Luận văn Thạc sĩ Assessing Risk Factors For Sudden Infant Death Syndrome And Caregivers’ Perceptions Of The Cardboard Box For Infant Sleep

EliScholar – A Digital Platform for Scholarly Publishing at Yale January 2019 Assessing Risk Factors For Sudden Infant Death Syndrome And Caregivers’ Perceptions Of The Cardboard Box Fo

EliScholar – A Digital Platform for Scholarly Publishing at Yale

January 2019

Assessing Risk Factors For Sudden Infant Death Syndrome And Caregivers’ Perceptions Of The Cardboard Box For Infant Sleep Nisha Dalvie

Follow this and additional works at: https://elischolar.library.yale.edu/ymtdl

Assessing Risk Factors for Sudden Infant Death Syndrome and Caregivers’ Perceptions of

the Cardboard Box for Infant Sleep

A Thesis Submitted to the Yale University School of Medicine

in Partial Fulfillment of the Requirements for the

Degree of Doctor of Medicine

by Nisha Dalvie

2020

Nisha S Dalvie, Victoria Nguyen, Eve Colson, and Jaspreet Loyal Department of Pediatrics, Yale University, School of Medicine, New Haven, CT

Some US hospitals are giving out cardboard boxes as a way to address behaviors

associated with Sudden Infant Death Syndrome (SIDS) Our goal was to evaluate the

cardboard box for this purpose by quantifying current practices and qualitatively assessing caregivers’ perceptions of the cardboard box Study participants were English or Spanish-speaking caregivers of 2-16 week old infants presenting to primary care clinics in New Haven, CT Caregivers completed a survey asking about demographic data and SIDS risk factors, such as non-supine positioning and bed-sharing Some caregivers also participated in

a semi-structured interview about the cardboard box, created used a grounded theory

approach Of 120 survey respondents, 38% of all participants and 63% of Spanish-speaking participants reported bed-sharing at least some of the time Factors associated with bed-sharing included Spanish as the primary language (OR: 4.3 [95% CI: 1.9-9.9]) Factors associated with non-supine positioning included Hispanic ethnicity (OR: 2.6 [95% CI 1.2-5.8]), caregiver born outside the US (OR: 4.2 [95% CI: 1.8-9.6]), Spanish as the primary language (OR: 6.3 [95% CI: 2.7-14.7]), and less than high school education (OR: 3.4 [95% CI: 1.3-8.9]) Of 50 interview participants, 52% said they would use the cardboard box for their infant to sleep in compared with 48% who said they would not The following 3 themes emerged from the data: (1) safety of the cardboard box; (2) appearance and (3) variation in

planned use In conclusion, bed-sharing rates were higher in our study population compared

to the national average, highlighting the need for better resources; however, participants were divided about whether they would actually use the cardboard box, indicating it may not be a successful intervention in our community

Acknowledgements

Thank you to Dr Eve Colson for her introduction to this field and her crucial expertise Thank you to Dr Maryellen Flaherty-Hewitt and Camisha Taylor for their flexibility in the primary care clinic workflow so that this project could succeed Most importantly, thank you to Dr Jaspreet Loyal for her incredible mentorship, unwavering support, and life-long lessons in pediatric clinical care that all clinician-educators should

aspire to

This work was supported by the National Institutes of Health

[Grant 2 T35 HL 7649-31]

Introduction

Sudden Infant Death Syndrome: Background and Risk Factors

Barriers to Safe Sleep and Studied Interventions

The Cardboard Box for Infant Sleep

Our Project

Statement of Purpose and Specific Aims

Methods

Setting and Sample

Data Collection

Data Analysis

Results

Overall

Sleep Positioning

Sleep Location

Qualitative Themes

Perceptions of the Cardboard Box, Demographics, and Sleep Practices

Discussion

Our Caregiver Population

Comparing National Prevalence of Sleep Practices with Our Data

Evaluation of the Cardboard Box for Infant Sleep

Study Limitations and Opportunities for Future Work

References

Appendices

Appendix A: PDF of Yale Qualtrics Survey

1-10

1

3

6

9

11

12-16

12

12

15

17-30

17

20

22

26

29

30-45

29

31

37

41

46-51

52-62 52

Introduction

Sudden Infant Death Syndrome (SIDS): Background and Risk Factors

Sudden infant death syndrome (SIDS), a type of sudden unexpected infant death (SUID) often associated with sleep, is defined as the sudden unexpected death of a child less than 1 year of age and outside of the perinatal period that remains unexplained after thorough work-up, including a complete autopsy.1 It is the leading cause of post-neonatal mortality in the United States and the third leading cause of infant death overall,

responsible for 3,600 deaths in 2017.2 Although SIDS remains a diagnosis of exclusion, risk factors related to intrinsic biological factors as well as the external sleep environment have been identified.3 The most well-established risk factors are non-supine sleep

positioning, soft and loose bedding, presence of items such as pillows and blankets,

sleeping on surfaces other than cribs (i.e adult beds, sofas), and sharing, where sharing is defined as an infant sleeping on the same surface as another person.4 Other factors correlated with higher SIDS incidence include male sex, black race, families who identify as lower socio-economic status, mothers younger than 20, low birth weight / pre-term infants, and cigarette smoking during pregnancy.5 It is important to note that none of these risk factors are sufficiently strong enough to identify a pathophysiologic cause, but have assisted in creating a descriptive profile that associates maternal, neonatal, and

bed-environmental factors with SIDS risk, as illustrated in Figure 1

Based on this emerging profile, the American Academy of Pediatrics has published recommendations for pediatricians to counsel families on modifiable factors to prevent SIDS The first guideline, published in 1992, recommended that infants be placed in a non-prone position for sleep; in 1994, this guideline became the basis for the “Back-to-Sleep”

Figure 1: Maternal, Neonatal, and Environmental Risk Factors for SIDS (Triple Risk

Model adapted from Filiano and Kinney6)

campaign (later becoming the “Safe-to-Sleep” campaign), a collaboration between the AAP and the National Institute of Child Health and Development (NICHD).7 Over the next

8 years, the percent of infants placed on their backs to sleep increased from 17% to 70%, and the incidence of SIDS decreased by over 40%.8-9 Some papers note that part of the reason for the drop in SIDS incidence may simply be because of diagnostic shift, with more cases of SUID being ruled as accidental suffocation rather than SIDS as more

thorough death scene investigations were performed later in the decade, but it is unlikely for changing classifications to account for all the decrease throughout the decade.10-12 The safe sleep recommendations have been updated several times since the 1990s, with the most recent 2016 recommendations expanding to specify placing infants in the supine position for sleep, avoiding cigarette smoke during and after pregnancy, using a firm sleep

surface with tight fitting bedding and no other loose articles such as pillows, and to avoid bed-sharing.5

Despite the improved strength of these recommendations and their uptake by

pediatricians and parents alike, SIDS incidence has not significantly decreased in the past two decades in the United States as a whole, although there are wide variations between states.13 Several large-scale studies have identified non-adherence to the AAP

recommendations and associated factors as a potential explanation for this plateau An analysis of results from the Web-based National Child Death Review Case Reporting System (NCDR-CDS) showed that, out of over 3000 cases of SIDS across 9 states, 70% of cases identified the infant on a surface not intended for infant sleep and 64% of infants were sharing a sleep surface with an adult or older child.14The nationally representative Study of Attitudes and Factors Effecting Infant Care (SAFE), which surveyed over 3000 caregivers about infant sleep practices between 2011-2014, found that although 77.3% of mothers usually place their child supine, only 43.7% intentionally place their child

exclusively supine.15 In addition, this study found that black mothers and mothers with less than a high school education were more likely to place their child in a non-supine position compared to white mothers and mothers with at least a high school education, aligning with results from a prior national survey from 1993-2007 and older studies on SIDS risk factors.5,16

Barriers to Safe Sleep and Successful Interventions

These findings lead to the all-important question: why are caregivers still practicing sleep positions that are non-adherent to current safety recommendations? It seems unlikely

to be primarily caused by lack of adequate education, as caregivers who practice

non-supine positioning indicate they are aware of their doctors’ recommendations, although studies have found caregivers that use prone positioning are less likely to be aware of the associated SIDS risk.16 Studies on the “ABC” messaging of safe sleep (Alone, on the Back, and in a Crib) have found no statistically significant changes in sleep positioning before

and after caregivers receive this information via crib card, as more than 80% of them were already aware supine positioning is the safest.This study found significant changes in sleep environment before and after patients communicated with nursing about safe sleep

practice, including a 40% reduction in loose articles within the crib, but could not attribute this to “ABC” messaging due to low compliance of using the crib card.17 These findings indicate that such communication methods may not be the most effective target to reducing SIDS risk factors, possibly because lack of knowledge is no longer the biggest barrier to safe sleep practices as it was in the 1990s and early 2000s: in 2015, 99% of caregivers at one hospital were aware of supine positioning and crib recommendations both at time of discharge and at 6 month follow-up, a significant increase compared to the National Infant Sleep Position (NISP) study results from 1993-2010.16,18

Interventions based in health messaging have been more successful if they gave caregivers specific rationales rather than re-iterating the best practices This has been demonstrated by randomized controlled trials in Washington, DC and Porto Alegre, Brazil that showed reduced bed-sharing rates and increased supine positioning after educational sessions designed to elicit reasons for choosing sleep positions.19-20 Other examples of successful education-based interventions include a nursing quality improvement (NQI) pilot to provide postpartum teaching about safe sleep practices prior to discharge, and a mobile health texting service to deliver tailored messages to caregivers about safe sleep for

2 months post-discharge These were both evaluated through the Social Media and Reduction Training (SMART) clinical trial, which demonstrated that caregivers who received both the NQI and the mobile health interventions for safe sleep reported the highest percentages of adherence to safe sleep practices.21 The success of all these

Risk-initiatives emphasizes the importance of understanding families’ attitudes about safe sleep practices in order to actually counteract barriers adherence: the one-on-one discussions, mobile health messages, and nursing education time were to address each caregiver’s unique concerns about safety recommendations, specifically about the comfort of supine positioning and reminders that their children are not immune to SIDS

This was not the first study to identify caregivers’ attitudes around the AAP

recommendations as a potential barrier to safe sleep practices In 2005, qualitative findings from focus groups of mainly black mothers in urban areas, a population which has been identified as high-risk for non-adherent practices since the 1990s, demonstrated concerns about choking in supine position, lack of trust in health providers compared to mothers in their families, and the perception that infants would be more comfortable on their

stomachs.22 The previously mentioned Study of Attitudes and Factors Effecting Infant Care (SAFE) from 2011-2014 also examined caregivers’ attitudes about sleep practices, and identified that mothers who believed they did not have control over their infants’ choice of sleeping position were much more likely to include prone sleep in their intended practices.15These findings make it clear that simply stating AAP recommendations to caregivers is not enough to ensure their uptake- successful interventions must address the root causes of parents’ concerns, whether that means anticipatory explanations about choking risk in the supine position or being culturally respectful of mothers’ traditions

while explaining the dangers of bed-sharing This framework is particularly important for populations that are already at a higher risk for SIDS, particularly pre-term infants, black families, and younger or less formally educated mothers

Cardboard Box for Infant Sleep

With the context of SIDS risk factors, AAP recommendations, and the best

interventions to improve adherence, we can now focus on a proposed intervention that has captured the attention of pediatricians around the globe: a cardboard box for infant sleep The government of Finland has utilized this resource since the 1930s, during which time infant mortality rate was recorded as high as 9%.23-24 Initially, only low-income mothers who had attended all their prenatal care appointments were eligible, making the box both

an incentive for mothers to attend all their appointments and a public health intervention for mothers who could potentially not afford another sleeping space; the box itself came with gauze diapers, muslin to stitch baby clothes, and a baby mattress.25 Although it is impossible to determine the effect of these kits on maternal health or infant outcomes such

as SIDS, especially with other important interventions such as vaccinations and midwife delivery beginning during this time period, the infant mortality rate in Finland decreased to 3% by 1950 and is now 0.17%, one of the lowest in the world.26 The cardboard box kit is now offered to all new caregivers, including those who adopt, and includes indoor and outdoor baby clothes, diapers, toys, bibs, bathing products, and a picture book in addition

to the fitted mattress.27 Over 95% of caregivers choose the kit over an alternative cash voucher, indicating its popularity and long-standing place in Finnish culture.23

Its popularity is expanding to other countries, both in the form of public health interventions and commercial products In 2017, Scotland approved the distribution of

baby boxes with a mattress, fitted sheet, clothes, a thermometer, bath towels, and a

changing mat to any mother who fills out a request form at her 24-week perinatal

appointment, at a £6 million annual cost.28 The Finnish baby box has also been cited as an inspiration for products such as the Barakat Bundle, a kit which includes a foldable cradle and sterile delivery supplies for rural Indian mothers, and the Thula Baba Box, a plastic bin for South African mothers to use as an infant bath tub complete with bathing supplies.29-30

In the United States and Canada, the baby box has become a phenomenon largely due to The Baby Box Company, a company that sells baby boxes directly to parents as well as to hospitals for large-scale distribution.31 All boxes come with a mattress and fitted sheet, but can also include various clothes, toys, and diapers for a higher cost; all boxes also come with an online educational course created by The Baby Box Company on SIDS risk factors and safe sleep practices.32

Part of the cardboard boxes’ popularity can be attributed to Dr Meghan Heere’s work at Temple University Hospital In 2016, as director of the well-baby nursery, she set-

up a large pilot study including over 2,500 women who delivered at Temple University Hospital Mothers were surveyed over the phone about bed-sharing and breastfeeding practices within the first week of their hospital discharge 1,264 of these women received

no education safe sleep practices or other resources after delivery; 423 of them received face-to-face education on safe sleep practices prior to discharge; and 391 received a

cardboard box for their infants to sleep in as well as face-to-face education on safe sleep practices Analysis demonstrated that women who received both the cardboard box and the inpatient education reported 27% less bed-sharing with their infant in the first week of life compared to women in the control group, and exclusively breastfeeding mothers reported

nearly 50% less bed-sharing compared to women in the control group Half of the mothers reported using the cardboard box for infant sleeping, with 12% using it as the primary sleeping space; many mothers also reported satisfaction with the box, especially as

proximity to the infant facilitated breast-feeding.33

The cardboard box was deemed a successful intervention based on these results, prompting the creation of the Sleep Awareness Family Education at Temple (SAFE-T) Program at Temple University Hospital This program was created to continue funding the distribution of cardboard boxes and face-to-face safe sleep education from specially trained inpatient nursing staff The boxes are purchased from The Baby Box Company, with funding from donations by Temple University Hospital and the Lewis Katz School of Medicine at Temple University; the SAFE-T program has given out over 10,000 boxes with safe sleep education since 2016.34 Dr Heere’s research efforts are now focused on quality improvement cycles for the SAFE-T program as well as long-term effects on

sleeping practices during the first year of life and Philadelphia’s SUID mortality rate.35

The results of this program, combined with the reputation of baby boxes from Finland, prompted other hospitals in the US and Canada to partner with The Baby Box Company to give out cardboard boxes and a membership for their online safe sleep

education program.36 After year-long pilots, New Jersey and Texas now have universal state-wide programs for every mother who wants to receive a box, which totaled to about 400,000 boxes given away from each state in 2017.37-38 Alabama’s public health

department sponsored 60,000 boxes between 2017 and 2018 with the plan to examine their effect on bed-sharing rates before increasing distribution plans.39 Ohio state government launched a partnership with several Cincinnati hospitals to give out 160,000 boxes in 2017,

and similar pilot programs have launched in Alberta and Toronto, Canada to a few

thousand expecting mothers in 2018.40-42

Despite the growing popularity of baby boxes in the US, many pediatricians,

government officials, and parents have reservations about the use of the cardboard box for infant sleep The AAP has declined to state that cardboard boxes are safe, citing both the lack of evidence in preventing infant deaths as well as the lack of regulation around them.43

Since the boxes do not meet the federal definition of a crib, bassinet, play yard, or

handheld carrier, they are not required to meet the same regulations set by the Consumer Product Safety Commission.44 Experts, including members of the AAP’s Task Force on SIDS, have expressed concern about how popular the cardboard boxes are, especially given their somewhat vague intended use: per the company’s instructions, the box is meant

to be “placed on the floor or a sturdy wide surface, such as a coffee table” and not placed

in the adult bed or used as a carrier, yet the way they are designed easy for parents to do both.45-46 Pediatricians have also raised specific concerns about the durability of cardboard, the lack of visibility in a cardboard box compared to a crib or bassinet, and the risk of injury if the box is placed on the floor or a high surface47; these exact concerns were

echoed by a focus group of mothers when asked interviewed about the cardboard boxes.48

Our Project

Amidst the abundance of controversy, the fact remains that there is limited

evidence on cardboard boxes as an intervention to improve safe sleep practices and SIDS outcomes Their safety and efficacy, especially in populations at higher risk for SIDS, are

of particular research interest as their usage expands into larger academic hospital centers Therefore, we sought to evaluate cardboard boxes as a resource for caregivers at Yale New

Haven Hospital’s Pediatric Primary Care Centers (PCC), a population that has been

previously identified as at high risk for unsafe sleep practices.49 By collecting baseline data about current sleep practices and SIDS risk factors among these caregivers, our objective was to better understand our own community as well as analyze whether the cardboard box would address the same barriers that Dr Heere identified at Temple University Hospital Secondarily, we would collect data on attitudes towards safe sleep practices and

perceptions of the cardboard box itself to understand what caregivers’ response would be if the boxes were to be distributed by the hospital, especially in the context of The Baby Box Company considering a partnership with Yale New Haven Hospital To this end, we

designed a mixed-methods study combining a quantitative survey with a qualitative

interview in order to capture both of these key steps in designing a successful intervention against unsafe sleep and SIDS

Methods

Setting and Sample

The study was conducted at two pediatric primary care clinics in New Haven, Connecticut Our sample included English and Spanish-speaking mothers of infants ages 2

to 16 weeks who presented for well-child visits at our pediatric primary care clinics from June to August 2017 We attempted to approach every family with an infant aged 2 to 16 weeks on any given day in clinic Our inclusion criteria were designed to identify

participants who have experienced a key concept being explored in the study and/or have membership in a subgroup with distinct characteristics; in this case, the subgroup in this study were mothers of young infants, thus the use of the cardboard box would be relevant

to them and would allow them to make salient comments about its usage for the qualitative portion Patients were screened for inclusion / exclusion criteria by Jaspreet Loyal, the primary investigator (JL) on a weekly basis, with the list being passed down to the student Nisha Dalvie (ND) once reviewed and approved by Maryellen Flaherty-Hewitt, the clinic director (MFH) We chose the pediatric primary care clinics to access families at higher risk of not following AAP recommendations for safe sleep, as identified in the background, and were also likely to use our hospital maternity services.49

Data collection

Our mixed-methods approach included in-person surveys and audio-recorded interviews, both of which were performed by the student (ND) with caregivers at their child’s well visits between 2-16 weeks after birth The quantitative survey was adapted from the Infant Care Practices survey, a validated tool administered nationally by the Slone Epidemiology Center.51 Survey data included questions about where the infants sleeps,

infant sleep positioning practices and intentions, bed-sharing practices and intentions, and other risk factors such as cigarette smoking and breast feeding; demographic data collected included age, race/ethnicity, years of education, and health insurance (see Appendix A for full survey) The semi-structured interviews were conducted using a grounded theory

approach, where each new interview was discussed by the research team in order to inform the structure of the next interview.50, 52 An initial interview guide was created based on current literature and expert opinion The interview guide (Table 1) was revised in an

iterative process as new information emerged from the data

Table 1: Semi-structured Interview Guide

Interview question

1 Have you heard of the cardboard box for babies to sleep in?

If no, research associate shows picture or actual box

2 What do you think of the cardboard box?

3 What are some things you like about the cardboard box? What are some things

you dislike about the cardboard box?

4 What do you think you would use it for? (Probing question: Would you use it for your baby to sleep in?)

5 (If participant stated he/she would not use it for their infant to sleep in) The

hospital is planning to give the cardboard box to parents at no cost, what do you think about that? How would this affect your decision to use the cardboard box?

6 Where would you put the cardboard box in your home?

7 Do you have anything else to share?

Verbal was obtained from each participant in English or Spanish by the student (ND) at the time of their appointment The survey was conducted via secure Yale Qualtrics link on an encrypted electronic tablet held by the student (ND), while verbally asking each question to the caregiver in either English or Spanish Surveys were conducted during caregivers’ waiting time in the exam room and took approximately 10 minutes each Semi-structured interviews were conducted by the student (ND) in a private room in the clinic space after the conclusion of the medical visit, each lasting for 15 to 20 minutes During the interview, participants were shown a picture of the cardboard box (Figure 2) or the

physical box itself – at the time of this study, the cardboard boxes were being distributed with a lid Interviews were audiotaped and transcribed verbatim by the student (ND)

Interviews conducted in Spanish were translated into English during the transcription

process by the student (ND), who is a certified Spanish language translator in the Yale New Haven Hospital system Approval from the Yale University Human Investigation Committee

as well as the primary care clinic directors was obtained prior to beginning the project

Figure 2: Image of the cardboard box shown to qualitative interview participants

Data Analysis

Quantitative survey data was exported from Qualtrics to Microsoft Excel 2016 by the student (ND) and organized by anonymous, randomized response ID Variable names were calibrated for further analysis and relationships of interest were identified by the student (ND) before being sent to the primary investigator (JL) who had access to the necessary software Data analysis, including calculated chi-square values, odds ratios, and associated 95% confidence intervals, was completed in SPSS (Armonk, NY) Reported behaviors were compared demographic data to quantify outcomes such safe infant

positioning, bed-sharing frequency, and other practices of interest Unadjusted odds ratios for which the confidence interval did not include 1.0 were considered statistically

significant

Data from the qualitative transcripts were analyzed using coding techniques

common to qualitative research using grounded theory methodology.53-54 Data analysis was conducted in an iterative process, with data collection and analysis continuing concurrently until no new themes emerged (‘thematic saturation’) In the first part of the analysis, an initial code list was created based on the first read-through of transcripts Codes, defined as participant’s words, phrases, or authors’ concept words, served as labels for important

participant data Transcripts were coded by 4 independent investigators: the student (ND), the primary investigator (JL), an expert in the field of safe sleep practices (EC), and a

nursing trainee (VN) Transcripts were then compared and discussed as a group to share reflections and abstract commonalities in the codes each author had assigned From these codes, the initial code list was created This list was iteratively revised using the constant comparative method as new interviews were conducted and coded In the second part of the

analysis, codes were clustered into cohesive categories To reduce redundancy among the categories and to ensure the category linkages were firmly established, all researchers came

to agreement in the coding schema, which was then reviewed for data that expressed the main ideas or themes In the third part of the analysis, data were reviewed for evidence of relationships among themes

Trustworthiness in the data was established through 1) ongoing debriefing sessions

by the authors to discuss reflections, insights and incoming data; 2) coding development over

3 months, enabling prolonged engagement with the data to recognize biases or distortions and 3) member checking during interviews to ensure correct interpretation of what was being shared, and by discussing tentative themes and interpretations with a subset of research participants Data was organized in Microsoft Excel 2016

Results

Overall

Of 129 caregivers approached, 120 caregivers (93%) consented to fill out the

survey Out of the participants who consented to the survey, 50 caregivers (42%) also consented to participate in the semi-structured interview Most of the mothers who did not consent to either the survey or the interview portion cited time as their principal reason for not participating Characteristics of survey and interview participants are shown in Table 2

Sex of infants was almost equally split between male and female in both the total surveyed group and subset who also participated in the interview There was representation

of several infant ages, with 42.5% presenting at their 2 week or 4 week well-child visits, 25.8% presenting at their 4 month well-child visit, and 31.7% presenting in between those visits; distribution was comparable in the interview-participant subset 59.2% of

respondents identified as mothers and the primary caregiver of their infant, compared to 35% of respondents identifying equal caregiving between mother and father and a small group (5.8%) identifying as non-parent caregivers, consisting of grandparents, an aunt, and

a non-relative In the interview-participating group, 90% of participants were mothers who identified as the primary caregiver; only 2 mother-father pairs were interviewed, as well as one grandmother Caregiver age was nearly equally distributed between younger than 30 years and older than 30 years, with no caregivers younger than 20 years For 26.7% of caregivers, the infant at the appointment was their first child compared to 73.3% with at least one other child at home In the interview-participating subset, this distribution was similar with 36% respondents having only one child and 67% having more than one

Table 2: Demographics of Participants (Total N = 120, Interview N = 50)

(Percent)

Interview Number (Percent) Infant Sex

Infant’s age (weeks)

Less than 1 month

Caregiver’s country of birth

United States incl Puerto Rico

Highest Education Level of Caregiver

Less than high school

Breastfeeding Status of Caregiver

Mostly or only breastmilk

Equal breastmilk and formula

Mostly or only formula

A Other includes grandparents (5), an aunt (1), and a non-relative such as babysitter or friend (1)

B Includes Ecuador (9), Mexico (9), the Dominican Republic (1), El Salvador (1), Guatemala (2), Honduras (2), Trinidad (1), Spain (2), Albania (1), Greece (1), Barbados (1), Jamaica (1), Grenada (1), China (1), and Togo (1)

C Includes Pacific Islander (1) and Native American (1)

Of 120 total participants, 50% identified as black and 41.7% identified as Hispanic, compared to 4% of participants who identified as white In the interview-participating subset, 47% of respondents identified as black and 28% identified as Hispanic and

compared to 16% of participants who identified as white The majority of participants (71.7%) were either from the continental United States or Puerto Rico compared to 28.3% from various other countries in Central America, South America, Europe, and Asia;

distribution was similar in the interview-participating subset The majority of participants identified English as their primary language (70.8%) compared to 29.2% of primarily Spanish speakers For the interview portion, 76% of interviews were conducted in English and 24% were conducted in Spanish

Of 120 participants, 45.8% reported their highest level of education was high school or equivalent compared to 16.7% with less than high school education and 37.5% with at least some college education; distribution was similar in the interview subset Most participants reported they had never smoked (83.3%) compared to 12.5% who quit before pregnancy and 4.2% who smoked during pregnancy or currently; distribution was similar

in the interview subset Of 120 participants, 42.5% reported mostly or exclusively

breastfeeding compared to 34.2% who used mostly or exclusively formula and 23.3% who did an equal mix of breastfeeding and formula feeding In the interview subset, 50% of participants reported mostly or exclusively breastfeeding compared to 20% who mostly or exclusively formula fed and 30% who reported and equal mix of formula and

breastfeeding

Sleep practices

Survey participants were asked about how they position their infant to sleep (on the back, on the side, or on the stomach), where their infant sleeps (free text response that was grouped into categories), and the environment their infant sleeps in (i.e with a firm

mattress, with a swaddle blanket, or with other items such as thick blankets, pillows, or toys.) In the context of each of these behaviors, they were also asked to identify their initial plans for sleep practices before bringing their infant home, the most commonly occurring practice since bringing their infant home, and any other practices that sometimes occur; these are signified by “Intended Practice”, “Most Common Practice” and “Practice occurs Sometimes” respectively in Table 3

Table 3: Intended and Most Commonly Practiced Sleep Behaviors (Total N = 120)

Intended Practice A (% of Total)

Most Common Practice (% of Total)

Practice occurs Sometimes A (% of Total) Positioning

all responses is greater than total N of 120

responses is greater than total N of 120

Participants that intended a specific practice and practiced it exclusively versus participants that intended a specific practice but ended up also practicing other behaviors were categorized in Table 4; participants whose current practices were not included in their initially intended practices are also listed in Table 4 Any participant whose current

practice differed from their intended practices were asked to explain the reason for the change in a free text option Participants were only asked about intended behaviors in the context of sleep position and location, so no data on intentions for sleep environment could

be organized Participants were also asked about intentions for sleeping in the same room, but not the bed, as their infant, but the majority of these responses had complicated

explanations based on number of caregivers in the home, presence of other children, and infants’ ages, so these results were not further organized or analyzed

Table 4: Changes in Intentions: Exclusive Practices, Non-Exclusive Practices,

and Divergent Practices (Total N = 120)

Practice was Intended and Occurs Exclusively (% of Intended Practitioners)

Practice was Intended but Does Not Occur Exclusively (% of Intended Practitioners)

Practice was Not Intended but is Now Practiced (% of Non-Intended Practitioners)

all responses is greater than total N of 120

Sleep positioning

Out of 120 participants, 105 caregivers (87.5%) reported supine positioning as their most common practice compared to 15 caregivers (12.5%) who most commonly practiced prone or side positioning, but a total 37 caregivers (30.8%) reported non-supine

positioning at least sometimes Only 78 caregivers (65% of total participants and 75.5% of those who intended to practice supine positioning) practiced supine positioning

exclusively, leaving 25 caregivers who intended to practice supine positioning but also practiced side and prone positioning Of 77 caregivers whose plans only included supine positioning, 5 caregivers ended up switching to side positioning (6.5%) and 14 caregivers ended up switching prone (18.2%) Some reasons these caregivers cited for their switch included “feeling that [their baby] had a preference for [their] stomach”, “getting advice that [their baby] might choke [in supine position]”

Of 120 participants, 43 caregivers (35.8%) reported that non-supine positioning was part of their intended practice (20.8% planned on side positioning compared to 15% who planned on prone positioning) Of these 43 caregivers, only 4 ended up switching to supine positioning (23.5%); these 4 caregivers all cited education about safe sleep from their pediatrician as their main reason for switching Unadjusted odds-ratio analysis for association between demographic data and supine versus non-supine positioning is

outlined in Table 5 Association was considered statistically significant if the 95%

confidence interval (CI) did not include 1.0; based on this criteria, caregivers born outside

of the US (including Puerto Rico), caregivers of Hispanic ethnicity, and caregivers with less than high school education, and primarily Spanish-speaking caregivers were

associated with greater odds of non-supine positioning

Table 5: Factors Associated with Non-Supine Positioning (Total N = 120)

Characteristic Any Non-Supine

Positioning N=42

Only Supine Positioning N=78

Unadjusted OR (95% CI) A

Caregiver

One child in the home 11 (26.2%) 21 (26.9%) (0.3-1.8) 0.7

Non-US / Puerto Rico

4.2 (1.8-9.6)

Less than high school

Any tobacco smokingC 7 (16.7%) 13 (16.7%) (0.5-2.1) 1.0

A OR indicates Odds Ratio; CI indicates Confidence Interval

B Includes mothers who co-identified as primary caregivers with fathers

C Defined as past or current tobacco smoking

Sleep location

Out of 120 survey responses, 115 caregivers (95.8%) indicated the crib or bassinet

as their infant’s most common sleeping space, leaving 1 caregiver (0.8%) who identified their infant’s usual sleeping space as a Pack and Play, 2 caregivers (1.7%) who identified a Moses basket or similar product designed to go in the adult bed as their infant’s usual sleeping space, and 2 caregivers (1.7%) who identified bed-sharing in an adult mattress or sofa as their infant’s usual sleeping space 45 of these caregivers (37.5%) exclusively placed their infant in a crib or bassinet to sleep, leaving 85 caregivers (62.5%) who

planned to place their infant in a crib or bassinet but at least sometimes placed their infant elsewhere No caregivers identified Pack and Plays or car seats as part of their child’s intended sleep area, but 41 caregivers (34.2%) reported using them occasionally

Out of 120 respondents, 25 caregivers (20.7%) included bed-sharing in their

intended practices However, a total of 46 caregivers (38.3%) reported actually bed-sharing

at least some of the time Out of 95 caregivers that did not plan to bed-share, 9 respondents (7.5% of total and 9.5% of caregivers that did not plan to bed-share) ultimately practiced bed-sharing at least some of the time, citing convenience with breast-feeding or wanting to spend more quality time with their child as their main reasons for switching

Odds-ratio analysis for association between demographic data and bed-sharing is outlined in Table 6 Association was considered statistically significant if the 95%

confidence interval (CI) did not include 1.0; based on this criteria, only caregivers who identified Spanish-as their primary language were associated with greater odds for bed-sharing

Table 6: Factors Associated with Bed-Sharing (Total N = 120)

Characteristic Any Bed-sharing

N=46 No Bed-sharing N=74 Unadjusted OR (95% CI) A

Less than high school

Any tobacco smokingC 8 (17.4%) 12 (16.2%) (0.4-2.9) 1.1

A OR indicates Odds Ratio; CI indicates Confidence Interval

B Includes mothers who co-identified as primary caregivers with fathers

C Defined as past or current tobacco smoking

Qualitative Themes

We identified 3 major themes: (1) safety, (2) appearance and (3) planned use Themes, subthemes, and exemplar quotes are compiled in Table 5 with additional quotes in the text below These results are also reported in a published study.55

1 Safety

Participants expressed concern with the safety of the cardboard box material as not being sturdy enough and that the lid had the potential for suffocation if the cardboard box was left covered while the baby was inside Another concern was the possibility of the infant rolling out of the cardboard box in their sleep In contrast, others felt that the

cardboard box would be safe for the baby For example, one participant discussed her

friends’ usage of a similar product which influenced her perspective She stated, “In my group there are moms who have baskets for the baby in their beds So, this is like that so I will use it.” Another participant valued a physician’s recommendation, “Well if the doctors give it to you, you’re supposed to use it right? Like if this is safer than a crib or whatever then I would prefer to use that.”

2 Appearance

For some participants, the simplicity of the design was appealing One participant

stated, “It’s cute, I like it.” For another participant, the cardboard was unattractive She stated, “It’s not appealing at all, it literally looks like a box that you would…use to mail something.” When the physical box was brought to clinic, many participants commented

on its apparent bulkiness Some participants compared the cardboard box to existing

products like cribs or bassinets, which were perceived as being more socially acceptable than the cardboard box Some participants associated use of the cardboard box with a

negative social status: “You look like you’re giving away your baby…it looks like a dog box.”

3 Variation in Planned Use

When participants were asked how they would use the box, some planned to use it in

their bed One participant stated, “We could have it in the bed, it could go in the middle”

Another participant said she would only use the box for storage, and one participant

thought of using it during the day, stating “I think it is also nice to use during the day when he’s playing, and I’m trying to keep watch on him.” Participants commented on the ease of

travel with the cardboard box and the convenience of having the cardboard box close by when breastfeeding Some participants were excited at the prospect of receiving the

cardboard box and contents at no cost For some participants who initially stated they wouldn’t use the cardboard box, the no cost provision changed their attitude and many stated they would use the product if given at no cost Other participants stated that they had already purchased cribs and therefore had no use for the cardboard box

Perceptions of the Cardboard Box, Demographics, and Sleep Practices

Out of 50 interview participants, 26 (52%) ultimately said they would use the cardboard box for their infant to sleep in, 21 (42%) said they would not use it for their infant to sleep in, and 3 (6%) were unsure For the interview-participating subset, analysis was performed to determine if parents who stated they would or would not use the cardboard box for their infant to sleep in had similar demographic factors or reported similar sleep behaviors – results are shown in Table 9 No association based on unadjusted ORs was found between indicated interest in using the cardboard box for infant sleep and any demographics, including those known to be SIDS risk factors, nor with unsafe sleep practices

Table 8: Caregivers’ Perspectives on Cardboard Boxes: Themes and Subthemes

“He rolls around his crib all the time so I feel like

he would just fall out of that.”

“ I don’t know if it’s made out of cardboard, that’s not safe either, it could always split.”

“I would not buy that product out of a store, it looks just like a cardboard box.”

It would be like the baby is, I don’t want to say, homeless or something? Like something out of the shelter?

“It’s cute, it looks like the bassinet without the legs.”

“That’s a big box It looks uncomfortable to carry.”

“So yeah it’s a good idea, it’s convenient.”

“The mattress is so thin too.”

PLANNED USE When traveling

Breastfeeding

In bed

Storage During playtime Cost

Necessity

“You can move it wherever you want.”

“Say if we’re at someone’s house and I don’t want

to put her in someone else’s bed, I would use that.”

“It’s much better for breastfeeding at night when the baby is sleeping next to you.”

“It’s perfectly sized for the bed, which is great, because sometimes we want to lay on the bed with him, but we’re scared we’re going to turn over or something.”

“I guess it might be good for storage but I would never let my baby sleep in that.”

“I think it is also nice to use during the day when he’s playing.”

“I think it’s helpful for the people that can’t afford

he cribs and stuff like that.”

“Well we had already bought her a crib so if it was given to me now I wouldn’t use it.”