Study 1: a randomized controlled trialDesign and participants

Study 1 was set up as a randomized controlled trial, which is considered the golden standard for testing causal claims, because it minimizes threats to internal validity [31]. A total of 267 Dutch mothers participated in an online study that employed a 1 × 2 (lactation room quality: high versus low) between-subjects experimental design. Mothers were randomly assigned to either the high-quality lactation room condition (n = 136) or the low-quality lactation room condition (n = 121), using pictures and descriptions for the manipulation of lactation room quality. Environmental sensitivity was added to the design as a continuous variable. Inclusion criteria were: (1) current or previous experience with breastfeeding and (2) being employed. Exclusion criteria were: (1) not meeting the inclusion criteria, and (2) age and/or completion time deviating more than 3 SD from the mean. The mothers had a mean age of 32.5 years (SD = 4.3), and worked on average 27.1 hours per week (SD = 6.8).

Procedure

Mothers were recruited through a message on the Facebook page of a popular Dutch website with breastfeeding information and were informed that breastfeeding and/or parenting books would be raffled among the participants who completed the survey. All mothers provided their informed consent before initiating the survey. First, we assessed environmental sensitivity, then mothers were randomly assigned to either the high-quality or the low-quality lactation room condition. They were shown pictures and a description of either the high-quality or the low-quality lactation room, and asked to imagine a scenario where they made use of this lactation room to express milk. After viewing the pictures and reading the description, they answered the survey questions comprising a manipulation check, the dependent variables, and demographic items.

Manipulation of lactation room quality

The manipulation of lactation room quality was based on the premise that a high-quality lactation room should not only meet the basic functional requirements, but should also follow the recommendations from the Theory of Supportive Design [6]. The stimulus materials for the high-quality and the low-quality lactation room conditions consisted of design drawings created by a professional interior designer, accompanied by a matching description of the room. The design drawings of the low-quality lactation room were based on examples of existing Dutch lactation rooms that only met the minimum requirements for lactation rooms according to Dutch law and guidelines, but did not foster perceptions of control, positive distraction, or social support. These design drawings showed a white room, containing a chair, a table, and a hospital bed. The design drawings of the high-quality lactation room met the minimum requirements, and in addition they aimed at fostering perceptions of control (e.g., adjustable lighting and pillows), positive distraction (e.g., nature images and decoration), and social support (e.g., supportive messages about breastfeeding). The drawings in this condition showed a room decorated with green paint on one wall and a forest-photo-wallpaper on another wall, containing a comfortable chair, a table, a bed, and many decorations, such as: pillows, a mood light, a bulletin board, books, ceramic plants, a radio etc. Both design drawings were accompanied by the following text: ‘Below you can see images of one lactation room from three different viewpoints, and a list of the available facilities. Study these images and the accompanying text carefully. Imagine expressing milk in such a room; try to imagine what this would feel like.’ For the low-quality lactation room, the text proceeded as follows: ‘This lactation room contains the following: A chair, a table for the breast pump, and a bed. There is also an adjoining room with a sink, and a door with a lock.’ In contrast, for the high-quality lactation room the text proceeded as follows: ‘This lactation room contains the following: a chair, a table for the breast pump, a bed with pillows, a mood light, a bulletin board, a card with the text: ‘Good that you are here! Take your time’, two shelves, a breastfeeding book, two picture books with nature images, a radio and 3 ceramic plants, wallpaper with an image of sun rays in the forest, a cabinet with two drawers. There is also an adjoining room with a sink, and a door with a lock.’ Because the study took place during the COVID-19 pandemic, we added information in both conditions about hygienic measures (indicating that the room is cleaned daily and that water and soap, paper towels, hygienic wipes, and disinfecting hand gel are also provided). See Fig. 2a and b for the design drawings.

Fig. 2

a Drawings of the high-quality lactation rooms. b Drawings of the low-quality lactation rooms

Manipulation check

To verify that our manipulation of lactation room quality based on the Theory of Supportive Design was successful, we developed a 4-item scale. Items were: ‘This room contains images of nature’, ‘This room contains nice, beautiful, or interesting things’, ‘This room is adjustable to my needs’, ‘This room makes me feel supported in milk expression at work’. Mothers were asked to indicate their agreement on a seven-point Likert scale from (1) ‘totally disagree’ to (7) ‘totally agree’ (α = .79). Furthermore, we asked mothers to award a report grade for lactation room quality on a scale of 1 to 10 (1 = very bad; 10 = very good). As intended, one-way ANOVAs showed that mothers perceived the high-quality lactation room as being more consistent with the Theory of Supportive Design (M = 6.01, SD = 0.68) than the low-quality lactation room condition (M = 3.13, SD = 0.88, F(1,265) = 893.60, p <  0.001). Moreover, mothers awarded a higher report grade for lactation room quality in the high-quality lactation room condition (M = 8.93, SD = 1.14) than in the low-quality lactation room condition (M = 6.69, SD = 1.57, F(1,265) = 981,01, p <  0.001). We therefore conclude that our manipulation of lactation room quality was successful.

Measures Environmental sensitivity

Environmental sensitivity was assessed before participants saw the design drawings and consisted of the 12-item short version of the HSP Scale [7, 32]. Example items of the HSP-scale are ‘Do you notice and enjoy delicate or fine scents, tastes, sounds, works of art?’ and ‘Are you bothered by intense stimuli, like loud noises or chaotic scenes?’ Answering options ranged from 1 ‘not at all’ to 7 ‘extremely’. The internal consistency of the scale was good (α = .82).

Stress

Anticipated stress was measured using the short version of the State-Trait Anxiety Inventory for adults [33]; this short version [34] is well validated and has been shown to correlate highly with physiological measures of stress [35]. Mothers could indicate on a four-point Likert scale, ranging from (1) ‘not at all’, to (4) ‘very much so’ the extent to which they would feel calm/ tense/ upset/ relaxed/ content/ worried in the room that was shown to them (α = .82).

Cognitions about milk expression at work

Anticipated attitude, perceived support, and perceived behavioural control towards milk expression at work were operationalized according to the guidelines by Ajzen [21, 36]. Attitude was measured by presenting mothers with the following statement: ‘For me expressing milk at work in the room that was shown would be…’. This statement was followed by three 7-point, semantic, differential adjective scales: ‘unenjoyable – enjoyable, unpleasant – pleasant, negative – positive’ (α = .94). Perceived support was measured with four bipolar items: ‘Judging from the room that was shown I think that my supervisor approves of me expressing breast milk at work’ and ‘Judging from the room that was shown I think that my supervisor supports me expressing breast milk at work’. These two items were then repeated, replacing ‘my supervisor’ with ‘my co-workers’. All of the items were answered using a 7-point Likert scale, ranging from (1) ‘strongly disagree’ to (7) ‘strongly agree’ (α = .93). Perceived behavioural control was measured by two items: ‘In the room that was shown, expressing milk at work would be…for me’, rated on a scale from (1) ‘impossible’ to (7) ‘possible’, and ‘In the room that was shown, I could express milk at work if I wanted to’, rated on a scale from 1 ‘strongly disagree’ to 7 ‘strongly agree’ (α = .61). Anticipated intention to express milk at work was measured with a single item, based on an Australian study on breastfeeding duration [37]. The item was: ‘How long would you like to express milk at work if the lactation room shown was available at work? In that case, I would like to express milk at work until my baby is ... months old’. Participants were asked to indicate their intended duration of milk expression at work as a whole number of months.

Perceived organizational support

Perceived organizational support was measured by selecting eight high-loading items (loadings from .71 to .84) from the Survey of perceived organizational support [38]. Examples of items that were used are: ‘The organization fails to appreciate any extra effort from me’ (reversed), ‘The organization really cares about my well-being’, ‘The organization cares about my general satisfaction at work’, ‘The organization shows very little concern for me’ (reversed). The statements were preceded by the sentence: ‘Taking into account the room that was shown I would think that…’. Participants indicated their agreement with each item using a 7-point Likert-type scale (1) ‘strongly disagree’, (7) ‘strongly agree’ (α = .92).

Subjective well-being

Subjective well-being was measured based on the 2-item scale developed by Statistics Netherlands [39]. The items were: ‘On a scale from 1 to 10 can you indicate to what extent you would consider yourself to be a happy person if you expressed milk in the room that was shown? (1 = completely unhappy, 10 = completely happy)’ and ‘On a scale from 1 to 10 can you indicate how satisfied would you be with the life you lead at the moment if you expressed milk in the room that was shown? (1 = completely dissatisfied and 10 = completely satisfied)’ (α = .93).

Study 2: a field experimentDesign and participants

To complement the results of Study 1 and improve the ecological validity of our research findings, a second experimental study was conducted in a real-life setting. A total of 61 lactating employees from a large hospital in Groningen, the Netherlands, participated in the research. Since on average 90 mothers make use of the lactation rooms on the maternity ward each year according to the secretary of the ward (Mollema, Y., personal communication, August 15, 2017), 61 participants over a two-year period reflects a response rate of approximately 34%. We used a 1 × 2 (lactation room quality: high versus low) between-subjects experimental design, with two measurement points: the first (T1) as soon as the mother returned to work (or maximally four weeks afterwards), and the second (T2) four weeks after their return to work (thereby making sure mothers could have used the lactation room for at least four weeks). Although the intention was that mothers filled in the T1 questionnaire as soon as they returned to work, most mothers signed up somewhat later. It was decided that the T1 questionnaire could be filled in maximally four weeks after the return to work. Environmental sensitivity was added to the design as a continuous variable. Inclusion criteria were: (1) returning from maternity leave no more than 4 weeks prior to T1, and (2) making use of the lactation rooms on the maternity ward of the hospital at work at T1. Exclusion criteria were: (1) no longer making use of the respective lactation rooms at work at T2. The experiment took place over a two-year period: from June 2018 until June 2020. In the first year, all participating mothers were assigned to the low-quality lactation room condition (n = 32) and in the second year all participating mothers were assigned to the high-quality lactation room condition (n = 29). The mothers had a mean age of 31.5 years (SD = 3.1) and worked on average 30.3 hours per week (SD = 7.1). On average mothers used the lactation room 5.8 times per week (SD = 2.8). About two thirds of the mothers (62.7%) also used an alternative lactation room (M = 3.4 times per week; SD = 2.4). There were no significant differences between mothers in the experimental group and the control group with regard to these characteristics.

Procedure

Mothers were recruited by placing flyers in the three lactation rooms in the maternity ward at the hospital. The flyers pointed out that participants for a study on experiences with milk expression at work were sought and that breastfeeding and/or parenting books would be raffled among the participants who completed the survey. Mothers could receive further information and an invitation to participate, by leaving their name, e-mail, and the date they had returned from maternity leave on a participation form. Every mother who handed in the participation form (at the front desk of the maternity ward), received a chocolate bar as a token of our gratitude. Invitations for the pre-test questionnaire were sent as soon as the mothers signed up for the study, mostly in the first week after they returned to work. Invitations for the post-test questionnaire were sent four weeks after the mothers returned to work. We emphasized that participation in the study was anonymous and voluntary and that they could withdraw from the study at any time. All mothers provided their informed consent before continuing to the survey. In the pre-test, mothers answered survey questions about their environmental sensitivity and demographic information. In the post-test, when mothers had been using the hospital’s lactation room for at least four weeks, they answered survey questions comprising a manipulation check and the dependent variables.

Manipulation of lactation room quality

The manipulation of lactation room quality corresponded to that in Study 1, but in the field experiment, we used and adapted the existing lactation rooms in the maternity ward of the hospital. In the low-quality condition, mothers made use of three identical standard lactation rooms in the hospital maternity wards where the research took place. These low-quality lactation rooms were basic white hospital rooms, containing a chair, a table, a hospital bed, and a hospital grade breast pump (which prevented unwanted individual variance in pumping experiences due to the breast pump used.) After one year the three lactation rooms were refurbished and painted in order to create the high-quality condition, based on the design drawings that had been created for Study 1. Similar to Study 1, these high-quality lactation rooms were identically decorated with green paint on one wall and a forest-photo-wallpaper on another wall, they contained a comfortable chair, a table, a bed with multiple pillows, a mood light, a bulletin board, with a card that welcomed mothers to the lactation room, a breastfeeding information book, two picture books with nature images, ceramic plants, a cabinet with two drawers, and a hospital grade breast pump. For photographs of the lactation rooms in the high-quality and the low-quality condition, see Fig. 3a and b.

Fig. 3

a Photos of the high-quality lactation rooms. b Photos of the low-quality lactation rooms

Manipulation check

The manipulation checks (α = .86 for the 4-item scale) were measured exactly as in Study 1. As intended, one-way ANOVAs showed that mothers perceived the high-quality lactation room as being more consistent with the Theory of Supportive Design (M = 5.45, SD = 0.90) than the low-quality lactation room condition (M = 2.28, SD = 0.86, F(1,59) = 196.18, p <  0.001). Moreover, mothers awarded a higher report grade for lactation room quality in the high-quality lactation room condition (M = 7.79, SD = 0.94) than in the low-quality lactation room condition (M = 6.22, SD = 1.52, F(1,59) = 23.12, p <  0.001). We therefore conclude that our manipulation of lactation room quality was again successful.

Measures

The measures we used corresponded to the ones we used in Study 1. We made some small adjustments in wording, taking into account that this was a field study instead of a scenario study. This, for instance, allowed us to use the present tense (e.g., I feel) instead of the conditional simple tense (e.g., I would feel).

Environmental sensitivity (α = .81) was measured exactly as in Study 1. Stress (α = .81), subjective well-being (α = .76), and perceived organizational support (α = .90) were measured using the same items as in Study 1, but stated in the present tense. To assess attitude (α = .88), perceived support (α = .87), and perceived behavioural control (α = .77) towards milk expression at work we used similar measures as in Study 1. However, we specified the behaviour of ‘expressing milk at work’ further by adding ‘until my baby is at least 6 months old’. Moreover, for the measurement of attitude we added two semantic, differential adjective scales: ‘worthless – valuable’ and ‘useless – useful’, to also include utilitarian aspects of attitude [40]. For perceived behavioural control we added 2 items to improve the reliability of the scale: ‘For me pumping milk at work until my baby is at least 6 months old is…’, rated on a scale from 1 ‘hard’ to 7 ‘easy’, and ‘It is mostly up to me whether or not I pump milk at work until my baby is at least 6 months old’, rated on a scale from 1 ‘strongly disagree’ to 7 ‘strongly agree’. We replaced the intention to express milk at work of Study 1 with a 3-item measure based on the guidelines developed by Ajzen [21]. Answer options were on a scale from (1) ‘strongly disagree’ to (7) ‘strongly agree’. The items were: ‘I intend to express milk at work until my baby is at least 6 months old’, ‘I will do my best to express milk at work until my baby is at least 6 months old’, and ‘I plan to express milk at work until my baby is at least 6 months old’ (α = .94).