The potential effects of a smart crib on infant crying and sleeping, and maternal sleep and downheartedness

(1)

Faculty of Social and Behavioural Sciences

Graduate School of Child Development and Education

The potential effects of a smart crib on infant crying and sleeping, and maternal sleep and downheartedness

Research Master Child Development and Education Research Master Thesis

Marije van Meegen Roos Rodenburg 11-12-2018

(2)

2 Abstract

Objective: Excessive infant crying is associated with reduced infant sleep, and with maternal sleep deprivation and post-partum depression (PPD). This study investigates the potential effects of a smart crib on infant crying and sleep, and on maternal sleep and

downheartedness.

Methods: Twelve mother-infant dyads were recruited from the general community. Mothers kept daily diaries on infant crying and sleep, own sleep and downheartedness – signaling risk for PPD - during baseline (A) and intervention phase (B). Simulation Modeling Analysis (i.e., idiographic effects) and Multilevel Data Analysis (i.e., group effects) were used to determine the effectiveness of the smart baby crib.

Results: SMA analysis showed that seven out of twelve infants reduced crying and two out of twelve infants improved sleeping. Maternal sleeping increased in three out of twelve mothers. One mother experienced a trend of less downheartedness during the intervention phase. The multilevel analysis identified a significant decrease in infant crying, a significant increase in infant sleep and a significant increase in maternal sleep. No effects were found for downheartedness.

Conclusion: The smart baby crib was associated with reduced crying and increased infant and maternal sleep, and as such the smart crib may help to break the vicious cycle of infant crying and parental exhaustion.

(3)

3 Background

Infant crying is a normal aspect of physiological and neurobehavioral development (Brazelton, 1986; Kurth, Kennedy, Spichiger, Hösli, & Zemp Stutz, 2011). From an evolutionary perspective crying is seen as "The Crying Reflex" via which infants communicate their basic needs (Brazelton, 1986; Lester & Zacharia Boukydis, 1985).

Through this signaling behavior, infants stimulate the parent to come closer. Therefore, infant crying is regarded as an early attachment behavior (Bell & Ainsworth, 1972).

A wide variety of definitions for excessive crying have been suggested. This study refers to excessive crying as the parental perception of unexplained and uncontrolled excessive crying in a healthy infant (Hiscock et al., 2014). Infant crying starts two weeks after the infants' original due date (Karp, 2015). Prevalence rates of excessive infant crying and fussiness varies weeks between 17% to 25% in the first six weeks, and decreases to 11% at 8-9 weeks (Wolke, Bilgin, & Samara, 2017). Excessive crying and fussiness drops to 0.6% among infants around 12 weeks. These prevalence rates vary across countries: Dutch, British, and Canadian infants more cry and fuss more compared to infants in German, Japanese or Denmark (Wolke et al., 2017). Excessive crying and fussiness is related to infant sleep

problems and is the common cause for parents to visit health professionals (Barr, 1998; Kurth et al., 2011). Between 14% and 30% of parents are affected by excessive infant crying

(Yalçın et al., 2010).

No consensus exists on the etiology of excessive crying (Kaley, Reid, & Flynn, 2011). To date, there are four widely used theories to explain the causes of excessive crying. From a medical viewpoint, it is hypothesized that gas, cramps, acid reflux (which induces pain), or food intolerance may cause excessive crying (Karp, 2015; Savino, 2007). Second, excessive crying is explained from a developmental psychology perspective, which assumes that

(4)

4 undeveloped self-regulatory behaviors and the still-developing circadian rhythm cause

excessive crying (Kaley et al., 2011). Third, from a transactional perspective, maternal

anxiety, depression, or parental stress are assumed to cause excessive infant crying (Landgren & Hallström, 2011; Petzoldt et al., 2014). However, the reverse relationship is also

considered: that excessive crying might be the cause of maternal anxiety, depression or parental stress (Kurth et al.,2011). Fourth, it is hypothesized that infants are born three months premature and as a result cry excessively (Karp, 2015). This assumption stems from the missing fourth-trimester theory.

The missing fourth-trimester theory assumes that infants are not fully developed and cannot cope with the stresses in the extrauterine environment. That is, in the intrauterine environment, the fetus experiences constant calming stimuli, which causes a “calming reflex” (Karp, 2015). This is a neonatal reflex consisting of tactile, vestibular, and auditory reflexes. This suggests that imitating the intrauterine environment would calm excessive crying. Furthermore, it has been found that infants adjust their respiratory sinus arrhythmia levels to their mothers' levels during the first two months after birth. This might indicate a continuing intrauterine effect during the first months of life. For infants passing three months, less dependence on intrauterine stimuli indicates a first developmental shift towards social orientation (Van Puyvelde et al., 2015). Thus, in the first phase of life infants are still in an adaptive period from the intrauterine to the extrauterine environment, which suggests that mimicking the intrauterine environment might reduce crying. Infants might, therefore thrive well with stimuli that represent the intrauterine effect. The maternal-infant physiological relatedness is indispensable here (Van Puyvelde et al., 2015).

(5)

5 After birth, the calming reflex can be triggered by the parents through the five steps of The Happiest Baby-method (THB; Karp, 2015). The following five steps mimic the calming uterine stimuli: 1. Swaddling, 2. Side or stomach position while awake; 3. Shushing, 4.Swinging, and 5. Sucking. An RCT revealed that receiving THB resulted in reduced pain scores and a decreased duration of crying in infants at two and four months, compared to other infants receiving soothing as usual, or receiving sucrose water after immunization (Harrington et al., 2012). Another RCT of Paul and colleagues (2011, 2016) showed that using THB significantly increased infant sleep duration at 16 weeks of age when compared to controls. However, these RCTs were not conducted among excessively crying infants. A study that explored the individual effects of THB among excessively crying infants showed similar promising results in infant sleep and crying (Möller, Oort, & Rodenburg, submitted). Resuming, current evidence suggests that THB may be a promising intervention for

prolonging sleep and decreasing crying in excessively crying infants until six months of age. Excessive infant crying also has a negative impact on the mother, leading to a vicious circle of exhaustion in the infant and mother. The interplay between excessive crying and sleep disturbance leads to increased daytime sleepiness and exhaustion for the mother (Filtness, MacKenzie, & Armstrong, 2014). Between 18% and 64% of postpartum mothers experience exhaustion or sleeping problems (Schytt, Lindmark, & Waldenstrom, 2005). Exhausted mothers are at an increased risk for postpartum depression (PPD) (Dørheim, Bondevik, Eberhard-Gran, & Bjorvatn, 2009; Goyal, Gay, & Lee, 2009; Kurth et al., 2011). Prevalence rates for PPD vary from between 20% and 60% in Western developed countries (McCoy, Beal, Shipman, Payton, & Watson, 2006; Halbreich & Karkun, 2006). Consequently, mothers who suffer from exhaustion and PPD may become less sensitive and

(6)

6 less responsive to their infant’s needs. They may ultimately show inadequate parenting, which may further enforce infant crying (Field, 2010; Kurth et al., 2011).

In sum, exhaustion and symptoms of depression may exacerbate infant crying and sleep problems, leading to a vicious circle of exhaustion in the infant and mother (Kurth et al., 2011). This vicious circle may be broken by improving infant crying and sleep. This, in turn, may contribute to increased maternal sleep, which may reduce the risk for maternal PPD (Kurth et al., 2011).

A range of interventions have been conducted to increase infant and maternal sleep. A recent meta-analysis revealed positive effects of non-pharmacological sleep interventions, on maternal report of subjective sleep quality and on infant sleep, but no effects were found for maternal depression. Effect sizes appeared to be larger when interventions were carried out earlier in the postpartum period (Owais, Chow, Furtado, Frey, & Van Lieshout, 2018). However, this meta-analysis did not take into account infant crying in the context of maternal sleep. Targeting infant crying and infant sleep is important, because infant crying and sleep are related to maternal sleep and consequently exacerbate PPD. Behavioral interventions targeted at infant sleep in the first six months of life were not effective in reducing parental distress and did not prolong infant sleep (Douglas & Hill, 2013).

Although a responsive intervention such as THB has been associated with positive effects, it may be that some mothers with PPD do not respond to the intervention because of their feelings of depression and exhaustion. Responsive interventions help parents tailor their parenting to an affective–emotional style with positive affection and high levels of warmth and nurturance (Eshel, Daelmans, de Mello, & Martines, 2006). In clinical practice mothers withdraw from treatments (Möller, Oort, & Rodenburg, submitted), and this may also be the case in studies into infant crying and maternal mental health (Priddis, Keedle, & Dahlen,

(7)

7 2018). Applying an intervention might be too exhausting for mothers during the postpartum period.

In this study we examined a new intervention, a smart crib (from now on: crib) that was developed to soothe infant crying, fussiness and to ameliorate sleep (Karp et al., 2015). This crib comprises three components of the happiest baby-method: 1. The infant is swaddled in a swaddling sack; 2. It produces white noise that is comparable to shushing: 3. It gently rocks the infant. The crib responds to infant crying with increasing levels of white noise and rocking. When the infant is soothed, the crib reduces to the baseline level of sound and movement (i.e., baseline level).

Our research is in line with former studies. Möller, De Vente, and Rodenburg (2017) examined whether the crib was capable of inducing a calming response in an experimental study (Esposito et al., 2013). The group using the crib was compared to a group using THB. Behavioral observations and heart rate indicated that fussiness was significantly decreased in both treatment arms. This revealed that infants have a calming response (Möller, De Vente, & Rodenburg, 2017).

However, the crib and its effects on infants has not yet been investigated in a natural environment (i.e., at home). Investigating interventions that can soothe infants at home might prevent infants from being hospitalized, and it is known that hospitalizations can negatively affect parent-infant attachment (McKenzie, 2013). We used a single case experimental design (SCED; Smith, 2012) with time series to evaluate the effects of the crib on infant crying, infant sleep, maternal sleep, and downheartedness. SCED is valuable for testing intervention in a population with clinical problems. With SCED it is possible to test the functional causal relationship between the independent variable (i.e., the treatment) and the dependent variable (i.e., our variables of interest) (Cohen, Feinstein, Masuda, & Vowles, 2014). SCED thus

(8)

8 allows to systematically evaluating whether an intervention works for a specific individual (Branch & Pennypacker, 2013). Therefore it gains insight into the process of change (Borckardt et al., 2008). In the current study, we will refer to the changes in individuals as idiographic effects. Besides studying idiographic effects, group results will be analyzed using multilevel analysis.

Our study aims to evaluate the effects of the crib on infant crying, infant sleep, maternal sleep, and downheartedness. We will do this by evaluating idiographic, and multilevel effects using Simulation Modeling Analysis (SMA) and multilevel analysis. The following research questions will be examined: 1) After using the crib at home, does daily infant crying decrease? 2) will infant sleep increase? 3) After using the crib at home, will maternal sleep increase and 4) will feelings of maternal downheartedness decrease? It is hypothesized that after using the crib: 1) infant crying decreases; 2) infant sleep increases; 3) maternal sleep will increase and 4) downheartedness will decrease. Hypotheses will be tested by multilevel analyses only.

Method Participants

Participants were recruited via Rondom de Geboorte, an organization for infant and youth health care in Amsterdam and via social media. The sampling procedure is displayed in Figure 1 in appendix A. The sample in the current study consists of N= 12 mother-infant dyads. Mean gestational age was 39.40 weeks (SD = 1.20) and mean age was 73.92 days (SD= 44.27) at study inclusion. 41.7% of the infants were breastfed or formula fed, and 16.60% were both breast- and formula fed, or given a different type of feeding. 58.3% of the infants was the first born child of the mothers.

(9)

9 Inclusion criteria were: parents who perceived infant crying as excessive, and infant and maternal sleep as problematic and feeling downhearted. Infants were between 0 and five months of age. Infants were excluded if the crying was due to a medical condition. Mothers signed an informed consent before the start of the study. The ethical committee of the University of Amsterdam (UvA) granted approval for the conduct of this study.

Procedure

See Figure A2 for an overview of the design of the study. We used daily diaries to evaluate infants crying, infant sleep, maternal sleep, and downheartedness. We reached out to the parents to schedule the first home visit and brought the crib to the mother. During the first home visit, parents received additional information about the study and on how to complete the daily diaries. We provided mothers with time rulers and the Baby Connect App, Avent u or Grow Baby App to facilitate recordings of the daily diaries.

The baseline phase consisted of a minimum of 3 to a maximum of 8 days. During this period, mothers kept daily diaries and evaluated infant crying, infant sleep, maternal sleep, and downheartedness. During the second home visit, within 3 to 7 days after the first home visit, the crib was delivered to the parents’ home. Parents received instructions on how to use the crib and continued to score daily diaries during the intervention phase of at least 21 days. In addition, during the interventions phase mothers also was used on a daily basis (Table B2). Three days after the second home visit the researcher contacted the parents to discuss their experiences with the crib. The third home visit took place at the end of the intervention phase, three weeks or more after the second home visit. At this home visit parents returned the bed to the UvA. To facilitate the ending of the intervention phase, parents received information on the 5S’s (Swaddling, Side or stomach, Shhh or white noise, Swinging, and Sucking) and on how to apply them. To ensure use of the 5S's was as intended, parents were given a guide

(10)

10 and a video with instructions. In addition, parents were taught behavioral techniques to ensure the enhancement of (save) sleep.

The intervention

The crib provides swaddling, movement, and white noise to trigger a calming response. It responds to infant crying with increasing levels of sound and movement. The crib has a baseline level with gentle movement and low levels of white noise and can increase up until level four, with more dynamic movement and a higher level of white noise. It automatically stops after three minutes if continues, to signal parents to attend to their infant's needs, such as changing the diaper, cuddling, or a feeding. If the crying stops within 3 minutes, the crib will automatically return to the baseline level. Accompanied by the crib, parents received a swaddle sack, in which the infant is swaddled. The sack, needs to be attached to sensors on the crib. This is to assure that the infant will sleep on his/her back and to avoid rolling over. The sensors ensure that the crib will not start rocking when the infant is not safely clipped into the crib. Parents can regulate the bed with the corresponding app, to attend to their infant’s needs. Measurements

Daily diaries. Mothers kept online daily diaries to record infant crying, infant sleep, own sleep, and downheartedness (Table B1 and B2). The daily diaries were based on the model of Barr, Kramer, Boisjoly, McVey-White, and Pless (1988). The items of the daily diaries are displayed in Table 7 and 8. Questions reflecting downheartedness were drawn from the Profile of Mood States (Baker, Denniston, Zabora, Polland, & Dudley, 2002). Additionally, the daily intervention diary also asked whether mothers used the crib and in what frequency.

(11)

11 To evaluate intervention effects on infant crying, infant sleep, maternal sleep, and downheartedness, the daily measurements were analyzed using two procedures: Simulation Modeling Analysis (SMA; Borckardt et al., 2008) and Multilevel Analysis.

SMA uses a bootstrapping methodology to determine the treatment effects (i.e., phase-effect sizes) between baseline and intervention phase while accounting for

autocorrelation (Borckardt & Nash, 2014). SMA is developed for time series data, which allows the investigation of overtime changes in the participant (Smith, 2012). This is achieved by calculation and comparison of symptom scores across treatment phases.

SMA calculates phase-effects sizes which equal mean-level differences and slope differences. These values are expressed in Pearson's r. Mean level differences refer to level changes and intervention effects. Moreover, we evaluated slope changes between the baseline and intervention phase to examine the trend of improvement in our outcome variables. Slope changes provide an evaluation of the trend-change within a specific phase (Borckardt et al., 2008).

We conducted Multilevel Modeling to evaluate group level effects of the crib on each of the four outcomes. Multilevel models are used to analyze nested data. Nested data refer to lower level units which are nested within the higher-level units. In the current study, repeated measurement occasions at level two (i.e., daily diaries) are nested within the individual mother-infant dyads at level one. Due to the repeated measurements, the data are not independent of each other thus, violating the assumption of independence. Multilevel is robust against the violation of the assumption of independent observations (Snijders & Bosker, 2012). Moreover, multilevel models can deal with unbalanced data structures. Thus, if data are missing at random or entirely at random, participants of whom not all

(12)

12 measurements occasions are available can still be included in the study (Snijders & Bosker, 2012).

Two-level hierarchical models were computed for each of the four outcomes: (1) infant crying, (2) infant sleep, (3) maternal sleep, and (4) downheartedness. Level-one predictors reflect a specific measurement occasion, and level-two predictors remain the same at each of the different measurement occasions. We found intra-class correlation coefficients (ICC) between .21 and .54, which indicated that observations were clustered between individuals. To deal with this clustering, multilevel analysis was used (Table 1, Snijders & Bosker, 2012). The pre-set alpha level of p < 0.05 was used to determine significance. Lastly, we added a time code parameter to the multilevel models to be able to compare baseline to the intervention phase. As a result baseline measurements received negative, ascending values until the last day of the baseline period. This day received a zero, indicating the intercept. All intervention measurements received positive ascending values.

Our regression equation was: ŷ = β0 + β1 ∙ treatment + β2 ∙ timecode + β2 ∙β3 ∙ timecode + treatment + timecode ∙ measurement occasion. β0 reflects the last day of the baseline phase. β1 reflects the first day of the intervention phase. β2 represents the slope of the baseline phase. β3 is the changein the slope during the intervention phase. To calculate the actual slope of the intervention phase: β4 = β3+ β2. Analyses were performed using the

Statistical Package for Social Sciences (SPSS), Windows version 25. Results

Descriptive statistics of our variables of interest are displayed in Table 1.

Idiographic effects. Results of our variables of interest per mother-infant dyad are displayed in Tables 2 to 5. Graphs of the idiographic results are displayed in appendix C. SMA revealed a significant decrease in daily crying for five out of twelve infants (Baseline vs Intervention,

(13)

13 ranges: r = -.35, p = .041 to, r = -.71, p = .011). In addition, slope effects revealed a decreasing trend of daily crying during the intervention phase for four out of twelve infants (Baseline vs Intervention, ranges: r = -.46, p = .024 to, r = -.70, p = .001).

Furthermore, we identified an increase in infant sleep in three out of twelve infants (Baseline vs Intervention, ranges: r = .39, p = .022 to, r = .68, p = .005). In addition, slope effects revealed an increasing trend of sleep during the intervention phase for one infant (Baseline vs Intervention, r = .60, p = .044).

SMA showed an increase in maternal sleep for three out of twelve mothers (Baseline vs Intervention, ranges: r = .39, p = .022 to, r = .68, p = .005). Slope effects revealed an increasing trend of sleep during the intervention phase for one mother (Baseline vs Intervention, r = .60, p = .044).

No intervention effects were identified for downheartedness. However, for one mother we found a decreasing trend in downheartedness during the intervention phase (Baseline vs. Intervention, r = -.58, p = .014).

Multilevel results

Preliminary analysis. We checked the values of the skewness and the kurtosis of all the outcome variables to evaluate the assumption of normality. There was no evidence against the assumption of normality since the values of the skewness and kurtosis for all outcome variables of interest were smaller than two times their standard error (Table 1). We checked the QQ-plots of the level-one residuals of our multilevel models. There was no evidence for violations of this assumption for infant sleep, maternal sleep, and downheartedness. The QQ-plot of the level-one residuals of infants crying showed deviations to the left side of the normal distribution.

(14)

14 We computed log-transformations of minutes of crying and checked whether the distribution improved. The normal distribution did improve after the transformation. However, the models of the untransformed and transformed outcome yielded similar results. In light of interpreting our results, we used the untransformed data in our analysis.

Table 6 presents the results of the multilevel analysis for all outcomes variables. Graphs of the group results are displayed in appendix D. All final models contained five parameters: (1) intercept of the baseline phase, (2) slope of the baseline phase, (3) intercept of the intervention phase, (4) slope of the intervention phase, (5) change in the slope of the intervention phase.

Infant crying decreased during the baseline phase with β = -12.91 (p =.058) minutes on average per day. On the last day of the baseline phase, infants cried β =175.42, p =.000, minutes on average. On the first day of the intervention phase, this decreased with β = - 40.77, p =.000, to 134.65 minutes. During the intervention phase, this decrease continued with 2 minutes on a day to day basis.

Infant sleep decreased during the baseline phase with β = - 30.18, p =.001 minutes on average per day. On the last day of the baseline phase, infants slept β = 683.33, p =.000 minutes on average. At the start of the intervention phase, infants slept β = 80.32, p = 003, minutes more compared to the last day of the baseline phase. During the intervention phase, infant sleep increased with 1.65 minutes on average per day.

Maternal sleep increased during the baseline phase with β = 3.31, p =.628, (ns) minutes on average per day. On the last day of the baseline phase, maternal sleep had increased to β = 380.19, p =.000, minutes on average. This decreased with β = - 8.99 p = .655, to 371.20 minutes on the first day of the intervention phase. Maternal sleep increased with 2.98 minutes per day across the intervention phase.

(15)

15 Downheartedness increased during the baseline phase with β = .01, p =.937, ns on average per day. On the last day of the baseline phase, feelings of downheartedness increased with β = 2.44, p =.000 on average. The score of downheartedness increased further with β = .04, p =.858, ns On the first day of the intervention phase. Downheartedness did not increase nor decrease during the intervention phase.

Discussion

This study aimed to examine the effects of a smart baby crib on infant crying, infant sleep, maternal sleep, and downheartedness in mother-infant dyads with excessively crying babies. To our knowledge, this was the first study presenting idiographic and multilevel analysis results to evaluate the effects of the crib. In this study, the crib was associated with a decrease in infant crying and an increase in infant and maternal sleep. Below, we discuss our hypotheses one by one.

It was hypothesized that the crib would decrease infant crying. Our multilevel results were in line with our hypothesis: from baseline to intervention, infants cried 40 minutes less, and crying decreased further with 2 minutes per day across the intervention phase. This is in line with other studies that used physical responsive interventions to soothe infant crying (Harrington et al., 2012; Möller, oort., Rodenburg, submitted; Paul et al., 2016; Paul et al., 2011). This might be caused because these components offer a womb-like sensation that addresses the intrauterine effect and causes a calming reflex (Karp, 2015; Möller, De Vente., Rodenburg, 2017). Idiographic results revealed that eight infants cried more than 120 minutes per day during the baseline period, which is more than would be expected (Wolke et al., 2017). In the light of a natural decline in crying it is expected that it would decrease, within 2-3 weeks to 75 minutes of crying per day. Among these 8 infants crying decreased too much

(16)

16 less than 75 minutes per day, which suggests that the crib had amplified the natural decline in daily crying.

We expected that the crib would ameliorate infant sleep. Our hypothesis was

confirmed with the multilevel results. From baseline to intervention, infants slept 80 minutes more, which improved with an increase of almost 2 minutes per day, across the intervention phase. This is in line with other studies that used the calming steps of The Happiest Baby method (Harrington et al., 2012; Möller et al., 2017a; Paul et al., 2016; Paul et al., 2011). This suggests that the reduction in crying might be ascribed to the three calming components of the smart crib: swaddling, jiggling, and soothing sounds, which trigger the calming

response. (Karp, 2015; Möller, De Vente., Rodenburg).

In addition, the idiographic results identified that infant sleep improved between baseline and the intervention phase for seven out of twelve infants. This effect was significant for two out of twelve infants. For five out of twelve infants sleep duration decreased; one infant experienced a significant decrease in sleep duration. Two out of five infants with decreased sleep duration were 20 weeks or older, which suggests that their decrease in sleep might be due to emerged circadian rhythm (Davis, Parker, & Montgomery, 2004).

Another explanation might be a reduced susceptibility to the stimuli provided by the crib, because the first bio-behavioral shift takes place, around 3-4 months and the intrauterine effect disappears (Van Puyvelde et al., 2015).

It was hypothesized that using the crib at home would increase maternal sleep as well. Our multilevel analysis supported this hypothesis. Maternal sleep improved with an increase of almost 3 minutes from day to day during the intervention phase. In the idiographic

analysis, we found that maternal sleep improved between baseline and intervention phase in ten out of twelve mothers. This effect was significant for two mothers. This partially adds to

(17)

17 the model of Kurth et al., (2011): if infant crying and fussiness and impaired sleep improves, this may positively affect maternal sleep maternal tiredness and PPD. However, three

mothers slept less (nonsignificant) during the intervention phase. Perhaps adding relaxation techniques next to using a smart crib for parental sleep would be beneficial for these mothers. It has been suggested the sympathetic nervous system is aroused after nocturnal awakenings. Hence, parents are unable to unwind and are unable to go to sleep after nocturnal awakenings (Bootzin & Epstein, 2011; Whittingham & Douglas, 2014).

We expected that the crib would decrease maternal downheartedness. The multilevel analysis did not confirm our hypothesis. From baseline to intervention there was a slight nonsignificant increase in downheartedness. No effect was found across the intervention phase. The idiographic analysis revealed a decrease in downheartedness for three mothers during the intervention phase. 9 out of 12 mothers experienced a nonsignificant increase in downheartedness. This lack of effect in downheartedness might be due to several reasons. Perhaps our intervention phase was too short for the intervention to take effect. Studies that found effects for downheartedness generally lasted between one and three months (e.g., Fisher, Feekery, & Rowe, 2004). O’Hara (2009) therefore suggested to follow mothers with feelings of downheartedness for a longer period of time. In addition, mothers may have been less susceptible to the intervention due to exhaustion. We showed their sleep level was relatively low, and exhaustion is one of the main predictors of downheartedness (Field, 2010; Kurth et al., 2011).

The lack of effect might also be due to that mothers may still be exhausted, also due to the intensity that caring for a baby requires (Kurth et al., 2011). As stated before exhaustion is one of the main predictors of feeling downhearted (Field, 2010; Kurth et al., 2011).

(18)

18 Furthermore, the crib might have affected maternal feelings of competence, which may result in feelings of downheartedness (Giallo, Rose, & Vittorino, 2011). Further research is needed to reveal whether this claim holds.

Lastly, the discrepancies between some of the idiographic effects and the multilevel effects may be due to the increased number of subjects in the multilevel analysis (Branch & Pennypacker, 2013).

Limitations and Further Directions

This was a pilot study, and our results should be interpreted with caution. Our baseline period was relatively short. It is known that more extended baseline phases reduce threats to validity (Cohen et al., 2014; Smith, 2012). However, mothers participating in our study were in need of help, and were not able to extend our baseline phase due to ethical reasons.

Second, it appeared that we attracted a relatively homogeneous sample of highly educated white females in our study, which may limit the external validity of our findings. Further research should try to include participants with varying backgrounds and social, economic status, in order to be able to generalize the results. Furthermore, we cannot make causal inferences based on the current data. A randomized control is needed to draw more reliable conclusions about causality.

We found no effect on maternal downheartedness. Future research should include a longer intervention phase and follow-up assessments in order to detect an intervention effect concerning downheartedness (O’Hara, 2009).

We used parental ratings to measure our variables of interest. Especially regarding sleep parents might not detect when their infant is awake (Henderson, France, Owens, & Blampied, 2010). In addition to the used measures we used, further studies might include

(19)

19 other objective measures such as video-recordings of infant cry and sleep behavior. Together these measures might yield a more holistic view on our outcome variables.

To conclude, in this study a smart crib at home was associated with reduced infant crying and increased infant and maternal sleep. Such a smart crib might be a responsive intervention for excessively crying infants, and exhausted mothers who feel in need of help with soothing infant crying and for help with boosting infant sleep. As such it may help break the vicious cycle between infant crying and parental exhaustion.

(20)

20 References

Baker, F., Denniston, M., Zabora, J., Polland, A., & Dudley, W. N. (2002). A POMS short from for cancer patients: Psychometric and structural evaluation. Psycho-Oncology, 11, 273–281. https://doi.org/10.1002/pon.564.

Barr, R. G. (1998). Section 3. Colic and crying syndromes in infants. Pediatrics, 102(5), 1282–1286. https://doi.org/10.1542/peds.102.5.SE1.1282.

Barr, R. G., Kramer, M. S., Boisjoly, C., McVey-White, L., & Pless, I. B. (1988). Parental diary of infant cry and fuss behavior. Archives of Disease in Childhood, 63, 380–387. doi: 10.1136/adc.63.4.380.

Bootzin, R. R., & Epstein, D. R. (2011). Understanding and Treating Insomnia. Annual

Review of Clinical Psychology, 7(1), 435–458.

https://doi.org/10.1146/annurev.clinpsy.3.022806.091516.

Borckardt, J. J., & Nash, M. R. (2014). Simulation modelling analysis for small sets of single-subject data collected over time. Neuropsychological Rehabilitation, 24(3–4), 492–506. https://doi.org/10.1080/09602011.2014.895390.

Borckardt, J. J., Nash, M. R., Murphy, M. D., Moore, M., Shaw, D., & O’Neil, P. (2008). Clinical Practice as Natural Laboratory for Psychotherapy Research: A Guide to Case-Based Time-Series Analysis. American Psychologist, 63(2), 77–95. https://doi.org/10.1037/0003-066X.63.2.77.

(21)

21 findings. APA Handbook of Behavior Analysis, Vol. 1: Methods and Principles., 1, 151– 175. https://doi.org/10.1037/13937-007.

Cohen, L. L., Feinstein, A., Masuda, A., & Vowles, K. E. (2014). Single-case research design in pediatric psychology: Considerations regarding data analysis. Journal of Pediatric Psychology, 39(2), 124–137. https://doi.org/10.1093/jpepsy/jst065.

Davis, K. F., Parker, K. P., & Montgomery, G. L. (2004). Sleep in infants and young children: Part one: normal sleep. Journal of Pediatric Health Care, 18(2), 65–71. https://doi.org/10.1016/S0891-5245(03)00149-4.

Dørheim, S. K., Bondevik, G. T., Eberhard-Gran, M., & Bjorvatn, B. (2009). Sleep and Depression in Postpartum Women: A Population-Based Study. Sleep, 32(7), 847–855. https://doi.org/10.1093/sleep/32.7.847.

Douglas, P. S., & Hill, P. S. (2013). Behavioral sleep interventions in the first six months of life do not improve outcomes for mothers or infants: A systematic review. Journal of Developmental and Behavioral Pediatrics, 34(7), 497–507. https://doi.org/10.1097/DBP.0b013e31829cafa6.

Eshel N, Daelmans B, de Mello MC, Martines J. Responsive parenting: interventions and outcomes. Bull World Health Organ . 2006;84(12):991–998.

Esposito, G., Yoshida, S., Ohnishi, R., Tsuneoka, Y., Rostagno, M. del C., Yokota, S., … Kuroda, K. O. (2013). Infant Calming Responses during Maternal Carrying in Humans and Mice. Current Biology, 23(9), 739–745. https://doi.org/10.1016/J.CUB.2013.03.041. Field, T. (2010). Postpartum depression effects on early interactions, parenting, and safety

practices: a review. Infant Behavior & Development, 33(1), 1–6. https://doi.org/10.1016/j.infbeh.2009.10.005.

(22)

22 Daytime Sleepiness in Postpartum Women. PLoS ONE, 9(7), e103513. https://doi.org/10.1371/journal.pone.0103513.

Fisher, J., Feekery, C., & Rowe, H. (2004). Treatment of maternal mood disorder and infant behaviour disturbance in an Australian private mothercraft unit: a follow-up study. Archives of Women’s Mental Health, 7(1), 89–93. https://doi.org/10.1007/s00737-003-0041-5.

Goyal, D., Gay, C., & Lee, K. (2009). Fragmented maternal sleep is more strongly correlated with depressive symptoms than infant temperament at three months postpartum. Archives of Women’s Mental Health, 12(4), 229–237. https://doi.org/10.1007/s00737-009-0070-9.

Halbreich, U., & Karkun, S. (2006). Cross-cultural and social diversity of prevalence of postpartum depression and depressive symptoms. Journal of Affective Disorders, 91(2– 3), 97–111. https://doi.org/10.1016/J.JAD.2005.12.051.

Harrington, J. W., Logan, S., Harwell, C., Gardner, J., Swingle, J., McGuire, E., & Santos, R. (2012). Effective Analgesia Using Physical Interventions for Infant Immunizations. Pediatrics, 129(5), 815–822. https://doi.org/10.1542/peds.2011-1607.

Henderson, J. M. T., France, K. G., Owens, J. L., & Blampied, N. M. (2010). Sleeping through the night: the consolidation of self-regulated sleep across the first year of life. Pediatrics, 126(5), e1081-7. https://doi.org/10.1542/peds.2010-0976.

Hiscock, H., Cook, F., Bayer, J., Pscyh, C., Le, H. N., Mensah, F., … St James-Roberts, I. (2014). Preventing Early Infant Sleep and Crying Problems and Postnatal Depression: A Randomized Trial. Pediatrics (Vol. 133). Retrieved from www.aappublications.org/news.

(23)

23 Your Newborn Baby Sleep Longer. New York: Bantam Books.

Kaley, F., Reid, V., & Flynn, E. (2011). The psychology of infant colic: A review of current research. Infant Mental Health Journal, 32(5), 526–541. https://doi.org/10.1002/imhj.20308.

Kurth, E., Kennedy, H. P., Spichiger, E., Hösli, I., & Zemp Stutz, E. (2011). Crying babies, tired mothers: What do we know? A systematic review. Midwifery, 27(2), 187–194. https://doi.org/10.1016/J.MIDW.2009.05.012.

Landgren, K., & Hallström, I. (2011). Parents’ experience of living with a baby with infantile colic - a phenomenological hermeneutic study. Scandinavian Journal of Caring Sciences, 25(2), 317–324. https://doi.org/10.1111/j.1471-6712.2010.00829.x.

Lester, B. M., & Zacharia Boukydis, C. F. (1985). Infant Crying: Theoretical and Research Perspectives - Google Boeken (1st ed.). New York: Plenum.

McKenzie, S. A. (2013).Fifteen-minute consultation: Troublesome crying in infancy. Archives of Disease in Childhood - Education and Practice;98:209-211.

Möller, E., L., Oort, F.J., & Rodenburg, R. (in press). Effects of The Happiest Baby Method: A behavioral intervention to reduce excessive infant crying. Sleep Medicine.

Möller, E., L., De Vente, W., & Rodenburg, R. (2017). [Effects of the smart crib and The Happiest Baby Method: A calming respons].

O’Hara, M. W. (2009). Postpartum depression: what we know. Journal of Clinical Psychology, 65(12), 1258–1269. https://doi.org/10.1002/jclp.20644.

Owais, S., Chow, C. H. T., Furtado, M., Frey, B. N., & Van Lieshout, R. J. (2018). Non-pharmacological interventions for improving postpartum maternal sleep: A systematic review and meta-analysis. Sleep Medicine Reviews, 41, 87–100. https://doi.org/10.1016/j.smrv.2018.01.005.

(24)

24 Paul, I. M., Savage, J. S., Anzman-Frasca, S., Marini, M. E., Mindell, J. A., & Birch, L. L. (2016). INSIGHT Responsive Parenting Intervention and Infant Sleep. Pediatrics, 138(1), e20160762–e20160762. https://doi.org/10.1542/peds.2016-0762.

Paul, I. M., Savage, J. S., Anzman, S. L., Beiler, J. S., Marini, M. E., Stokes, J. L., & Birch, L. L. (2011). Preventing obesity during infancy: A pilot study. Obesity, 19(2), 353–361. https://doi.org/10.1038/oby.2010.182.

Petzoldt, J., Wittchen, H. U., Wittich, J., Einsle, F., Höfler, M., & Martini, J. (2014). Maternal anxiety disorders predict excessive infant crying: A prospective longitudinal study. Archives of Disease in Childhood, 99(9), 800–806. https://doi.org/10.1136/archdischild-2013-305562.

Priddis, H. S., Keedle, H., & Dahlen, H. (2018). The Perfect Storm of Trauma: The experiences of women who have experienced birth trauma and subsequently accessed residential parenting services in Australia. Women and Birth, 31(1), 17–24. https://doi.org/10.1016/J.WOMBI.2017.06.007.

Savino, F. (2007). Focus on infantile colic. Acta Paediatrica, International Journal of Paediatrics, 96(9), 1259–1264. https://doi.org/10.1111/j.1651-2227.2007.00428.x

Schytt, E., Lindmark, G., & Waldenstrom, U. (2005). Physical symptoms after childbirth: prevalence and associations with self-rated health. BJOG: An International Journal of Obstetrics and Gynaecology, 112(2), 210–217. https://doi.org/10.1111/j.1471-0528.2004.00319.x.

Smith, J. D. (2012). Single-case experimental designs: A systematic review of published research and current standards. Psychological Methods, 17(4), 510–550. https://doi.org/10.1037/a0029312.

(25)

25 Snijders, T. A. B., & Bosker, R. J. (2012). Multilevel analysis. An introduction to basic and

advanced multilevel modeling (2nd edition). London: Sage Publications.

Rebecca Giallo, Natalie Rose & Renzo Vittorino (2011) Fatigue, wellbeing, and parenting in mothers of infants and toddlers with sleep problems, Journal of Reproductive and Infant Psychology, 29:3, 236-249.

Van Puyvelde, M., Loots, G., Meys, J., Neyt, X., Mairesse, O., Simcock, D., & Pattyn, N. (2015). Whose clock makes yours tick? How maternal cardiorespiratory physiology influences newborns’ heart rate variability. Biological Psychology, 108, 132–141. https://doi.org/10.1016/j.biopsycho.2015.04.001.

Whittingham, K., & Douglas, P. (2014). Optimizing parent-infant sleep from birth to 6 months: A new paradigm. Infant Mental Health Journal, 35(6), 614–623. https://doi.org/10.1002/imhj.21455.

Wolke, D., Bilgin, A., & Samara, M. (2017). Systematic Review and Meta-Analysis: Fussing and Crying Durations and Prevalence of Colic in Infants. The Journal of Pediatrics, 185, 55–61.e4. https://doi.org/10.1016/j.jpeds.2017.02.020.

Yalçın, S. S., Örün, E., Mutlu, B., Madendağ, Y., Sinici, İ., Dursun, A., … Yurdakök, K. (2010). Why are they having infant colic? A nested case-control study. Paediatric and Perinatal Epidemiology, 24(6), 584–596. https://doi.org/10.1111/j.1365-3016.2010.01150.x.

(26)

26

Table 1

Descriptive Statistics and ICC for All Outcome Variables

Outcome N M SD Range ICC

Infant crying 299 134.68 110.90 5-770 .39 Infant sleep 299 781.07 144.82 130-1200 .38 Maternal sleep 299 401.09 100.45 60-640 .21 Downheartedness 299 2.40 1.28 1-5 .54

Table 2

Results of Time-Series Phase-Effects Analyzes for Daily Crying

Baby Baseline M(N size) Intervention M (N size) Level change r p Slope change r p 1 220(3) 137.38(21) -.48 .028 -.57 . 007 2 333.33(3) 186.90(21) -.43 .075 -.70 .001 3 336.25(4) 105.48(21) -.59 .006 -.51 .023 4 73,33 (3) 45(23) -.32 .255 .28 .505 5 236.67(3) 47.5 (12) -.71 .011 -.32 .382 6 86.25(4) 72.14(21) -.20 .348 -.46 .024 7 138.33(3) 105.91(22) -.19 .357 -.15 .473 8 377.5(4) 243.23(31) -.36 .041 -.33 .057 9 221(5) 145.48(21) -.27 .270 .14 .595 10 85.33(6) 86.75(20) .01 .972 -.08 .741 11 267.5(8) 135(18) -.60 .009 -.21 .450 12 56.67(6) 45.63(16) -.24 .443 -.37 .233

(27)

27

Table 3

Results of Time-Series Phase-Effects Analyzes .for Daily Infant Sleep

Baby Baseline M(N size) Intervention M (N size) Level change r p Slope change r p 1 743.33(3) 819.52(21) .25 .296 .00 1 2 755(3) 765.48(21) .03 .887 .02 .925 3 507.5(4) 844.52(21) .68 .005 .60 .044 4 632.67 (3) 723.91(23) .29 .230 .37 .151 5 505(3) 727.5 (12) .61 .064 .621 .096 6 741.25 (4) 697.62(21) -.21 .310 -.01 .963 7 776.67(3) 862.27(22) .39 .022 -.07 .685 8 841.25(4) 906.45(31) .24 .141 -.31 .055 9 619(5) 607.63(21) -.03 .871 .18 .307 10 869.17(6) 866(20) -.02 .909 .29 .101 11 957.5 (8) 837.22(18) -.41 .045 .39 .058 12 740(6) 718.25(16) -.17 .455 .28 .226 Note. Sleep is based on minutes within 24 hours..

Table 4

Results of Time-Series Phase-Effects Analyzes for Daily Maternal Sleep

Baby Baseline M (N size) Intervention M (N size) Level change r p Slope change r p 1 350(3) 381.90(21) .13 .649 .24 .507 2 365(3) 374.29(21) .04 .872 .16 .573 3 291.25 (4) 405.95(21) .48 .020 .45 .031 4 360 (3) 396.74(23) .15 .483 .25 .263 5 353.33(3) 407.5 (12) .21 .607 .69 .091 6 323.75 329.29(21) .02 .932 .23 .341

(28)

28 (4) 7 470 (3) 457.05(22) -.05 .772 .26 .095 8 325(4) 364.84(31) .12 .493 .37 .024 9 328(5) 382.38(21) .21 .468 .35 .218 10 480.83(6) 514.25(20) .18 .347 .12 .522 11 457.5 (8) 498.33(18) .25 .308 -.21 .403 12 383.33 (6) 375(16) -.14 .524 .21 .344

Note. Sleep is based on minutes within 24 hours.

Table 5

Results of Time-Series Phase-Effects Analyzes for Daily Feelings of Downheartedness

Baby Baseline M (N size) Intervention M (N size) Level change r p Slope change r p 1 1 (3) 1.19 (21) .17 .461 -.24 .314 2 1 (3) 1.28 (21) .22 .405 .06 .842 3 1.75 (4) 2.33(21) .22 .384 .12 .661 6 3.66 (3) 4.39 (23) .21 .449 .49 .163 7 4(3) 4.66 (12) .26 .372 .35 .219 8 1.75 (4) 2.28 (21) .22 .155 -.07 .622 9 3 (3) 2.48(22) -.16 .384 .07 .694 10 3.33(4) 1.90 (31) -.55 .015 -.58 .014 12 1.8(5) 2.14 (21) .14 .611 .41 .146 13 2(6) 2.05 (20) .02 .937 .09 .736 14 2.25 (8) 2.33 (18) .05 .832 .32 .177 15 3.33 (6) 2.75 (16) -.35 .096 -.07 .720

(29)

(30)

30 Table 6

Main Effects of the Smart Crib on Infant Crying, Infant Sleep, Maternal Sleep, and Downheartedness

Baseline Intervention

Outcome B(SE) T(df) P B(SE) T(df) p

Infant crying Intercept 175.42(25.18) 6.966(35.03) .000 -40. 77(16.50) -3.654(289,55) .000 Slope -12.92(6.79) -1.900(289.202) .058 -1.93 Change in Slope 10.99(6.83) 1.607(289.13) .109 Infant sleep Intercept 683.33(32.78) 20.84(38.61) .000 80.32(27.03) 2.97(287.77) .003 Slope -30.18(9.16) -3.30(289.44) .001 1.65 Change in Slope 31.83(9.21) 3.46(289.39) .001 Maternal sleep Intercept 380.19(22.76) 16.70(49.34) .000 -8.99(20.13) -.447(288.01) .655 Slope 3.31(6.81) .486(290.15) .628 2.98 Change in Slope -0.07(6.85) -.003(290.05) .998 Downheartedness

(31)

31

Intercept 2.44(.33) 7.31(23.16) .000 .04(.22) .18(287.18) .858

Slope .01(.07) .08(288.06) .937 0

(32)

32 Appendix A: Overview of sampling and design

Figure A1. Flowchart sampling.

Community

infants

Assessed for eligibility (n = 19

)

Excluded (n = 3); Crying already diminished after visiting pediatrician Allocated to intervention (n = 16) Completed intervention (n = 12)

Currently receiving the

(33)

33 Figure A2. Design overview.

(34)

34 Appendix B:Daily Diaries

Table B1 Baseline Diary

Item Response

How many hours/minutes did your baby cry during the

morning/afternoon/evening/night? Hours and Minutes

How many hours/minutes did your baby sleep during the

How many hours/minutes was your baby awake and fussy during

the morning/afternoon/evening/night? Hours and Minutes

How many feedings did your baby have during the

morning/afternoon/evening/night? Number of Feedings

How often has your baby woken up tonight? Number of Awakenings

How many times have you woken up tonight? Number of Awakenings

How many hours/minutes have you slept during the

To what extent did you feel downhearted or exhausted today? 5-point Likert scale

(35)

35 Table B2

Intervention Diary

Item Response

How many hours/minutes did your baby cry during the

How many hours/minutes was your baby awake and fussy during

the morning/afternoon/evening/night? Hours and Minutes

How many feedings did your baby have during the

morning/afternoon/evening/night? Number of Feedings

How often has your baby woken up tonight? Number of Awakenings

How many times have you woken up tonight? Number of awakenings

How many hours/minutes have you slept during the

To what extent did you feel downhearted or exhausted today? 5-point Likert scale

Did you use the crib in the last 24 hours? Yes/No

Where did your baby sleep during the night? 6 categories How many hours/minutes did your baby sleep in the crib during the

(36)

36 Appendix C: Graphs of intervention effects per mother-infant dyad

Figure C1. Graph mother-infant dyad 1. Note. A is the baseline phase, B is the intervention phase. .

Figure C2. Graph mother-infant dyad 2. Note. A is the baseline phase, B is the intervention phase. 0 1 2 3 4 5 0 200 400 600 800 1000 1200 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Infant Crying Infant Sleep Maternal Sleep Downheartedness

0 1 2 3 4 5 0 200 400 600 800 1000 1200 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

B

A

B

B

A

(42)

42 Appendix D: Graphs of Multilevel Effects per Outcome

Figure D1. Graph of infant crying. Note. A is the baseline phase, B is the intervention phase. Fewer participants after 21 measures.

(43)

43 Figure D2. Graph of infant sleep. Note. A is the baseline phase, B is the intervention phase. Fewer participants after 21 measures.

(44)

44 Figure D3. Graph of maternal sleep. Note. A is the baseline phase, B is the intervention phase. Fewer participants after 21 measures.

(45)

45 Figure D4. Graph of downheartedness. Note. A is the baseline phase, B is the intervention phase. Fewer participants after 21 measures.