Secondary Logo

Journal Logo

METHODS

Comparisons of Three Measures of Maternal Engagement Activities in the Neonatal Intensive Care Unit

Greenfield, Jennifer C.; Weikel, Blair W.; Bourque, Stephanie L.; Hwang, Sunah S.; Klawetter, Susanne; Roybal, Kristi L.; Palau, Mauricio A.; Scott, Jessica; Shah, Pari; Brown, Kyria; Neu, Madalynn

Author Information
doi: 10.1097/NNR.0000000000000582
  • Free

Abstract

As efforts increase to enhance bonding between mothers and their infants who are hospitalized in neonatal intensive care units (NICUs), emphasis is being placed on promoting maternal engagement activities such as increased visitation or presence in the NICU, kangaroo care (KC; infant on the mother’s chest in skin-to-skin contact), and general caregiving (e.g., bathing and diaper change). To determine effectiveness of interventions, accurate measurement of maternal engagement activities is vital. Of course, videotaping or monitoring these activities provides the most accurate measurements. Still, they can be intrusive and unfeasible for daily measurement in longitudinal studies, especially those that include home measurement. Methods often used for data collection in longitudinal studies include surveys, maternal time use diaries, and electronic health record (EHR) documentation. As we describe here, each has benefits and challenges, and no one method has been identified as a gold standard for measuring maternal activities.

Surveys

“Surveys are a collection of information from a sample of individuals through their responses to questions” (Check & Schutt, 2012, p. 160). The infant literature is most commonly used to measure types and amounts of maternal feeding. In one study, mothers completed a survey at 2, 6, and 12 months after birth to assess if they were currently breastfeeding, if and when this changed, and exclusively breastfeeding (Tully et al., 2016). The authors proposed that a limitation to the study was an unclear definition of breastfeeding that may have resulted in underreporting. In another study, mothers of 1- to 12-week-old infants completed an electronic survey. Questions included the hour after birth of the first breastfeeding attempt, if mothers continued breastfeeding during hospitalization, and if formula supplementation was used. The authors mentioned social desirability response as a limitation to self-report measures when participants overreport an activity to please the investigators (Matthay & Glymour, 2020; Wood et al., 2021). Little et al. (2021) examined use of an infant carrier on breastfeeding outcomes with electronic surveys at 2 and 6 weeks and 3 and 6 months after birth. Mothers were asked what they were feeding the infant, how many times in the last 24 hours they breastfed, and how many hours per day they used the carrier. Only data loss through attrition was mentioned as a limitation to the survey data. An online Likert scale was used in another study to indicate the frequency of father–infant engagement activities. The researchers added a qualitative component to assess fathers’ involvement (Atkinson et al., 2021). Survey recalls in the studies mentioned above ranged from 24 hours to several weeks. Validation studies on surveys suggest accurate responses over a 2-week or greater period (Cohen, 1988; Radloff, 1977).

Maternal Time Use Diaries

A time use diary collects the continuous time an activity is performed with estimated start/finish times. It is used to evaluate both durations and sequences of a specific activity (Gershuny et al., 2020). Often, measurement of maternal holding of her infant is calculated by a maternal diary or checklist without corroboration from a second instrument. Neu et al. (2013) used a detailed time use diary for 8 weeks to measure times, duration, and type of holding by both parents and did not report limitations with data collection. Mothers were reimbursed for diary completion. Diaries were collected every week by a research team member during a home visit and discussed with the mother. Raiskila et al. (2017) also used a time use diary to measure presence in the NICU of each parent and time, duration, and type of holding for 2–3 weeks. The researchers speculated that if a problem with data existed, it was slight underreporting. Tully et al. (2016) reported that mothers completed a checklist of engagement activities such as singing, rocking, and KC at each visit while their infants were hospitalized in the NICU. The checklists were collected weekly. Missing data occurred because mothers sometimes failed to complete the checklists or they were misplaced in the NICU.

Maternal diaries often are used in sleep research to measure amount and quality of children’s sleep. They have been compared to more objective measures such as actigraphy that assesses sleep by recording periods of activity and inactivity based on movement. An actigraph is a small device that can be worn on the wrist or ankle (Dayyat et al., 2011). In one study, the diaries showed an overall average of 2.5 hours greater nighttime sleep and 2.3 hours of 24-hour sleep than actigraphy (Covington et al., 2019). In other studies, parents reported higher total sleep time and fewer night awakenings of their children compared to actigraphy (Dayyat et al., 2011; Iwasaki et al., 2010). Results were reported individually for both instruments.

Electronic Health Record

EHRs—digital accounts of patient care—have been compared with other measures for accuracy. Avidan et al. (2014) compared real-time data by a trained observer on drug administration during elective surgeries with EHR data entered by anesthesiologists. For drug names, drug doses, and drug administration time agreements were 83.1%, 92.5%, and 96.2% respectively. Most disagreements were omissions in EHR charting (Avidan et al., 2014). In a study of nurse charting in an intensive care unit, an average of 1.4 charting deficiencies (omissions) per day per patient for 60 patients at the end of a 12-hour day shift was found (Liaw & Goh, 2018).

Use of Maternal Diaries and EHR in the NICU

Diaries or checklists and EHR measuring maternal engagement activities in the NICU have been previously compared. Pineda et al. (2012) examined parent visitation and holding between a nurse time use diary and the EHR in a study spanning 5 weeks. When a discrepancy arose, the method documenting higher engagement was analyzed. In a study measuring amount of KC and visitation, mothers recorded when they visited their infant and how long they engaged in KC. In addition, data on dates and times of maternal visitation and hours engaging in KC were extracted from the medical record. No information was provided on which data were used or whether data from the two measures were reconciled if divergent (Gonya & Nelin, 2013). Blomqvist et al. (2011) compared frequency and duration of KC holding between a very detailed parent diary and the patient medical record and reported that the parent diary was more complete. However, the diary was used for only 24 hours.

Summary

When limitations were mentioned for the survey, time use diary, or EHR data, omissions were common to all three measures (Avidan et al., 2014; Liaw & Goh, 2018; Raiskila et al., 2017; Tully et al., 2016). Surveys and diaries share social desirability responses—a limitation to self-report measures (Wood et al., 2021). Misinterpretation of questions or unclear questions are other limitations (Tully et al., 2016). Compared to objective measures, mothers have been found to overreport in a diary (Covington et al., 2019; Dayyat et al., 2011; Iwasaki et al., 2010). When diaries and EHR data are collected, there are no guidelines to reconcile discrepancies (Gonya & Nelin, 2013; Pineda et al., 2012). Powers et al. (2022) reviewed 27 studies dating from 1972 through 2018 and found no systematic method of recording parental presence in the NICU or engagement activities. Although no gold standard exists for measuring maternal engagement activities, using two data collection instruments is recommended to improve accuracy (Kirkpatrick et al., 2018). Lack of accuracy in data collection threatens study validity (Matthay & Glymour, 2020). Data on maternal engagement are essential for planning and evaluating interventions to increase engagement between mothers and infants during NICU hospitalization and the months following hospitalization. Determining differences in values captured from instruments used in this study is a first step in deciding how to reconcile discrepancies.

This study was part of a larger study examining barriers and facilitators of maternal engagement when preterm infants are hospitalized in the NICU. The aim of this component of the study was to compare data from a survey instrument; a time use maternal diary; and EHR on maternal time spent visiting, holding, and general caregiving. The secondary aim was to identify patterns in detected differences in measures of engagement.

METHODS

Design and Sample

This methodological comparative study included mothers admitted to three NICUs in Colorado between June 2017 and December 2019. Two NICUs were Level III with a capacity for 25–50 patient beds; one served as the primary academic delivery hospital for the state, whereas the other performed in a city approximately 60 miles north of the central metro area. The third NICU was a Level IV unit with 84 beds, situated on the academic medical campus in a suburban area, and served as the regional referral unit for a multistate catchment area. Infant inclusion criteria were birth at ≤32 weeks of gestational age, no congenital birth defects, and at least 2 weeks of NICU exposure. Maternal inclusion criteria were English or Spanish as the primary language, no diagnosed psychiatric disorders, and no recorded or stated illicit substance use. The study was approved by the Colorado Multiple Institutional Review Board that was responsible for research oversight at all three NICUs. Mothers were approached for consent by a bilingual research team member. The consent was explained, and mothers were given opportunities to ask questions before agreeing to participate.

Procedure

We enrolled mother–infant dyads when infants were 33–34 weeks postmenstrual age and had been admitted to the NICU for at least 2 weeks. At enrollment, mothers completed the consent form and a demographic questionnaire that included race/ethnicity, marital status, highest completed level of education, insurance provider, mode, time of travel to the NICU, and presence of other children in their households. Mothers also answered survey questions about their visits, including estimates of the number of days per week and the average time each day they spent in the NICU during the weeks before study enrollment. Because of differences in length of stay prior to survey completion, some mothers recalled time spent in the NICU over only the prior 2 weeks, whereas some recalled their engagement over longer periods of time. These survey questions are outlined in Figure 1.

F1
FIGURE 1:
Maternal survey questions and time use diary examples.

Maternal Time Use Diaries

After enrollment and completion of the initial survey, mothers were instructed to complete a daily time use diary noting when they were present at the bedside and type and duration of activities performed with their infant: KC, traditional blanket holding, and infant cares (e.g., changing diapers, bathing, and taking the infant’s temperature). KC was defined as mother holding her diaper-clad infant in skin-to-skin contact between her breasts, with her clothing wrapped around the infant. Blanket or traditional holding was specified as the mother holding the infant horizontally or vertically while the infant was dressed and wrapped in a blanket. Figure 1 shows two types of time use diaries. One consisted of two circles, one for a.m. and one for p.m., with each division representing 30 minutes. Mothers shaded in the time and added the letter representing the activity into the appropriate division. The other included a table for mothers to record the time and activity.

The survey was completed on paper or electronically, and the diaries were completed on paper. A research assistant then entered the diary data into a Health Insurance Portability and Accountability Act-compliant data platform (Research Electronic Data Capture; Harris et al., 2009). The survey and diaries were available in English and Spanish. A bilingual research team member instructed mothers to complete the surveys and diaries and then collected diary sheets weekly. She also answered any questions from the mothers about diary documentation.

Electronic Health Record

EHR input of maternal caregiving activities by NICU nurses was completed differently in each of the three units. In the two Level III NICUs, maternal caregiving activities included maternal feeding, holding, presence in the NICU, participation in care, participation in developmental therapy, and others. Nurses checked a caregiving activity when they noticed it or when the nurse was at the bedside, which produced binary data for each category rather than a measure of time engaged in each activity. In the Level IV NICU, nurses had a checklist of caregiving activities that included mother: active in care, at bedside, holding, KC, not engaged in plan of care, sleeping (mother at bedside but sleeping), and not present. Documentation was completed hourly, and often multiple categories were checked. At times, these categories seemed to conflict—for instance, if “not present” and “at bedside” were both checked. After consulting with nurses in the unit to better understand their customs around documenting mothers’ time, we interpreted and applied uniform coding rules to reflect the nurses’ intent in the unit. As an example, the conflicting categories of “not present” and “at bedside” often were used for 1 hour to indicate the mother was at the hospital but not at the bedside in that hour; similarly, checking “holding” and “not present” in the same hour indicated that the mother held the infant but not for the whole hour. Feeding was not included in the activity checklist and was charted separately.

The information on maternal activities was extracted from the medical records at each unit using the Health Data Compass program (https://www.healthdatacompass.org). This program serves as a health data warehouse and provides EHR data for the participating hospitals.

Data Analysis

Level III NICUs

Because of the way that unit nurses documented maternal activities in the Level III NICUs, we analyzed engagement at these facilities as the number of instances of holding, cares, and presence at the bedside not engaged in holding or cares, rather than the length of time engaged in each activity. Our research team counted each type of engagement instance whenever it was charted to analyze EHR data. For example, if at 12:00 p.m. and 4:00 p.m. it was charted that a mother was holding her infant, we counted two instances of holding for that day. Our research team reviewed time use diaries kept by mothers and similarly counted the number of times she reported engagement in each activity during each day. Unfortunately, because of the nature of data collection, we could not compare the diary and EHR count data with survey data from participants in the Level III NICUs. Therefore, the median number of instances engaged in each activity per week of NICU hospitalization were compared between EHR documentation and maternal time use diaries using Wilcoxon signed-ranks tests. EHR data were reconciled to only compare weeks in which mothers completed diaries, as the earliest they would have begun was 2 weeks after NICU admission. Some mothers stopped completing diaries near the end of their NICU stay.

Level IV NICU

In the Level IV NICU, nurses recorded maternal engagement every hour in the EHR, comparing the number of hours engaged in infant care, holding, KC, and time present at the bedside. This documentation captured family interaction by person, allowing for specific extraction of maternal engagement data. If only one of these areas was selected for 1 hour of charting, we considered this full 1 hour of engagement for that activity. If more than one activity was charted, we divided the total time between those activities. For example, if a mother was observed holding and performing care for 1 hour, we assigned 0.5 hours to each activity. We used the hours engaged in each activity noted in the maternal diaries for this analysis.

We conducted pairwise comparisons of the median number of hours per week engaged in each activity over NICU hospitalization across the three tools, comparing EHR to diaries and EHR to survey data. Surveys could not be compared to time use diaries as they reflected engagement over different time periods. We tested for significant differences with Wilcoxon signed-ranks test. All statistical analyses were performed with SAS software, Version 9.4 (Cary, NC).

RESULTS

We analyzed data from 101 mother–infant dyads admitted to the two Level III NICUs and 45 dyads admitted to the Level IV NICU. Table 1 shows demographics for the total sample by NICU level. In our sample, 60% of mothers (n = 88) identified as White. Most mothers (82%, n = 119) were less than 35 years of age, and 86 mothers (59%) had private health insurance coverage. Sixty-six percent (n = 96) of mothers had been employed full time prior to NICU admission; 42% (n = 61) had annual household income of less than $50,000. Mothers were excluded from analysis if they withdrew from the study and were excluded from diary analysis if they did not complete any diaries while participating in the study. Eight dyads withdrew from the Level IV NICU during the study period, and two withdrew from the Level III NICUs.

TABLE 1 - Demographic Characteristics of Cohort
Total Level IV NICU Level III NICUs
N = 146 n = 45 n = 101
Maternal characteristic n (%) n (%) n (%)
Race
 Non-Hispanic White 88 (60) 30 (67) 58 (57)
 Hispanic 35 (24) 11 (24) 24 (24)
 Non-Hispanic Black 16 (11) 2 (4) 14 (14)
 Other 6 (4) 1 (2) 5 (5)
 Missing 1 (1) 1 (2) 0
Age (years)
 <25 23 (16) 9 (20) 14 (14)
 25–35 96 (66) 28 (62) 68 (67)
 >35 27 (18) 8 (18) 19 (19)
Insurance
 Private 86 (59) 26 (58) 60 (59)
 Medicaid 56 (38) 18 (40) 38 (38)
 None 4 (3) 1 (2) 3 (3)
Education
 High school or less 38 (26) 11 (24) 27 (27)
 Some college/associate’s 37 (25) 10 (22) 27 (27)
 Bachelor’s 47 (32) 16 (36) 31 (31)
 Master’s or higher 23 (16) 7 (16) 16 (16)
 Missing 1 (1) 1 (2) 0
Marital status
 Married 119 (82) 41 (91) 78 (77)
 Other status 27 (18) 4 (9) 23 (23)
Household income
 <$50,000 61 (42) 18 (40) 43 (43)
 $50,000–$99,999 46 (32) 10 (22) 36 (36)
 >$100,000 35 (24) 15 (33) 20 (20)
 Missing 4 (3) 2 (4) 2 (2)
Pre-NICU employment
 Working, full time 96 (66) 27 (60) 69 (69)
 Other 49 (34) 18 (40) 31 (31)
 Missing 1 (1) 0 1 (1)
Note. NICU = neonatal intensive care unit.

In comparing counts of engagement between EHR documentation and maternal time use diaries in the Level III NICUs, we found significant differences for all types of engagement between the two sources at p < .01 with more instances of each engagement type recorded in the EHR than in the maternal diaries (Table 2).

TABLE 2 - Level III Neonatal Intensive Care Unit Electronic Health Record Versus Maternal Diaries: Instances of Engagement Per Week
EHR
Median [IQR]
Diaries
Median [IQR]
p a
Holding 13.0 [7.1, 23.7] 5.5 [3.2, 7.0] <.01
Cares 15.5 [8.6, 30.7] 5.4 [2.9, 6.5] <.01
Present 3.5 [2.2, 5.7] 1.4 [0.2, 4.3] <.01
Note. EHR = electronic health record; IQR = interquartile range.
aWilcoxon signed-ranks test.

In the pairwise comparison of median weekly hours of engagement between EHR documentation and cross-sectional survey in the Level IV NICU, we found a significant difference between the tools for KC (p < .01) and blanket holding (p < .01). More holding hours were reported in the EHR, whereas more hours of KC were reported on the survey. Hours engaged in caregiving and present in the NICU did not differ significantly (Table 3).

TABLE 3 - Level IV Neonatal Intensive Care Unit Electronic Health Record Versus. Surveys: Hours of Engagement Per Week
EHR
Median [IQR]
Survey
Median [IQR]
p a
Holding 1.1 [0.7, 2.8] 0 [0, 4] <.01
Cares 12.4 [6.9, 19.8] 10 [7,14] .25
Present 19.8 [11.7, 36.5] 19 [5.3, 42] .31
Kangaroo care 2.2 [1.0, 3.7] 13.8 [7, 20] <.01
Note. EHR = electronic health record; IQR = interquartile range.
aWilcoxon signed-ranks test.

The pairwise comparison between EHR documentation and maternal time use diaries in the Level IV NICU showed significant differences in the median hours captured for holding (p = .05), cares (p < .01), and presence (p < .01), but not KC (Table 4). Each significant difference showed more hours reported in the EHR than in the diaries.

TABLE 4 - Level IV Neonatal Intensive Care Unit Electronic Health Record Versus Diaries: Hours of Engagement Per Week
EHR
Median [IQR]
Diaries
Median [IQR]
p a
Holding 4.2 [1.9, 8.5] 2.6 [0.2, 7.3] .05
Cares 22.7 [11.4, 33.6] 3.0 [1.2, 7.1] <.01
Present 27.1 [13.4, 44.3] 11.6 [1.7, 31.1] <.01
Kangaroo care 1.5 [0.7, 3.1] 2.4 [0.4, 5.5] .48
Note. EHR = electronic health record; IQR = interquartile range.
aWilcoxon signed-ranks test.

DISCUSSION

The objectives of this study were to compare three measurement methods of maternal engagement (a five-item maternal cross-sectional survey, time use diaries, and EHRs) to identify (a) whether these methods capture consistent data and (b) patterns in detected differences in measures of engagement. Consistent with other literature, data captured from different measurement tools were discordant (Blomqvist et al., 2011; Covington et al., 2019; Dayyat et al., 2011; Pineda et al., 2012). However, there was some consistency between data captured in surveys and in the EHR in the Level IV NICU. Unfortunately, there was no comparison where differences were nonsignificant across all three tools. We also did not detect consistent differences among the three measurement tools. In the Level III NICUs, significantly more instances of engagement were recorded in the EHR across all categories of engagement compared to the maternal diaries. In the Level IV data, the results were less consistent. When mothers’ survey data were analyzed, only hours spent doing KC and blanket holding were considerably different between surveys and EHR. Mothers reported many more hours of KC than were recorded in the EHR. Meanwhile, only time spent doing KC did not differ substantially between the EHR and maternal diary data; considerably more hours of cares were recorded in the EHR than in maternal diaries.

This analysis adds to the literature exploring best practices for measuring maternal engagement in NICUs. We found that collecting data about mothers’ physical presence and engagement via surveys and diaries is feasible and, in some situations, yields maternal report data that are comparable to that collected by nurses recording engagement data in medical records. Interestingly, asking mothers to estimate time engaged in specific activities during a survey administered at study enrollment produced data consistent with EHR data captured during the same time for the following measures: time spent caregiving and visiting (presence). This finding is similar to reported psychometric research with longer surveys in which respondents can accurately recall frequencies of emotions and activity for up to 1 month (Cohen, 1988; Craig et al., 2003; Radloff, 1977).

Remarkably, mothers’ recording of physical presence and provision of cares in the diaries was so much lower in every engagement activity in the Level III NICUs. This finding differs from other literature in which mothers overestimate time (Covington et al., 2019; Dayyat et al., 2011; Iwasaki et al., 2010). Visitation and caregiving also seemed underreported in the Level IV NICU by mothers, suggesting instrument decay (Flannelly et al., 2018; Gibson et al., 2017). Mothers may have felt too overwhelmed or fatigued to record everything in the diaries. It also is possible that they did not count visitation time if they left the bedside. For instance, if the mother left the bedside to eat in the cafeteria or go for a walk, she might not record the visit time. On the other hand, the bedside nurse could record the time that the mother entered the NICU until she left for the day as visitation. Providing care to the infant may have been interpreted differently by the nurse and the mother.

Conversely, KC may have been overreported per survey in the Level IV NICU. KC is the standard of holding in this NICU. If mothers held using the traditional blanket method, they might not have wished to report that. The difference was largest between the survey and EHR. Then again, the Level IV nursery had private rooms, and it is possible that the nurses were not aware when the mother was engaged in KC. Mothers recorded almost 14 hours per week of KC on the survey. Mothers may have been able to focus more on KC than the other activities, in part because it requires a shorter time period and is just one activity. It also may be that social desirability bias with mothers trying to please the research team or other considerations led to differences reporting estimates of time spent in this activity (Matthay & Glymour, 2020). Interestingly, mothers recorded less KC on the diaries than written in the survey, and there was no difference in time reported for KC between the EHR and diaries. There may have been omissions in both the diaries and the EHR.

Our findings also support concerns about the accuracy of the various strategies for recording maternal engagement. First, there are essential inconsistencies in how maternal activities are recorded across units. In the example of the Level IV NICU, recording engagement as a dichotomous checked/unchecked variable per hour—with conflicting categories checked in the same hour at times—limits the types of analyses that can be conducted and requires interpretation of the raw data, which may impose error. Furthermore, differences in recording strategies and norms across NICUs present challenges for researchers seeking to compare data and findings across studies of different units. For instance, because of differences in how data are recorded in EHRs at other units, it was not possible to compare EHR data across all units in this study in the same analysis. We attempted to circumvent this problem by collecting survey and maternal diary data to triangulate engagement recorded in the EHR; however, comparisons with EHR data make it unclear which is a more accurate measurement of actual time spent by mothers in these critically important activities.

Limitations

We experienced challenges with each of the data collection methods. In particular, although study participants were reasonably consistent in completing diaries in the first weeks after study enrollment, diary completion seemed to become more inconsistent as their stay in the NICU continued and missingness increased with weeks of participation, suggesting instrument decay (Flannelly et al., 2018; Matthay & Glymour, 2020). The use of text messages or app-based reminders may help maintain consistent diary completion over time. Also, reviewing diary entries weekly and reviewing the data with mothers soon after completing the diary sheets might help produce more consistent data. Though, the staff time required for this strategy may be cost-prohibitive and impractical.

Another challenge is that staff in each unit use creative strategies for charting complicated situations, such as recording two activities in the same time block. This requires analysts to employ judgment in interpreting these data and inconsistencies in how the staff do this charting could lead to systematic errors in interpretation. Slight discrepancies in training may result in systematic differences in recording these data. Also, charting maternal engagement may not always be a priority when new staff are trained to use the EHR. Providing training to all staff at the initiation of study data collection and refreshing the training at periodic intervals might help reduce data inconsistencies.

Finally, although survey data seemed the most consistent in matching EHR data, surveys could not be compared to EHR documentation and diaries because the data were collected in different formats (i.e., counts vs. hours of engagement) in the Level III NICUs from which most of the mothers were enrolled. Also, survey data could only be compared to EHR documentation in the Level IV nursery because the survey was administered only at enrollment before mothers initiated diary entry. Surveys administered at only one point during the NICU stay likely do not capture changes in NICU engagement over time. In addition, differences in the length of stay before completion of the initial survey may have resulted in differences in the reliability of mothers’ recall. Asking participants to complete surveys at a few time points during their NICU stay may yield more accurate data without the missingness present in daily diary collection.

Despite these challenges, maternal engagement remains an important focus of study, mainly because parents’ engagement is linked with several important outcomes for preterm infants and their parents (Klawetter et al., 2019). Study validity depends on accurate data collection. Researchers must continue to explore strategies for compiling reliable and valid engagement data in the NICU. Often only one measure of data collection is used (Mitha et al., 2019; Raiskila et al., 2017; Samra et al., 2015; Tully et al., 2016). Two or more measures of engagement data collection have the potential to provide more accurate data or at least reveal if there are significant discrepancies (Kimberlin & Winterstein, 2008) when more than one measure is used—reporting which data were chosen or how reconciliation of the data is vital for interpreting results.

Conclusion

Successful strategies for measuring maternal engagement in the NICU may require additional staff time for training and survey administration, additional data collection time points, and perhaps application development or exploration of other tech-based strategies to help minimize missingness. These strategies may be essential investments to improve scientific evidence about the role of maternal NICU engagement in supporting infant and maternal health and mental health outcomes.

FU1

ORCID iDs

Jennifer C. Greenfield https://orcid.org/0000-0002-5498-6333

Blair W. Weikel https://orcid.org/0000-0001-6748-2647

Stephanie L. Bourque https://orcid.org/0000-0003-4938-7296

Sunah S. Hwang https://orcid.org/0000-0002-2661-4049

Susanne Klawetter https://orcid.org/0000-0001-7931-531X

Kristi L. Roybal https://orcid.org/0000-0002-1274-5240

Mauricio A. Palau https://orcid.org/0000-0002-7068-487X

Pari Shah https://orcid.org/0000-0002-8229-967X

Kyria Brown https://orcid.org/0000-0001-6749-8331

Madalynn Neu https://orcid.org/0000-0002-2662-2495

REFERENCES

Atkinson L., Silverio S. A., Bick D., Fallon V. (2021). Relationships between paternal attitudes, paternal involvement, and infant-feeding outcomes: Mixed-methods findings from a global online survey of English-speaking fathers. Maternal & Child Nutrition, 17(Suppl. 1), e13147. 10.1111/mcn.13147
Avidan A., Dotan K., Weissman C., Cohen M. J., Levin P. D. (2014). Accuracy of manual entry of drug administration data into an anesthesia information management system. Canadian Journal of Anesthesia, 61, 979–985. 10.1007/s12630-014-0210-1
Blomqvist Y. T., Rubertsson C., Nyqvist K. H. (2011). Parent-infant skin-to-skin contact; how do parent records compare to nurse records?Acta Paediatrica, 100, 773–775. 10.1111/j.1651-2227.2011.02160.x
Check J., Schutt R. K. (2012). Research methods in education. SAGE. 10.4135/9781544307725
Cohen S. (1988). Perceived stress in a probability sample of the United States. In Spacapan S., Oskamp S. (Eds.), The social psychology of health (pp. 31–67). SAGE.
Covington L. B., Rogers V. E., Black M. M. (2019). Sleep measurement in toddlers from low-income families. Journal of Developmental & Behavioral Pediatrics, 40, 219–223. 10.1097/DBP.0000000000000651
Craig C. L., Marshall A. L., Sjöström M., Bauman A. E., Booth M. L., Ainsworth B. E., Pratt M., Ekelund U., Yngve A., Sallis J. F., Oja P. (2003). International physical activity questionnaire: 12-Country reliability and validity. Medicine & Science in Sports & Exercise, 35, 1381–1395. 10.1249/01.MSS.0000078924.61453.FB
Dayyat E. A., Spruyt K., Molfese D. L., Gozal D. (2011). Sleep estimates in children: Parental versus actigraphic assessments. Nature and Science of Sleep, 3, 115–123. 10.2147/NSS.S25676
Flannelly K. J., Flannelly L. T., Jankowski K. R. B. (2018). Threats to the internal validity of experimental and quasi-experimental research in healthcare. Journal of Health Care Chaplaincy, 24, 107–130. 10.1080/08854726.2017.1421019
Gershuny J., Harms T., Doherty A., Thomas E., Milton K., Kelly P., Foster C. (2020). Testing self-report time use diaries against objective instruments in real time. Sociological Methodology, 50, 318–349. 10.1177/0081175019884591
Gibson R. S., Charrondiere U. R., Bell W. (2017). Measurement errors in dietary assessment using self-reported 24-hour recalls in low-income countries and strategies for their prevention. Advances in Nutrition, 8, 980–991. 10.3945/an.117.016980
Gonya J., Nelin L. D. (2013). Factors associated with maternal visitation and participation in skin-to-skin care in an all referral Level IIIc NICU. Acta Paediatrica, 102, e53–e56. 10.1111/apa.12064
Harris P. A., Taylor R., Thielke R., Payne J., Gonzalez N., Conde J. G. (2009). Research Electronic Data Capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics, 42, 377–381. 10.1016/j.jbi.2008.08.010
Iwasaki M., Iwata S., Iemura A., Yamashita N., Tomino Y., Anme T., Yamagata Z., Iwata O., Matsuishi T. (2010). Utility of subjective sleep assessment tools for healthy preschool children: A comparative study between sleep logs, questionnaires, and actigraphy. Journal of Epidemiology, 20, 143–149. 10.2188/jea.je20090054
Kimberlin C. L., Winterstein A. G. (2008). Validity and reliability of measurement instruments used in research. American Journal of Health-System Pharmacy, 65, 2276–2284. 10.2146/ajhp070364
Kirkpatrick S. I., Collins C. E., Keogh R. H., Krebs-Smith S. M., Neuhouser M. L., Wallace A. (2018). Assessing dietary outcomes in intervention studies: Pitfalls, strategies, and research needs. Nutrients, 10, 1001. 10.3390/nu10081001
Klawetter S., Neu M., Roybal K. L., Greenfield J. C., Scott J., Hwang S. (2019). Mothering in the NICU: A qualitative exploration of maternal engagement. Social Work in Health Care, 58, 746–763. 10.1080/00981389.2019.1629152
Liaw Y. Q., Goh M. L. (2018). Improving the accuracy of fluid intake charting through patient involvement in an adult surgical ward: A best practice implementation project. JBI Database of Systematic Reviews and Implementation Reports, 16, 1709–1719. 10.11124/JBISRIR-2017-003683
Little E. E., Cioffi C. C., Bain L., Legare C. H., Hahn-Holbrook J. (2021). An infant carrier intervention and breastfeeding duration: A randomized controlled trial. Pediatrics, 148, e2020049717. 10.1542/peds.2020-049717
Matthay E. C., Glymour M. M. (2020). A graphical catalog of threats to validity: Linking social science with epidemiology. Epidemiology, 31, 376–384. 10.1097/EDE.0000000000001161
Mitha A., Piedvache A., Glorieux I., Zeitlin J., Roué J. M., Blondel B., Durox M., Burguet A., Kaminski M., Ancel P. Y., Pierrat V.; EPIPAGE-2 Neurodevelopmental Care Writing Group (2019). Unit policies and breast milk feeding at discharge of very preterm infants: The EPIPAGE-2 cohort study. Paediatric and Perinatal Epidemiology, 33, 59–69. 10.1111/ppe.12536
Neu M., Robinson J., Schmiege S. J. (2013). Influence of holding practice on preterm infant development. MCN: The American Journal of Maternal/Child Nursing, 38, 136–143. 10.1097/NMC.0b013e31827ca68c
Pineda R. G., Stransky K. E., Rogers C., Duncan M. H., Smith G. C., Neil J., Inder T. (2012). The single-patient room in the NICU: Maternal and family effects. Journal of Perinatology, 32, 545–551. 10.1038/jp.2011.144
Powers S. A., Taylor K., Tumin D., Kohler J. A. Sr. (2022). Measuring parental presence in the neonatal intensive care unit. American Journal of Perinatology, 39, 134–143. 10.1055/s-0040-1715525
Radloff L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. 10.1177/014662167700100306
Raiskila S., Axelin A., Toome L., Caballero S., Tandberg B. S., Montirosso R., Normann E., Hallberg B., Westrup B., Ewald U., Lehtonen L. (2017). Parents’ presence and parent-infant closeness in 11 neonatal intensive care units in six European countries vary between and within the countries. Acta Paediatrica, 106, 878–888. 10.1111/apa.13798
Samra H. A., Dutcher J., McGrath J. M., Foster M., Klein L., Djira G., Hansen J., Wallenburg D. (2015). Effect of skin-to-skin holding on stress in mothers of late-preterm infants: A randomized controlled trial. Advances in Neonatal Care, 15, 354–364. 10.1097/ANC.0000000000000223
Tully K. P., Holditch-Davis D., White-Traut R. C., David R., O’Shea T. M., Geraldo V. (2016). A test of kangaroo care on preterm infant breastfeeding. Journal of Obstetric, Gynecologic & Neonatal Nursing, 45, 45–61. 10.1016/j.jogn.2015.10.004
Wood N. K., Odom-Maryon T., Smart D. A. (2021). Factors associated with perceived insufficient milk in the first three months of breastfeeding. MCN American Journal of Maternal Child Nursing, 46, 223–229. 10.1097/NMC.0000000000000723
Keywords:

electronic health records; kangaroo care; maternal engagement; NICU; surveys; time use diaries

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.