The major product lines used for common cleaning tasks included general purpose cleaners, glass cleaners, washroom cleaners, and floor finishing products. Examples of products and their ingredients are given in Additional file 1. A list of chemical ingredients identified from MSDSs is given in Additional file 2. The most frequent ingredients (that occurred in more than three products) are highlighted in bold in Additional file 2.

Ingredients of concern identified based on the previously expanded criteria, included: quaternary ammonium chlorides or "quats", glycol ethers such as 2-butoxyethanol, ethanolamine, several alcohols such as benzyl alcohol, ammonia and several phenols. Additional file 3 presents a summary of ingredients' chemical and physical properties, health effects of their inhalation and dermal exposures, and the purpose of the application in cleaning products.

Common cleaning tasks identified included: preparation of cleaning solutions, floor cleaning, window cleaning, mirror cleaning, toilet bowl cleaning, sink cleaning, and floor finishing tasks (buffing, waxing and stripping).

Qualitative exposure assessment of inhalation exposures resulted in classification of cleaning tasks into three major exposure groups: low, medium, and high exposures.

Two examples of step by step estimations of potential skin exposures for mirror and floor cleaning are given in Additional files 4 and 5. The results of DREAM estimates for five cleaning tasks are presented in a graph given in Figure 2. This graph presents the total body potential skin exposure for five tasks along with contribution of three exposure routes emission, deposition and transfer for each task. The results of individual body part contribution to the potential total body dermal exposure/task for the five cleaning task evaluated is presented in Figure 3.

According to the DREAM categories, cleaning tasks create moderate (such as in floor cleaning tasks) and high potential for dermal exposure (such as in mirror/window cleaning, sink cleaning and toilet bowl cleaning tasks). We identified the relative contribution of three dermal exposure routes for different tasks as shown in Figure 2. As seen from this graph, the "emission" route contributes more to the overall exposure compared to "transfer" and "deposition" routes during mirror and toilet bowl cleaning. One possible explanation to this finding is related to the spraying activities that generate liquid particles with aerodynamic diameter >100 μm that potentially reach the skin. In the case of sink cleaning, the "emission" is lower because the potential for aerosol particles to reach the head and upper body parts here is lower compared to mirror and toilet bowl spraying. Transfer contributed more during floor cleaning, probably due to the continuous hand contact with the mop handle contaminated with cleaning solution. Overall, floor cleaning tasks were associated with the lowest potential for dermal exposures. Hands were identified as having the highest potential for dermal exposure for most of the tasks. Forearms were at the next highest risk of exposure during sink, toilet bowl & mirror cleaning while for floor cleaning, feet and lower legs were most prone to exposure.

The chemical ingredients identified in the products included disinfectants, surfactants, solvents, and fragrances. These ingredients are representative of different chemical classes such as ethers, alcohols, amines, acids and have a very wide range of volatilities and other chemical properties. The same chemical ingredients we have identified here have been previously reported by several studies [1,26,27].

When investigating ingredients using product MSDSs, health and safety professionals should review not only MSDSs of concentrated product forms, but also the ready to use forms. We found that many ingredients reported in the concentrated form were missing in the RTU form, because MSDSs are required to list only ingredients at concentrations greater than 1% in the product. This is important for identifying ingredients that are sensitizers in the workplace; given the fact that sensitization may occur even at trace concentrations.

One important finding is related to the high frequency of use of disinfectants among different product groups. Disinfectants are added to the cleaning products with the main goal to destroy microbial life. On the other hand, cleaning is done with the goal of mechanically removing the surface contaminants. An important question that can be raised is: Can disinfectants achieve their goal if they are applied in combination cleaner-disinfectant product? In order for disinfection to be effective, it should follow surface cleaning and the disinfectant should reside on the surface for about 10–15 minutes after application[28]. In the case of combination product (cleaner-disinfectant) application, these procedures can not be followed. The effectiveness of disinfectants used for common cleaning activities has been questioned in the literature [29-31]. Although the evidence to date is minimal, repeated application of disinfectants may increase the risk of microbial resistance, which will require the use of stronger disinfectants in order to be destroyed [32,33]. Given: 1) the uncertainty of disinfectant effectiveness in cleaning public areas, 2) the risk of inducing bacteria resistance, and 3) the health concerns related to the use of disinfectants, it is critical to further evaluate disinfectants' effectiveness for common cleaning activities and to develop workplace strategies for preventing workers from exposures to disinfectants. Such strategies may include purchasing of green cleaning products, identification of the areas where disinfection is needed, and following the necessary disinfection procedures in the cases when disinfection is necessary.

2-Butoxyethanol (2-BE), a glycol ether with boiling point (BP) of 168°C, was commonly used in cleaning products including glass/window cleaners, carpet cleaners and other surface cleaners[36]. Indoor exposures to its vapors at a concentration threshold of 2 ppm (10 mg/m3) and above may result in sensory irritation [34]. The OSHA permissible exposure limit (PEL) is 50 ppm for 8 h time weighed average (TWA), the ACGIH threshold limit value (TLV) is 20 ppm (8 h TWA) and the NIOSH recommended exposures limit (REL) is 5 ppm (10 hour exposure). There is a skin designation for 2-BE from both OSHA and NIOSH, indicating that 2-BE can be absorbed through the skin. The presence of 2-BE in cleaning products has been reported by several studies [35,36]. Concentrations of 2-BE in the air generated during window cleaning reported by Vincent 1993 ranges from 0.1–7.33 ppm, lower than existing occupational standards. The study suggested that dermal exposure may be the most important exposure route in the workplace[37]. Because 2-BE was one of the most frequent solvents in our products and had the highest concentrations in the bulk products, it is important to further assess its workplace exposures. Quantitative workplace investigations are necessary to measure the degree of exposure intensity and relationship with irritation symptoms reported among cleaning workers.

Quaternary ammonium compounds, or quats, were widely used in many of the products investigated. Quats have been identified by Nielsen 2007 as one of the indoor agents that may promote development of airway allergy[27]. In his review, Nielsen summarizes the evidence from animal and humans studies that relates quats exposures with allergy-promoting effects. He concludes that the mechanism of asthma from quats remains unknown. One important consideration in the investigation of asthma symptoms from quats is the understanding of exposure routes in the workplace. Although several case reports identify an asthma-quats exposure relationship, they lack the clarification on how exposure had occurred [2,5,6]. In the second case report of a study by Purohit 2000, the nurse developed symptoms after entering the room that was cleaned with a surface cleaning product that contained benzalkonium chloride[5]. Because the nurse was not involved in cleaning activities the most probable exposure route would be inhalation and not through skin. A study by Vincent 2006 showed non-detectable levels of quats in the hospital indoor air [38]. Inhalation to quats may happen in two ways: 1) by inhaling aerosolized liquid particles generated during product application, or 2) by inhaling quats absorbed into the dust particles that are re-suspended in the air. Further quantitative workplace investigations of inhalation and dermal exposures will provide important evidence for understanding actual exposure routes for quats.

Mono-ethanolamine, used as surfactant, was found in most of the product types investigated, with exception of the floor cleaners. It has a boiling point of 171°C and dissolves very well in water. Exposures to its vapors can irritate the nose, throat, and lungs, causing coughing, wheezing and shortness of breath. The OSHA PEL for mono-ethanolamine is 3 ppm and the ACGIH 15 min short term exposure limit (STEL) is 6 ppm. Exposure to mono-ethanolamines from cleaning agents have been related to occupational asthma [4]. To understand the exposure response relationship and the mechanism of asthma, it is necessary to investigate exposure patterns in the workplace including short term peak exposures. Dermal exposure assessment should be considered in further workplace exposure assessment strategies, given the concern that mono-ethanolamine can be absorbed through the skin [19].

Fragrances were used commonly in bathroom cleaners. Exposure to fragrances is a topic of special interest because they may cause secondary emissions due to reactions of the primary exposures with oxidizers present in indoor air (e.g. terpenes, a family of chemicals common in fragrances, reacting with ozone in indoor air) [39]. These reactions can produce secondary gaseous and aerosol ultrafine particles that may be responsible for the airway irritation symptoms [40,41]. Very recently, Wolkoff (2008) found that gaseous products generated from ozone-limonene reactions are causative of acute irritation effects measured in a mice bioassay[42]. Another important consideration related to fragrance use in cleaning products is the presence of odor during and after cleaning. Even at low concentrations, the presence of compounds with low odor thresholds may cause perceived respiratory irritation because of odor annoyance[41]. Furthermore, this is a topic of special interest due to the reported sensitization effects associated with the fragrance use [43].

Surprisingly, bleach was not used in any of these products compared to findings from other studies, which found that bleach can be responsible for asthma symptoms among domestic cleaners [14,44].

Volatile compounds identified in cleaning products covered a wide range of volatilities, from highly volatile ingredients such as ammonia (BP = -33°C) and isopropyl alcohol (BP = 82°C) and relatively less volatile ingredients such as 2-butoxyethanol (BP = 168°C) and mono-ethanolamine (BP = 171°C). The highest intensity of VOC exposures in the workplace is expected during the use of floor strippers and general purpose cleaners because they contain the highest concentrations of VOCs in the bulk. Inhalation exposure to aerosol particles of volatile and non-volatile ingredients can be facilitated during product spraying. The worst exposure scenarios can happen when several cleaning tasks are performed in small and poorly ventilated spaces, such as bathrooms.

Hazardous exposures related to cleaning products are an important public health concern because these exposures impact not only cleaning workers, but also other occupants in the building. Data from laboratory studies indicate a two phase decay of the air concentrations in the room. The first phase decay happens very fast (in the first 10 minutes) and the second phase decay happens slowly (about 1–2 hours for the air concentrations to reach the background level). Furthermore, experimental studies have shown that some compounds such as glycol ethers are released slowly from the surfaces. This creates potential for exposure of other occupants in the building, hours after the cleaning activities are performed [35,36]. The intensity of exposures after the completion of cleaning has not been investigated in field studies. In a follow up study we will conduct quantitative assessment airborne exposures during cleaning and will provide evidence on the exposure levels after cleaning.

Application of the DREAM method in this pilot study confirmed the applicability of this method for categorization of cleaning tasks in different dermal exposure categories. Exposure categories identified included two groups: "high" (for sink, mirror and toilet bowl cleaning) and "moderate" (for floor cleaning with two different methods) exposures. The difference between these two groups of tasks may reflect the product applications procedures, such as spraying (typical for the first group of tasks) versus mopping (the second group of tasks). The DREAM method did not find differences within tasks that involve spraying and within the tasks that do not employ spraying. Both floor cleaning methods were in the same exposure category, even though there were important changes in the cleaning procedures (such as dipping the hands into cleaning solution during the microfiber mop method). This limitation has also been observed by the DREAM authors, who recommend the method is most appropriate for detecting high contrast exposure levels[23]. Furthermore, for all cleaning tasks, DREAM identified hands as the body part at higher potential for dermal exposure compared to other body parts. Overall, our results suggest that dermal exposure prevention should focus mostly on hands and the activities that involve product spraying.

The DREAM observational analyses applied here showed that dermal exposure can be an important route for chemicals in the body. Recent literature suggests that some chemical ingredients, such as isocyanates, may be able to penetrate the skin and cause systemic respiratory effects [45]. Dermal exposure should be evaluated in future studies of health effects of cleaning.

A comprehensive approach to exposure prevention will account for the method with which a product is applied and the task requirements, as well as assessing of the chemical ingredients and implementation of safer alternatives to cleaning products.

The results of this work are based on a small number of products. While we selected a few representative hospitals, it is possible that other products with additional ingredients are used elsewhere.

This study does not address the lack of quantitative data in the literature regarding the concentrations of cleaning compounds in workplace air. Quantitative characterization of exposures would better identify activities that produce the highest exposure, important for control measures. This work serves as preparation for a detailed quantitative assessment of airborne exposures from cleaning tasks.