, may be associated with increased risk of infection transmission75,76. Comparable group infection risks have been seen previously in other contexts such as sport77. In fact, the risk of communal disease transmission via musical instruments was deemed significant enough to warrant hygiene measures enacted into US law78,79. Although environmental and occupational factors may contribute, wind instruments nonetheless constitute a modifiable source of exposure to potentially-harmful microbes. Targeted hygiene practices may help reduce the risk of exposure. It is therefore essential to understand what wind players are actually doing in terms of hygiene practices, including their frequency, consistency, and underlying motivations. Inconsistent hygiene Musicians vary substantially with respect to many musical factors such as expertise-related goals and emotion regulation strategies54,55,80. It is therefore reasonable to expect variation in how they approach instrument care. Overall, we observed that wind players engaged in inconsistent cleaning and drying, and very infrequent disinfecting of the instrument. The inconsistent and low levels of hygiene activity we observed are surprising, given that the instrument is regularly in sustained contact with a wind players’ mouth and skin, often multiple times per day. Certain aspects of the hygiene inconsistency we observed are, to an extent, understandable. For example, it can be extremely difficult to access the internal sections of certain brass instruments. A French horn is a good example of this, its long, coiled tubing and narrow bends making it practically impossible to physically reach, even for elongated flexible cleaning brushes. Moisture has to be emptied by other means such as drip-drying, emptying water keys, or blowing air through the instrument. The constraints posed by certain instrument designs may account for why the vast majority of brass players reported never drying their instrument, and should not be taken to indicate neglect or disinterest in instrument care. In contrast, we observed that the vast majority of woodwind players dried their instrument after every use. Whilst disinfection was often not performed by woodwind players, regular drying may be very beneficial nonetheless, limiting microbial growth by removing moisture, the primary substrate needed by bacteria and mould to survive and multiply. That being said, it is important to reiterate that the efficacy of drying—including the long-standing practice of drip-drying among wind players—remains largely untested. One factor which may contribute to hygiene inconsistency is music-related stress. A plethora of research suggests that musicians routinely suffer from frequent or severe stress related to issues including performance demands81,82, career instability83, and playing-related pain84. It may be the case that some musicians prioritise these concerns over basic instrument hygiene, especially if the perceived risk of illness is low. Equipment cost is also a potential factor: many professional musicians and instrument teachers use extremely expensive instruments. They may be reluctant to apply chemical disinfectants or other cleaning substances to valuable equipment to avoid unintended damage or perceived changes to sound quality. If the factors contributing to instrument contamination are to be better understood, then any mechanism underlying insufficient or absent hygiene must be taken seriously. This is not merely a matter of convenience or access, but may also reflect a broader misunderstanding between perceived and actual risk of illness. Understanding and addressing such misconceptions is crucial if hygiene is to be properly deployed and serve its intended role of reducing contamination and the risk of respiratory illness. Compensatory over preventive hygiene Contrary to our predictions, we observed evidence that more frequent hygiene predicted a greater likelihood that a wind player reported respiratory symptoms. This is a surprising finding as hygiene measures are typically framed as preventive in nature. We interpret our findings to instead indicate a behavioural mis-timing. In other words, some musicians only adopt more rigorous hygiene reactively— in response to symptoms—rather than preventively. If performed consistently and frequently, preventive hygiene may in principle help limit instrument contamination and potentially reduce the risk of symptom onset—a premise that future longitudinal research could directly evaluate. However, if musicians intensify their efforts only after symptoms arise, the protective effects of hygiene measures may be obscured in cross-sectional studies. Delaying hygiene until symptoms appear is problematic and impractical as it allows microbes to persist in the instrument and fosters a cycle in which increased hygiene brings temporary relief, but symptoms may return once hygiene activity declines. Alternative explanations for the association between frequency of disinfection and probability of symptom reporting may also be considered. For example, repeated exposure to chemical agents, including aerosols or vapours from disinfecting products, may contribute to respiratory irritation, and may also pose risks to the skin and eyes85,86. However, commercially available disinfectants for musical equipment are generally considered safe when used as intended, and in case studies on HP in wind players21–23, equipment disinfection has been associated with symptom improvement. Additionally, confounding by indication may be present, as musicians with pre-existing respiratory conditions may be advised to adopt more intensive hygiene practices, potentially influencing the observed association between hygiene frequency and symptom reporting. The tendency to increase hygiene only after respiratory symptoms develop highlights a potential knowledge gap. Musicians may not fully appreciate that the absence of symptoms or disease does not equate to the absence of microbial risk. After all, not only are anti-fungal antibodies present in the blood of wind players who are themselves healthy7, but both fungal and bacterial biofilms can be harmless for long periods of time before becoming pathogenic87–89. Hygiene-related education and attitudes Receiving hygiene-specific education was highly inconsistent among wind players. The vast majority (85%) shared that they never received instruction about disinfecting and only about half learned about drying their instrument. In fact, more than one in four had never learned about any hygiene topic at all. Low levels of reported hygiene education is concerning, especially given just how desirable and effective education may be: when received, not only do educated wind players believe these topics to be more essential but they also engaged in higher levels of hygiene activity. That being said, these findings are not causal. Although education was associated with stronger beliefs in the importance of hygiene and more frequent behaviour, how effectively hygiene is implemented in practice remains unclear. We also observed that instrument teachers—the most common source of hygiene education – did not strongly emphasise the importance of hygiene topics. Institutional emphasis of hygiene was also extremely low. These findings support our earlier point that other musical responsibilities may be given greater priority, so hygiene topics are seen as less important by comparison. Alternatively, some instrument teachers may not emphasise hygiene topics as they simply lack sufficient knowledge themselves. Taken together, these findings raise the question of how standardised, evidence-based, and accessible hygiene guidelines might benefit wind players. Although the present data do not determine what form such guidelines should take nor whether current practices are effective, existing hygiene guidelines (for example, see41–46) may provide a starting point for future evaluation. Instrument makers, in particular, may be well placed to advise musicians on which agents and methods can be safely applied to their instruments. In other domains such as sport, it is common for educators to undergo supplementary accreditation (e.g., first-aid, anti-doping, safeguarding children in sport, etc.). Whether an equivalent infrastructure in woodwind and brass music – which could include training on cleaning, drying, and disinfection – would be feasible and effective remains an open question for future work. Limitations and future directions Data collection took place in Germany during the final months of lockdown restrictions related to the COVID-19 global pandemic. Although suspected history of a COVID-19 infection was an exclusionary criterion for this study, the pandemic situation, which offered fewer opportunities for playing and performing music, may have influenced wind players’ retrospective reporting of respiratory symptoms and how often they engaged in instrument hygienics. However, this concern may be somewhat mitigated by including a community control group of non-wind players, meaning that any population-wide effects of the pandemic would likely have affected both groups similarly. As recruitment occurred via an open email advertisement, the total number of musicians who received the invitation to participate could not be determined; therefore, a formal response rate could not be calculated. This recruitment approach also raises the possibility of selection bias, as those experiencing respiratory symptoms may have been more motivated to take part than those without. A similar consideration is that wind instrumentalists are likely to be highly attuned to the state of their respiratory health because respiratory illness can directly interfere with their professional musical activities. As a result, they may report respiratory symptoms with greater sensitivity, unlike non-wind players who may underestimate or even fail to recall respiratory issues if they had little impact on their ability to perform. Potential differences in the subjective perception of symptoms between wind and non-wind controls may have influenced reporting patterns in this study. Additionally, allergy history, familial predisposition to asthma, and detailed occupational history beyond musical activity were not assessed and may represent residual confounders. The frequentist comparisons presented in Table 1 should be interpreted with caution given the size imbalance between the wind (n = 245) and non-wind (n = 43) groups, and the absence of a power analysis for these tests. The Bayesian models, which are better suited to unequal group sizes, are prioritised throughout the analyses for this reason. Critiquing current hygiene practices is beyond the scope of the current research. A systematic review may help clarify which hygiene strategies have evidence for reducing contamination, but also which are most feasible and acceptable for musicians in terms of cost, complexity, and outcome. Such an evaluation could take into account hygiene strategies from non-musical domains where good respiratory conditions are critical (e.g., scuba-diving, fire-fighting). Future studies could consider how best to reliably assess hygiene behaviours. A Likert-scale point representing, for example, drying after every use may at face value suggest regular drying behaviour, but does not necessarily capture how often usage occurs nor how effectively drying is performed. Alternative approaches such as experience sampling90 or app-based gamified learning91 may be useful in this respect and may help to frame hygiene behaviours as part of one’s regular musical schedule. Health-conscious music-making can be supported by other technologies which should ideally not depend upon specialist equipment, expert knowledge, nor wealth. For example, auto-fluorescence imaging can successfully identify multiple species of bacteria92–95; common household paint can be embedded with antimicrobial silver nanoparticles which can kill various human pathogens when applied to wood, steel, and polymers96. These options warrant investigation but, to the authors’ knowledge, have not yet been applied to the field of wind players’ health. Ongoing and open collaboration between researchers and instrument manufacturers may help to establish the applied value of these technologies and may inspire new instrument designs and early-warning microbial detection systems in non-laboratory settings. Conclusion Certain respiratory symptoms/diseases may occur more often in wind players than non-wind instrumentalists. Microbial contamination of wind instruments is one potential explanation of this. Developing standardised hygiene routines and ensuring consistent hygiene-specific education remain important challenges for this field, relevant to woodwind and brass players, educators, and instrument manufacturers alike. Addressing these challenges may help to strengthen institutional awareness of occupational hygiene, support wind players’ health, and contribute to sustainable, health-conscious music-making. Acknowledgements The authors thank the unisono German Music and Orchestra Association (unisono Deutsche Musik- und Orchestervereinigung), the German Music Council (Deutscher Musikrat e.V.), and the Federal Music Association for Choir and Orchestra (Bundesmusikverband Chor und Orchester e.V.) for supporting our data collection by publicising this project. The authors also thank Maja Grützmann, Arne Samatin, Antonia Bähr, and Dorothea Schmitt for their assistance at various stages of this project including preparation of the ethics application, reviewing literature, and data processing; Sebastian Bauer for insightful discussions on the topic; Carmel Breaden C.Psychol., Dr William Breaden Madden, Dr Dejana Mutavdžin, Dr Thomas Thompson, and Dr Liliana Araújo for helpful discussions and comments on an earlier version of the manuscript. Finally, the authors express their gratitude to the musicians who participated in this project. Author contributions GBM and HCJ: study conception and design, data collection, data processing and statistical analysis, manuscript writing and review. SAH: statistical analysis, manuscript writing and review. EA, MS, MP: discussion of the statistical analysis, manuscript review. All authors contributed to the manuscript and approved the submitted version. Funding This work was financially supported by the German Federal Ministry of Education and Research – BMBF (#01PL17063). Additionally, this study was financed by the Saxon State government out of the State budget approved by the Saxon State Parliament. SAH is supported by the Australian Government through the Australian Research Council (ARC) under the Discovery Early Career Researcher Award (DECRA, DE220100961) and by the University of Sydney, through the Sydney Horizon Fellowship. Responsibility for the content lies with the authors. Data availability The public availability of this data set is not covered by the ethics application and approval provided for this study (State Chamber of Physicians of Saxony [IRB: 425931; Application Reference: EK-BR-74/21-1]). Requests to access the data set of this study should be directed to GBM. Declarations Ethics statement This study involved human participants and was reviewed and approved by the Institutional Review Board of the State Chamber of Physicians of Saxony (IRB: 425931; Application Reference: EK-BR-74/21-1). The participants provided their informed consent to take part in this study. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The public availability of this data set is not covered by the ethics application and approval provided for this study (State Chamber of Physicians of Saxony [IRB: 425931; Application Reference: EK-BR-74/21-1]). Requests to access the data set of this study should be directed to GBM.