The Journal of Maternal-Fetal & Neonatal Medicine
ISSN: 1476-7058 (Print) 1476-4954 (Online) Journal homepage: http://www.tandfonline.com/loi/ijmf20
Characteristics associated with intervention and follow-up attendance in a secondhand smoke exposure study for families of NICU infants Thomas F. Northrup, Charles Green, Patricia W. Evans & Angela L. Stotts To cite this article: Thomas F. Northrup, Charles Green, Patricia W. Evans & Angela L. Stotts (2015) Characteristics associated with intervention and follow-up attendance in a secondhand smoke exposure study for families of NICU infants, The Journal of Maternal-Fetal & Neonatal Medicine, 28:10, 1208-1213, DOI: 10.3109/14767058.2014.947950 To link to this article: http://dx.doi.org/10.3109/14767058.2014.947950
Accepted author version posted online: 23 Jul 2014. Published online: 14 Aug 2014. Submit your article to this journal
Article views: 22
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ijmf20 Download by: [Deakin University Library]
Date: 06 November 2015, At: 08:14
http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, 2015; 28(10): 1208–1213 ! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2014.947950
ORIGINAL ARTICLE
Characteristics associated with intervention and follow-up attendance in a secondhand smoke exposure study for families of NICU infants Thomas F. Northrup1, Charles Green2,3, Patricia W. Evans2, and Angela L. Stotts1,3 Department of Family and Community Medicine, 2Department of Pediatrics, and 3Department of Psychiatry and Behavioral Sciences, University of Texas Medical School at Houston, Houston, TX, USA
Downloaded by [Deakin University Library] at 08:14 06 November 2015
1
Abstract
Keywords
Objective: The neonatal intensive care unit (NICU) is an ideal setting to intervene with an under served population on secondhand smoke exposure (SHSe). Unfortunately, attrition may compromise outcomes. Baseline characteristics associated with intervention and follow-up attendance were investigated in mothers who participated in a novel SHSe prevention study designed for households with a smoker and a NICU-admitted infant. Methods: Intervention participants received two motivational, NICU-based counseling sessions; usual care participants received pamphlets. Home-based follow-up assessments occurred at 1, 3 and 6 months. Sociodemographic, smoking history, and psychosocial factors were analyzed. Results: Mothers from households with greater numbers of cigarettes smoked and fewer children had higher odds of both intervention and follow-up attendance. Maternal smoking abstinence (lifetime), more adults in the home and higher perceived interpersonal support were also associated with higher odds of follow-up visit completion. Conclusions: Innovative strategies are needed to engage mothers in secondhand smoke interventions, especially mothers who smoke, have lower levels of social support and have greater childcare responsibilities.
Attrition, neonatal intensive care unit, NICU, secondhand smoke exposure, SHSe
Introduction The neonatal intensive care unit (NICU) may provide an ideal setting for intervention on harmful health practices with an underserved, hard-to-reach, young parent population. Young, low-socioeconomic status adults are known to use tobacco products at higher rates [1,2] and young parents with an infant in a NICU may not be aware of or consider the dangers of secondhand smoke exposure (SHSe) to their fragile infant. Very low birth weight infants (VLBW; 1500 g; 3 lbs 5 oz), and infants requiring mechanical ventilation are particularly vulnerable to the effects of SHSe. Many develop bronchopulmonary dysplasia (BPD) after birth, resulting in higher risks of pneumonia, asthma, hospitalization and death, especially in the first 6 months after discharge [3,4]. While this population of infants is not extremely large, these infants demand extensive resources. Effective household smoking interventions would reduce the significant infant morbidity, mortality and associated costs. A recently completed SHSe-reduction trial with mothers of VLBW infants in a NICU demonstrated that a brief Address for correspondence: Thomas F. Northrup, PhD, Department of Family and Community Medicine, University of Texas Medical School at Houston, 6431 Fannin, JJL 324, Houston, TX 77030, USA. Tel: +1(713)500-6869. Fax: + 1(713)500-7598. E-mail: [email protected]
History Received 10 January 2014 Accepted 21 July 2014 Published online 13 August 2014
two-session motivational interviewing-based intervention is feasible and potentially efficacious [5]. Specifically, participants who received intervention and/or intensive home-based assessment reported significantly higher car and total (home and car) smoking bans 6-months post-discharge, compared to groups who received no intervention and intensive assessment. Moreover, motivational interviewing participants who did not have an indoor home smoking ban at baseline were significantly more likely to implement a home ban compared to usual care participants. SHSe programs in the NICU may have a significant impact, yet patient compliance can be challenging. Data from this initial study may identify characteristics associated with program participation. SHSe-reduction programs in medical settings face many barriers to participation [6–8]. For example, families with a primary caregiver who is female or a smoker were harder to retain in a SHSe-reduction trial among pediatric cancer patients [6]. Many of the mothers and families with an infant in the NICU have low household incomes, less education and are ethnic minorities with a strong potential for experiencing health disparities [9], and may not engage in their child’s care due to limited medical knowledge or access to resources [10]. A study with a similar population found that single, lesseducated, non-working and alcohol/drug using women were less likely to be retained in the intervention [8]. Infants born prematurely, or with other medical complications, can often
Intervention attend
Downloaded by [Deakin University Library] at 08:14 06 November 2015
DOI: 10.3109/14767058.2014.947950
spend weeks to months in this environment before being discharged [11], providing significant time to educate and intervene on important SHSe prevention. Thus, understanding characteristics of families caring for NICU infants that are associated with SHSe program attendance would be valuable in designing future programs. The post-discharge period is also critical as the most serious illnesses occur by 6 months for high-risk infants [3,4]. Research-based home visits serve as a marker of home-based intervention feasibility with a population. Previous NICU research investigations that had post-discharge follow-up assessments (e.g. 3, 6, 12 and 18 months) have reported high levels of attrition [12,13], and post-discharge medical care visit rates have been reported as low as 59% for graduates of the NICU [14]. Follow-up rates such as these appear to be the norm, and underscore the need to understand barriers to engagement. This exploratory study sought to gain critical information on a novel, ‘‘captive’’ population of smokers at a time during which they may be more motivated to make behavior changes. The primary aim of this secondary analysis was to identify baseline characteristics of mothers and families that may affect attendance at (i) a hospital-based SHSe program and (ii) home-based follow-up visits. Generally, it was hypothesized that mothers with fewer economic resources, greater psychosocial stress, more childcare responsibilities, and those who reported smoking and higher levels of household smoking, would have lower SHSe intervention attendance and follow-up, similar to other studies with this population [8,10,12–14].
Methods Participants and procedures This is a secondary analysis of data from The Baby’s Breath Project (www.clinicaltrials.gov, registration #: NCT00670280), one of the first studies to test a SHSe intervention with mothers caring for a NICU infant. All participants provided written informed consent. This research was approved by our local Institutional Review Board’s regulations. Full study design and outcome data are published [5]. In this study, mothers with a high-risk infant in the NICU who had at least one smoker living in their household were randomized to receive a brief hospital-based SHSe prevention intervention or were assigned to one of two control groups. Research assistants approached mothers in the NICU to determine the potential for household SHSe. Eligibility criteria included: (i) having an infant at high respiratory risk (HRR) in the NICU, with HRR defined as VLBW or having received mechanical ventilation for 412 h; (ii) reporting at least one smoker living in the household; (iii) being able to read English or Spanish and (iv) living within a 50-mile radius of the hospital (due to home-based follow-up assessments). Participants were compensated during hospital-based intervention visits (with gifts for infants) and home-based follow-up visits ($30/visit). Consented families (N ¼ 144) were randomized to one of three groups (Figure 1): (i) motivational interviewing (MI) intervention; (ii) usual care (UC) or (iii) usual care-reduced measurement (UC-RM). While their babies were in the NICU,
1209
MI participants received two individual hospital-based, MI sessions targeting SHSe, provided by experienced MI counselors. The UC and UC-RM groups received usual hospital care which was integrated with discharge planning and did not consist of separate intervention sessions. Follow-up home assessments were conducted at 1-, 3- and 6-months postdischarge for both the MI and UC groups. Research staff utilized rigorous retention procedures for all participants. For example, collecting a detailed (and frequently updated) baseline locator form with additional family and friends’ phone numbers to contact (or re-establish contact with) the participant and letters were sent to homes for missed appointments. UC-RM participants (n ¼ 40) completed a limited baseline interview and received only the 6-month assessment, in order to evaluate effects due to measurement [15] in the larger trial, and therefore were excluded from this secondary analysis, as key predictor measures were missing by design. Measures Attendance for both visit types was defined as completing a face-to-face visit; phone assessments were counted as nonattendance. Demographic, smoking-related and psychosocial data were collected via interviews or self-report. Information about the amount of smoking by the mother and others living in the household was collected with timeline follow back procedures [16]. The Interpersonal Support Evaluation Scale (ISEL) [17], the Miles Parental Perception of Stressors-NICU [18], the Perceived Stress Scale [19] and the Patient Health Questionnaire, 9-item version [20], were used to measure interpersonal support, NICU-related stress, general stress and symptoms of depression, respectively. The processes of change [21] adapted to secondhand smoke were used to measure important cognitive and behavioral change processes that potentially relate to study attendance. Data analyses Two separate analyses were conducted on: (i) intervention attendance and (ii) follow-up attendance. Participants whose infants died shortly after randomization (n ¼ 4) were removed from the intervention and follow-up assessment analyses yielding a final sample size of 67 for the intervention analysis of attendance at MI sessions (MI participants only). Eight more participants’ infants died shortly after NICU discharge and two participants moved away, with no opportunity for face-to-face follow-up. The final sample for the follow-up assessment analysis (of attendance at 1-, 3- and 6-month follow-up visits) was 91, which included MI and UC participants. The intervention attendance analysis (MI participants only) used a cumulative logit model (Proc LOGISTIC; SAS version 9.2) [22]; attendance (i.e. 0, 1 or 2 sessions) was regressed on the predictors. For the analysis of follow-up assessments (UC and MI participants), a Generalized Estimating Equation (GEE) model (Proc GENMOD; SAS version 9.2) [22] accounted for within-subject clustering due to repeated measures over time. Model building followed the general template recommended by Hosmer and Lemeshow [23]. Initial univariate
1210
T. F. Northrup et al.
J Matern Fetal Neonatal Med, 2015; 28(10): 1208–1213
Participants (N=145)
Intervention (MI) (Includes 1 Pilot; N=71)
Usual Care, Reduced Measurement (UC-RM) participants are missing key baseline, 1-month, and 3-month data (by study design) and were excluded from the analyses (n=40)
Usual Care (UC; N=34)
Exclusions: Infant deaths (n=4)
Downloaded by [Deakin University Library] at 08:14 06 November 2015
2 Hospital-Based Counseling Sessions (N=67)
Infants discharged from the hospital (N=101) Exclusions: Infant deaths (n=4) Relocation (n=2)
Exclusions: Infant deaths (n=4)
Home-Based Follow-Up Visits [1-, 3-, & 6-month visits] (N=91)
Figure 1. Study flow and attendance at study visits. Exclusions appear in boxes on the right and left side of the diagram.
analysis (i.e. one variable examined at a time) of all variables [i.e. maternal smoking status; number of household cigarette packs per month; all psychosocial measures (and subscales; see ‘‘Measures’’ section); maternal race, household income, work status, education and relationship status; and, number of kids 518 years and adults 18 years in the home] resulted in a preliminary list of predictors (i.e. variables with p50.25). The same steps identified variables for inclusion in the follow-up visit GEE model, with the exception that the model included time as a covariate. The final model selection began with all variables retained by univariate analyses examined in a single multivariate model. Next, removal of variables (with the highest p40.05) one at a time permitted evaluation of changes in relative model fit. This process was continued until the best-fitting, most parsimonious model was obtained. In the interest of space, only variables retained in the final multivariate model are reported.
Results Participants were predominantly ethnic minorities (47% Black; 29% Hispanic; 22% White, non-Hispanic and 3.0% Other), most were unemployed (74%), a majority had a high school diploma or GED or less formal education (68%) and, were unmarried (76%). Approximately half of the mothers reported being lifetime smokers, and 33% reported currently smoking. Socioeconomic status indicators showed that the average family supported five members (averaging two
children of age 18 years) and a majority of households were living on $35 000. See Stotts et al. [5] for additional details. Intervention attendance (MI group only) The model results for the intervention attendance analyses are presented in Table 1. The score test ( 2 ¼ 0.88, df ¼ 2, p ¼ 0.64) supported the assumption of proportional odds and the appropriateness of the cumulative logit model approach. Two variables were retained for the final multivariate model: the number of cigarette packs smoked per month by other household members and the number of children in the home (age518 years). For each additional cigarette pack smoked by someone other than the mother, the odds of attending intervention visits increased by a 6% (OR ¼ 1.06; 95% CI 1.00–1.11), suggesting that mothers who live with other smokers may have higher odds of attending hospital-based SHSe interventions. For each additional child in the home, the odds of attending visits decreased 42% (OR ¼ 0.58; 95% CI 0.36–0.92). Follow-up assessment attendance at 1-, 3- and 6-month visits (MI and UC groups) The model results for the follow-up assessment analyses are presented in Table 1. Five variables: mother’s lifetime smoking status, the number of cigarette packs smoked per month by other household members, number of children in
Intervention attend
DOI: 10.3109/14767058.2014.947950
1211
Table 1. Characteristics significantly associated with attendance at intervention and home-based follow-up visits. MI intervention visits Characteristic/variable Cigarette packs/month (others) Children518 years in the home
Mean (SD) 14.6 (14.6) 2.1 (1.2)
OR (95% CI)
SE
2
1.06 (1.00–1.11)* 0.58 (0.36–0.92)*
0.03 0.24
4.36 5.28
SE
Z
0.28 0.28 0.01 0.06 0.41 0.40 0.14
1.99 1.81 2.47 2.41 2.54 2.04 2.20
Follow-up home visits Characteristic/variable
Downloaded by [Deakin University Library] at 08:14 06 November 2015
Time (ref ¼ 6 month) 1 month 3 month Cigarette packs/month (others) ISEL (appraisal) Lifetime smoking (%) (mothers; ref ¼ yes) Children 518 years in the home (ref 3) Adults 18 years in the home
Mean (SD) N/A N/A 15.0 (16.3) 14.1 (2.6) 46.2% (n ¼ 42) 2.4 (1.3) 3.0 (1.3)
OR (95% CI) 0.57 0.60 1.03 1.16 2.83 2.25 1.36
(0.33–0.99)* (0.35–1.04) (1.01–1.06)* (1.03–1.31)* (1.27–6.29)* (1.03–4.88)* (1.03–1.79)*
Measures of depression, processes of change and perceived stress, along with maternal race, income, employment status and age were not significantly associated with intervention or follow-up attendance. ‘‘Children518 years in the home (ref 3)’’ represents the dichotomization of number of children in the home to 3 children or 53 children in the home (for increased parsimony in the follow-up visit analyses). ‘‘ref ¼’’ reference group for the categorical variables. SE, standard error of mean. SD, standard deviation. *p50.05.
the home, number of adults (18 years) in the home and the amount of perceived interpersonal support—specifically feeling as though there are people to talk to about stress (ISEL, appraisal subscale)—were significantly associated with home-based follow-up visit completion in the final multivariate model. Mothers who have never smoked had higher odds for completing follow-up assessments compared to those who reported lifetime smoking (OR ¼ 2.83, 95% CI 1.27–6.29). The odds of follow-up attendance by mothers increased for each additional cigarette pack smoked per month by other family members living in the home (OR ¼ 1.03, 95% CI 1.01–1.06). Mothers who lived in households with fewer than three children had increased odds of attending follow-up visits, compared to mothers who reported living in households with three or more children (OR ¼ 2.25, 95% CI 1.03–4.88). Mothers residing in homes with more adults had greater odds of attending follow-up appointments, as each additional adult in the home increased the odds of attendance by 36% (OR ¼ 1.36, 95% CI 1.03–1.79). These results suggest that mothers with fewer childrearing responsibilities, or help from other adults, may be more able to participate in SHSe program follow-up visits. The ISEL, appraisal subscale, measured the mother’s perception that she has someone with whom to talk about her difficulties, and mothers with higher scores were more likely to attend follow-up visits for each one-unit increase in this measure (OR ¼ 1.36, 95% CI 1.03–1.79). Thus, mothers perceiving more social support in their environments are more likely to comply with program requirements. The final multivariate model also included comparisons across time points (with the 6-month follow-up as the reference time point). Participants were less likely to attend 1-month (OR ¼ 0.57; 95% CI 0.33–0.99) and 3-month (OR ¼ 0.60; 95% CI 0.35–1.04) visits compared to 6-month visits. However, the 3-month odds ratio was not statistically significant (p ¼ 0.07).
Discussion Studying characteristics associated with intervention and follow-up attendance among mothers of infants admitted to a NICU allows for improved conceptualization and planning of SHSe and other health behavior programs and research. Results from this study indicated that greater reported levels of household smoking by members other than the mother and fewer children in the home were associated with better hospital-based intervention attendance and home-based follow-up attendance. Home-based follow-up attendance was also associated with mothers’ report of being a smoker at some point during her life, a greater number of adults in the home and higher perceived interpersonal support. Some of these findings are likely specific to mothers who have smoked or live in homes with smokers, yet other results may be more generally applicable to health behavior intervention research with NICU mothers and families. One unanticipated finding was that mothers who reported higher levels of household smoking by others were more likely to attend the hospital-based SHSe intervention and follow-up research appointments. Mothers residing in a home with smokers may be especially concerned about the impact of the SHSe on their infants and therefore are more interested in and willing to receive SHSe interventions. Not surprisingly, mothers who had more children (age 518 years) did not attend intervention and follow-up visits as often, even when visits were scheduled to be conducted in the home. The demands of caring for other children likely make it more difficult for these mothers to visit the NICU or participate in health programs, compared to first-time mothers, who may have fewer responsibilities. Additional family and maternal characteristics were associated with follow-up attendance. Mothers living with higher numbers of other adults attended more follow-up visits, suggesting that other adults in the home may have lessened childrearing or other home responsibilities. In line with a
Downloaded by [Deakin University Library] at 08:14 06 November 2015
1212
T. F. Northrup et al.
priori hypotheses, mothers who reported lifetime smoking at baseline were less likely to schedule and complete homebased follow-up appointments with the study staff. The latter finding is made more complex when considering that mothers who reported higher levels of baseline smoking by other household members had greater odds to be followed across time points. There may be an interaction between maternal smoking status and the level of reported smoking by other household members. Specifically, mothers who have never smoked, yet experience greater levels of household smoking by others may be more concerned about their infants’ exposure to secondhand smoke and thus more likely to continue with the study. Extra incentives may prove necessary to engage mothers who smoke or who have smoked previously [24]. Likewise, additional resources or novel methods may be needed to assist those with more children in attending SHSe interventions. The tendency for mothers with higher perceived interpersonal support to attend follow-up appointments deserves further consideration. Perhaps mothers with higher perceived support are better at finding help to care for their children. Or, they may feel better buffered against the stressors of caring for a vulnerable child and are more capable of handling the extra demands of a SHSe program. Furthermore, it may be that mothers of infants in the NICU with strong social skills may be able to draw on and perceive research staff as additional supports. Also, 1- and 3-month assessment visits were less likely to be obtained with mothers of an infant in the NICU in general, which may be a function of the initial stress of having a new, fragile infant at home following discharge. Past research on barriers to study retention has advised providing transportation assistance, providing bilingual staff, increasing flexibility around meeting times, including other caregivers, and reducing stigma around the area being investigated [25]. Research conducted on medical appointment follow-ups for graduates of the NICU suggested that increased patient contact (e.g. phone check-ins within 1 week of discharge), early intervention, and working with hospital staff to reduce patient transfers, when possible, may improve attendance [14]. Following these recommendations is encouraged. In addition, methods to assess and increase interpersonal support, offering childcare assistance, and implementing novel methods (e.g. incentives) to reduce resistance from maternal and other household smokers are needed. These post-discharge follow-up rates were higher than in previous research with NICU samples [12], suggesting benefit of home-based intervention and follow-up. This study had limitations that deserve comment; most notably a small-to-moderate sample size may limit the generalizability of the findings. We believe the modest sample size does not invalidate the overall importance of this exploratory work; however, given the relatively small body of research on SHSe prevention intervention and the difficulty engaging this understudied minority population of low-income mothers of high-risk, high-cost NICU infants. While in need of replication and extension to new settings, study findings from our large, urban NICU provide a starting point for generating and testing critical program engagement targets.
J Matern Fetal Neonatal Med, 2015; 28(10): 1208–1213
Overall, the limitations of this work are outweighed by the strengths of this novel investigation and this research is believed to be a crucial hypothesis-generating step for future work in the area of SHSe, and perhaps other behavioral programs with NICU mothers and families. NICU hospitalization offers a rare opportunity to reach young, diverse and typically low-income populations who rarely present in health-care settings, thereby precluding exposure to SHSe and other health behavior change messages important for young families at higher risk for tobacco-related health disparities [9]. Much more research is needed to understand the complexities of retention and intervention in NICU populations, with the ultimate goal of identifying effective behavioral strategies to assist in improving health outcomes among NICU infants and their families.
Acknowledgements The authors would like to thank Jeff Fox, Shireen Hayatghaibi, Tiffany Dean, Lora Bunge, and the Children’s Memorial Hermann Hospital and its NICU staff for their assistance in completing this study.
Declaration of interest The authors report no conflicts of interest. This study was supported by grant R40MC08962 through the US Department of Health and Human Services, Maternal and Child Health Research Program.
References 1. Center for Disease Control and Prevension (CDC). Tobacco use among adults – United States, 2005. MMWR Morb Mortal Wkly Rep 2006;55:1145–8. 2. Center for Disease Control and Prevension (CDC). Vital signs: current cigarette smoking among adults aged 18 years – United States 2005–2010. MMWR Morb Mortal Wkly Rep 2011;60: 1207–12. 3. Morris BH, Gard CC, Kennedy K. Rehospitalization of extremely low birth weight (ELBW) infants: are there racial/ethnic disparities? J Perinatol 2005;25:656–63. 4. Martin JA, Hamilton BE, Sutton PD, et al. Births: final data from 2003. Natl Vital Stat Rep 2005;54:1–116. 5. Stotts A, Green C, Northrup T, et al. Feasibility and efficacy of an intervention to reduce secondhand smoke exposure among infants discharged from a neonatal intensive care unit. J Perinatol 2013;33: 811–16. 6. Klosky JL, Tyc VL, Lawford J, et al. Predictors of non-participation in a randomized intervention trial to reduce environmental tobacco smoke (ETS) exposure in pediatric cancer patients. Pediatr Blood Cancer 2009;52:644–9. 7. Wilson SR, Yamada EG, Sudhakar R, et al. A controlled trial of an environmental tobacco smoke reduction intervention in low-income children with asthma. Chest 2001;120:1709–22. 8. El-Khorazaty MN, Johnson AA, Kiely M, et al. Recruitment and retention of low-income minority women in a behavioral intervention to reduce smoking, depression, and intimate partner violence during pregnancy. BMC Public Health 2007;7:233. 9. Stotts A, Evans P, Green C, et al. Secondhand smoke risk in infants discharged from a NICU: potential for significant health disparities? Nicotine Tob Res 2011;13:1015–22. 10. Coyne IT. Parental participation in care: a critical review of the literature. J Adv Nurs 1995;21:716–22. 11. McGrath JM. Developmentally supportive caregiving and technology in the NICU: isolation or merger of intervention strategies? J Perinat Neonatal Nurs 2000;14:78–91.
DOI: 10.3109/14767058.2014.947950
Downloaded by [Deakin University Library] at 08:14 06 November 2015
12. Churcher E, Egan M, Walop W, et al. Fine motor development of high-risk infants at 3, 6, 12 and 24 months. Phys Occup Ther Pediatr 1993;13:19–37. 13. Meck NE, Fowler SA, Claflin K, Rasmussen LB. Mothers’ perceptions of their NICU experience 1 and 7 months after discharge. J Early Interv 1995;19:288–301. 14. Nehra V, Pici M, Visintainer P, Kase JS. Indicators of compliance for developmental follow-up of infants discharged from a regional NICU. J Perinat Med 2009;37:677–81. 15. Hovell MF, Meltzer SB, Zakarian JM, et al. Reduction of environmental tobacco smoke exposure among asthmatic children: a controlled trial. Chest 1994;106:440–6. 16. Lewis-Esquerre JM, Colby SM, Tevyaw TO, et al. Validation of the timeline follow-back in the assessment of adolescent smoking. Drug Alcohol Depend 2005;79:33–43. 17. Cohen S, Mermelstein R, Kamarck T, Hoberman HM. Measuring the functional components of social support. In: Sarason IG, Sarason BR, eds. Social support: theory, research, and applications 1985:73–94. 18. Miles MS, Funk SG, Carlson J. Parental Stressor Scale: neonatal intensive care unit. Nurs Res 1993;42:148–52.
Intervention attend
1213
19. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav 1983;24:385–96. 20. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001;16: 606–13. 21. Cancer Prevention Research Center (CPRC). Smoking: processes of change (short form). 2011. Available from: http://www.uri.edu/ research/cprc/Measures/Smoking01.htm [last accessed 22 Jul 2011]. 22. Allison PD. Logistic regression using the SAS system: theory and application. Cary (NC): SAS Institute Inc.; 1999. 23. Hosmer DW, Lemeshow S. Applied logistic regression, 2nd edn. New York: John Wiley & Sons; 2000. 24. Stotts AL, Northrup TF, Schmitz JM, et al. Baby’s Breath II protocol development and design: a secondhand smoke exposure prevention program targeting infants discharged from a neonatal intensive care unit. Contemp Clin Trials 2013;35: 97–105. 25. Woodal A, Morgan C, Sloan C, Howard L. Barriers to participation in mental health research: are there specific gender, ethnicity and age related barriers? BMC Psychiatry 2010;10:103.
ISSN: 1476-7058 (Print) 1476-4954 (Online) Journal homepage: http://www.tandfonline.com/loi/ijmf20
Characteristics associated with intervention and follow-up attendance in a secondhand smoke exposure study for families of NICU infants Thomas F. Northrup, Charles Green, Patricia W. Evans & Angela L. Stotts To cite this article: Thomas F. Northrup, Charles Green, Patricia W. Evans & Angela L. Stotts (2015) Characteristics associated with intervention and follow-up attendance in a secondhand smoke exposure study for families of NICU infants, The Journal of Maternal-Fetal & Neonatal Medicine, 28:10, 1208-1213, DOI: 10.3109/14767058.2014.947950 To link to this article: http://dx.doi.org/10.3109/14767058.2014.947950
Accepted author version posted online: 23 Jul 2014. Published online: 14 Aug 2014. Submit your article to this journal
Article views: 22
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ijmf20 Download by: [Deakin University Library]
Date: 06 November 2015, At: 08:14
http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, 2015; 28(10): 1208–1213 ! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2014.947950
ORIGINAL ARTICLE
Characteristics associated with intervention and follow-up attendance in a secondhand smoke exposure study for families of NICU infants Thomas F. Northrup1, Charles Green2,3, Patricia W. Evans2, and Angela L. Stotts1,3 Department of Family and Community Medicine, 2Department of Pediatrics, and 3Department of Psychiatry and Behavioral Sciences, University of Texas Medical School at Houston, Houston, TX, USA
Downloaded by [Deakin University Library] at 08:14 06 November 2015
1
Abstract
Keywords
Objective: The neonatal intensive care unit (NICU) is an ideal setting to intervene with an under served population on secondhand smoke exposure (SHSe). Unfortunately, attrition may compromise outcomes. Baseline characteristics associated with intervention and follow-up attendance were investigated in mothers who participated in a novel SHSe prevention study designed for households with a smoker and a NICU-admitted infant. Methods: Intervention participants received two motivational, NICU-based counseling sessions; usual care participants received pamphlets. Home-based follow-up assessments occurred at 1, 3 and 6 months. Sociodemographic, smoking history, and psychosocial factors were analyzed. Results: Mothers from households with greater numbers of cigarettes smoked and fewer children had higher odds of both intervention and follow-up attendance. Maternal smoking abstinence (lifetime), more adults in the home and higher perceived interpersonal support were also associated with higher odds of follow-up visit completion. Conclusions: Innovative strategies are needed to engage mothers in secondhand smoke interventions, especially mothers who smoke, have lower levels of social support and have greater childcare responsibilities.
Attrition, neonatal intensive care unit, NICU, secondhand smoke exposure, SHSe
Introduction The neonatal intensive care unit (NICU) may provide an ideal setting for intervention on harmful health practices with an underserved, hard-to-reach, young parent population. Young, low-socioeconomic status adults are known to use tobacco products at higher rates [1,2] and young parents with an infant in a NICU may not be aware of or consider the dangers of secondhand smoke exposure (SHSe) to their fragile infant. Very low birth weight infants (VLBW; 1500 g; 3 lbs 5 oz), and infants requiring mechanical ventilation are particularly vulnerable to the effects of SHSe. Many develop bronchopulmonary dysplasia (BPD) after birth, resulting in higher risks of pneumonia, asthma, hospitalization and death, especially in the first 6 months after discharge [3,4]. While this population of infants is not extremely large, these infants demand extensive resources. Effective household smoking interventions would reduce the significant infant morbidity, mortality and associated costs. A recently completed SHSe-reduction trial with mothers of VLBW infants in a NICU demonstrated that a brief Address for correspondence: Thomas F. Northrup, PhD, Department of Family and Community Medicine, University of Texas Medical School at Houston, 6431 Fannin, JJL 324, Houston, TX 77030, USA. Tel: +1(713)500-6869. Fax: + 1(713)500-7598. E-mail: [email protected]
History Received 10 January 2014 Accepted 21 July 2014 Published online 13 August 2014
two-session motivational interviewing-based intervention is feasible and potentially efficacious [5]. Specifically, participants who received intervention and/or intensive home-based assessment reported significantly higher car and total (home and car) smoking bans 6-months post-discharge, compared to groups who received no intervention and intensive assessment. Moreover, motivational interviewing participants who did not have an indoor home smoking ban at baseline were significantly more likely to implement a home ban compared to usual care participants. SHSe programs in the NICU may have a significant impact, yet patient compliance can be challenging. Data from this initial study may identify characteristics associated with program participation. SHSe-reduction programs in medical settings face many barriers to participation [6–8]. For example, families with a primary caregiver who is female or a smoker were harder to retain in a SHSe-reduction trial among pediatric cancer patients [6]. Many of the mothers and families with an infant in the NICU have low household incomes, less education and are ethnic minorities with a strong potential for experiencing health disparities [9], and may not engage in their child’s care due to limited medical knowledge or access to resources [10]. A study with a similar population found that single, lesseducated, non-working and alcohol/drug using women were less likely to be retained in the intervention [8]. Infants born prematurely, or with other medical complications, can often
Intervention attend
Downloaded by [Deakin University Library] at 08:14 06 November 2015
DOI: 10.3109/14767058.2014.947950
spend weeks to months in this environment before being discharged [11], providing significant time to educate and intervene on important SHSe prevention. Thus, understanding characteristics of families caring for NICU infants that are associated with SHSe program attendance would be valuable in designing future programs. The post-discharge period is also critical as the most serious illnesses occur by 6 months for high-risk infants [3,4]. Research-based home visits serve as a marker of home-based intervention feasibility with a population. Previous NICU research investigations that had post-discharge follow-up assessments (e.g. 3, 6, 12 and 18 months) have reported high levels of attrition [12,13], and post-discharge medical care visit rates have been reported as low as 59% for graduates of the NICU [14]. Follow-up rates such as these appear to be the norm, and underscore the need to understand barriers to engagement. This exploratory study sought to gain critical information on a novel, ‘‘captive’’ population of smokers at a time during which they may be more motivated to make behavior changes. The primary aim of this secondary analysis was to identify baseline characteristics of mothers and families that may affect attendance at (i) a hospital-based SHSe program and (ii) home-based follow-up visits. Generally, it was hypothesized that mothers with fewer economic resources, greater psychosocial stress, more childcare responsibilities, and those who reported smoking and higher levels of household smoking, would have lower SHSe intervention attendance and follow-up, similar to other studies with this population [8,10,12–14].
Methods Participants and procedures This is a secondary analysis of data from The Baby’s Breath Project (www.clinicaltrials.gov, registration #: NCT00670280), one of the first studies to test a SHSe intervention with mothers caring for a NICU infant. All participants provided written informed consent. This research was approved by our local Institutional Review Board’s regulations. Full study design and outcome data are published [5]. In this study, mothers with a high-risk infant in the NICU who had at least one smoker living in their household were randomized to receive a brief hospital-based SHSe prevention intervention or were assigned to one of two control groups. Research assistants approached mothers in the NICU to determine the potential for household SHSe. Eligibility criteria included: (i) having an infant at high respiratory risk (HRR) in the NICU, with HRR defined as VLBW or having received mechanical ventilation for 412 h; (ii) reporting at least one smoker living in the household; (iii) being able to read English or Spanish and (iv) living within a 50-mile radius of the hospital (due to home-based follow-up assessments). Participants were compensated during hospital-based intervention visits (with gifts for infants) and home-based follow-up visits ($30/visit). Consented families (N ¼ 144) were randomized to one of three groups (Figure 1): (i) motivational interviewing (MI) intervention; (ii) usual care (UC) or (iii) usual care-reduced measurement (UC-RM). While their babies were in the NICU,
1209
MI participants received two individual hospital-based, MI sessions targeting SHSe, provided by experienced MI counselors. The UC and UC-RM groups received usual hospital care which was integrated with discharge planning and did not consist of separate intervention sessions. Follow-up home assessments were conducted at 1-, 3- and 6-months postdischarge for both the MI and UC groups. Research staff utilized rigorous retention procedures for all participants. For example, collecting a detailed (and frequently updated) baseline locator form with additional family and friends’ phone numbers to contact (or re-establish contact with) the participant and letters were sent to homes for missed appointments. UC-RM participants (n ¼ 40) completed a limited baseline interview and received only the 6-month assessment, in order to evaluate effects due to measurement [15] in the larger trial, and therefore were excluded from this secondary analysis, as key predictor measures were missing by design. Measures Attendance for both visit types was defined as completing a face-to-face visit; phone assessments were counted as nonattendance. Demographic, smoking-related and psychosocial data were collected via interviews or self-report. Information about the amount of smoking by the mother and others living in the household was collected with timeline follow back procedures [16]. The Interpersonal Support Evaluation Scale (ISEL) [17], the Miles Parental Perception of Stressors-NICU [18], the Perceived Stress Scale [19] and the Patient Health Questionnaire, 9-item version [20], were used to measure interpersonal support, NICU-related stress, general stress and symptoms of depression, respectively. The processes of change [21] adapted to secondhand smoke were used to measure important cognitive and behavioral change processes that potentially relate to study attendance. Data analyses Two separate analyses were conducted on: (i) intervention attendance and (ii) follow-up attendance. Participants whose infants died shortly after randomization (n ¼ 4) were removed from the intervention and follow-up assessment analyses yielding a final sample size of 67 for the intervention analysis of attendance at MI sessions (MI participants only). Eight more participants’ infants died shortly after NICU discharge and two participants moved away, with no opportunity for face-to-face follow-up. The final sample for the follow-up assessment analysis (of attendance at 1-, 3- and 6-month follow-up visits) was 91, which included MI and UC participants. The intervention attendance analysis (MI participants only) used a cumulative logit model (Proc LOGISTIC; SAS version 9.2) [22]; attendance (i.e. 0, 1 or 2 sessions) was regressed on the predictors. For the analysis of follow-up assessments (UC and MI participants), a Generalized Estimating Equation (GEE) model (Proc GENMOD; SAS version 9.2) [22] accounted for within-subject clustering due to repeated measures over time. Model building followed the general template recommended by Hosmer and Lemeshow [23]. Initial univariate
1210
T. F. Northrup et al.
J Matern Fetal Neonatal Med, 2015; 28(10): 1208–1213
Participants (N=145)
Intervention (MI) (Includes 1 Pilot; N=71)
Usual Care, Reduced Measurement (UC-RM) participants are missing key baseline, 1-month, and 3-month data (by study design) and were excluded from the analyses (n=40)
Usual Care (UC; N=34)
Exclusions: Infant deaths (n=4)
Downloaded by [Deakin University Library] at 08:14 06 November 2015
2 Hospital-Based Counseling Sessions (N=67)
Infants discharged from the hospital (N=101) Exclusions: Infant deaths (n=4) Relocation (n=2)
Exclusions: Infant deaths (n=4)
Home-Based Follow-Up Visits [1-, 3-, & 6-month visits] (N=91)
Figure 1. Study flow and attendance at study visits. Exclusions appear in boxes on the right and left side of the diagram.
analysis (i.e. one variable examined at a time) of all variables [i.e. maternal smoking status; number of household cigarette packs per month; all psychosocial measures (and subscales; see ‘‘Measures’’ section); maternal race, household income, work status, education and relationship status; and, number of kids 518 years and adults 18 years in the home] resulted in a preliminary list of predictors (i.e. variables with p50.25). The same steps identified variables for inclusion in the follow-up visit GEE model, with the exception that the model included time as a covariate. The final model selection began with all variables retained by univariate analyses examined in a single multivariate model. Next, removal of variables (with the highest p40.05) one at a time permitted evaluation of changes in relative model fit. This process was continued until the best-fitting, most parsimonious model was obtained. In the interest of space, only variables retained in the final multivariate model are reported.
Results Participants were predominantly ethnic minorities (47% Black; 29% Hispanic; 22% White, non-Hispanic and 3.0% Other), most were unemployed (74%), a majority had a high school diploma or GED or less formal education (68%) and, were unmarried (76%). Approximately half of the mothers reported being lifetime smokers, and 33% reported currently smoking. Socioeconomic status indicators showed that the average family supported five members (averaging two
children of age 18 years) and a majority of households were living on $35 000. See Stotts et al. [5] for additional details. Intervention attendance (MI group only) The model results for the intervention attendance analyses are presented in Table 1. The score test ( 2 ¼ 0.88, df ¼ 2, p ¼ 0.64) supported the assumption of proportional odds and the appropriateness of the cumulative logit model approach. Two variables were retained for the final multivariate model: the number of cigarette packs smoked per month by other household members and the number of children in the home (age518 years). For each additional cigarette pack smoked by someone other than the mother, the odds of attending intervention visits increased by a 6% (OR ¼ 1.06; 95% CI 1.00–1.11), suggesting that mothers who live with other smokers may have higher odds of attending hospital-based SHSe interventions. For each additional child in the home, the odds of attending visits decreased 42% (OR ¼ 0.58; 95% CI 0.36–0.92). Follow-up assessment attendance at 1-, 3- and 6-month visits (MI and UC groups) The model results for the follow-up assessment analyses are presented in Table 1. Five variables: mother’s lifetime smoking status, the number of cigarette packs smoked per month by other household members, number of children in
Intervention attend
DOI: 10.3109/14767058.2014.947950
1211
Table 1. Characteristics significantly associated with attendance at intervention and home-based follow-up visits. MI intervention visits Characteristic/variable Cigarette packs/month (others) Children518 years in the home
Mean (SD) 14.6 (14.6) 2.1 (1.2)
OR (95% CI)
SE
2
1.06 (1.00–1.11)* 0.58 (0.36–0.92)*
0.03 0.24
4.36 5.28
SE
Z
0.28 0.28 0.01 0.06 0.41 0.40 0.14
1.99 1.81 2.47 2.41 2.54 2.04 2.20
Follow-up home visits Characteristic/variable
Downloaded by [Deakin University Library] at 08:14 06 November 2015
Time (ref ¼ 6 month) 1 month 3 month Cigarette packs/month (others) ISEL (appraisal) Lifetime smoking (%) (mothers; ref ¼ yes) Children 518 years in the home (ref 3) Adults 18 years in the home
Mean (SD) N/A N/A 15.0 (16.3) 14.1 (2.6) 46.2% (n ¼ 42) 2.4 (1.3) 3.0 (1.3)
OR (95% CI) 0.57 0.60 1.03 1.16 2.83 2.25 1.36
(0.33–0.99)* (0.35–1.04) (1.01–1.06)* (1.03–1.31)* (1.27–6.29)* (1.03–4.88)* (1.03–1.79)*
Measures of depression, processes of change and perceived stress, along with maternal race, income, employment status and age were not significantly associated with intervention or follow-up attendance. ‘‘Children518 years in the home (ref 3)’’ represents the dichotomization of number of children in the home to 3 children or 53 children in the home (for increased parsimony in the follow-up visit analyses). ‘‘ref ¼’’ reference group for the categorical variables. SE, standard error of mean. SD, standard deviation. *p50.05.
the home, number of adults (18 years) in the home and the amount of perceived interpersonal support—specifically feeling as though there are people to talk to about stress (ISEL, appraisal subscale)—were significantly associated with home-based follow-up visit completion in the final multivariate model. Mothers who have never smoked had higher odds for completing follow-up assessments compared to those who reported lifetime smoking (OR ¼ 2.83, 95% CI 1.27–6.29). The odds of follow-up attendance by mothers increased for each additional cigarette pack smoked per month by other family members living in the home (OR ¼ 1.03, 95% CI 1.01–1.06). Mothers who lived in households with fewer than three children had increased odds of attending follow-up visits, compared to mothers who reported living in households with three or more children (OR ¼ 2.25, 95% CI 1.03–4.88). Mothers residing in homes with more adults had greater odds of attending follow-up appointments, as each additional adult in the home increased the odds of attendance by 36% (OR ¼ 1.36, 95% CI 1.03–1.79). These results suggest that mothers with fewer childrearing responsibilities, or help from other adults, may be more able to participate in SHSe program follow-up visits. The ISEL, appraisal subscale, measured the mother’s perception that she has someone with whom to talk about her difficulties, and mothers with higher scores were more likely to attend follow-up visits for each one-unit increase in this measure (OR ¼ 1.36, 95% CI 1.03–1.79). Thus, mothers perceiving more social support in their environments are more likely to comply with program requirements. The final multivariate model also included comparisons across time points (with the 6-month follow-up as the reference time point). Participants were less likely to attend 1-month (OR ¼ 0.57; 95% CI 0.33–0.99) and 3-month (OR ¼ 0.60; 95% CI 0.35–1.04) visits compared to 6-month visits. However, the 3-month odds ratio was not statistically significant (p ¼ 0.07).
Discussion Studying characteristics associated with intervention and follow-up attendance among mothers of infants admitted to a NICU allows for improved conceptualization and planning of SHSe and other health behavior programs and research. Results from this study indicated that greater reported levels of household smoking by members other than the mother and fewer children in the home were associated with better hospital-based intervention attendance and home-based follow-up attendance. Home-based follow-up attendance was also associated with mothers’ report of being a smoker at some point during her life, a greater number of adults in the home and higher perceived interpersonal support. Some of these findings are likely specific to mothers who have smoked or live in homes with smokers, yet other results may be more generally applicable to health behavior intervention research with NICU mothers and families. One unanticipated finding was that mothers who reported higher levels of household smoking by others were more likely to attend the hospital-based SHSe intervention and follow-up research appointments. Mothers residing in a home with smokers may be especially concerned about the impact of the SHSe on their infants and therefore are more interested in and willing to receive SHSe interventions. Not surprisingly, mothers who had more children (age 518 years) did not attend intervention and follow-up visits as often, even when visits were scheduled to be conducted in the home. The demands of caring for other children likely make it more difficult for these mothers to visit the NICU or participate in health programs, compared to first-time mothers, who may have fewer responsibilities. Additional family and maternal characteristics were associated with follow-up attendance. Mothers living with higher numbers of other adults attended more follow-up visits, suggesting that other adults in the home may have lessened childrearing or other home responsibilities. In line with a
Downloaded by [Deakin University Library] at 08:14 06 November 2015
1212
T. F. Northrup et al.
priori hypotheses, mothers who reported lifetime smoking at baseline were less likely to schedule and complete homebased follow-up appointments with the study staff. The latter finding is made more complex when considering that mothers who reported higher levels of baseline smoking by other household members had greater odds to be followed across time points. There may be an interaction between maternal smoking status and the level of reported smoking by other household members. Specifically, mothers who have never smoked, yet experience greater levels of household smoking by others may be more concerned about their infants’ exposure to secondhand smoke and thus more likely to continue with the study. Extra incentives may prove necessary to engage mothers who smoke or who have smoked previously [24]. Likewise, additional resources or novel methods may be needed to assist those with more children in attending SHSe interventions. The tendency for mothers with higher perceived interpersonal support to attend follow-up appointments deserves further consideration. Perhaps mothers with higher perceived support are better at finding help to care for their children. Or, they may feel better buffered against the stressors of caring for a vulnerable child and are more capable of handling the extra demands of a SHSe program. Furthermore, it may be that mothers of infants in the NICU with strong social skills may be able to draw on and perceive research staff as additional supports. Also, 1- and 3-month assessment visits were less likely to be obtained with mothers of an infant in the NICU in general, which may be a function of the initial stress of having a new, fragile infant at home following discharge. Past research on barriers to study retention has advised providing transportation assistance, providing bilingual staff, increasing flexibility around meeting times, including other caregivers, and reducing stigma around the area being investigated [25]. Research conducted on medical appointment follow-ups for graduates of the NICU suggested that increased patient contact (e.g. phone check-ins within 1 week of discharge), early intervention, and working with hospital staff to reduce patient transfers, when possible, may improve attendance [14]. Following these recommendations is encouraged. In addition, methods to assess and increase interpersonal support, offering childcare assistance, and implementing novel methods (e.g. incentives) to reduce resistance from maternal and other household smokers are needed. These post-discharge follow-up rates were higher than in previous research with NICU samples [12], suggesting benefit of home-based intervention and follow-up. This study had limitations that deserve comment; most notably a small-to-moderate sample size may limit the generalizability of the findings. We believe the modest sample size does not invalidate the overall importance of this exploratory work; however, given the relatively small body of research on SHSe prevention intervention and the difficulty engaging this understudied minority population of low-income mothers of high-risk, high-cost NICU infants. While in need of replication and extension to new settings, study findings from our large, urban NICU provide a starting point for generating and testing critical program engagement targets.
J Matern Fetal Neonatal Med, 2015; 28(10): 1208–1213
Overall, the limitations of this work are outweighed by the strengths of this novel investigation and this research is believed to be a crucial hypothesis-generating step for future work in the area of SHSe, and perhaps other behavioral programs with NICU mothers and families. NICU hospitalization offers a rare opportunity to reach young, diverse and typically low-income populations who rarely present in health-care settings, thereby precluding exposure to SHSe and other health behavior change messages important for young families at higher risk for tobacco-related health disparities [9]. Much more research is needed to understand the complexities of retention and intervention in NICU populations, with the ultimate goal of identifying effective behavioral strategies to assist in improving health outcomes among NICU infants and their families.
Acknowledgements The authors would like to thank Jeff Fox, Shireen Hayatghaibi, Tiffany Dean, Lora Bunge, and the Children’s Memorial Hermann Hospital and its NICU staff for their assistance in completing this study.
Declaration of interest The authors report no conflicts of interest. This study was supported by grant R40MC08962 through the US Department of Health and Human Services, Maternal and Child Health Research Program.
References 1. Center for Disease Control and Prevension (CDC). Tobacco use among adults – United States, 2005. MMWR Morb Mortal Wkly Rep 2006;55:1145–8. 2. Center for Disease Control and Prevension (CDC). Vital signs: current cigarette smoking among adults aged 18 years – United States 2005–2010. MMWR Morb Mortal Wkly Rep 2011;60: 1207–12. 3. Morris BH, Gard CC, Kennedy K. Rehospitalization of extremely low birth weight (ELBW) infants: are there racial/ethnic disparities? J Perinatol 2005;25:656–63. 4. Martin JA, Hamilton BE, Sutton PD, et al. Births: final data from 2003. Natl Vital Stat Rep 2005;54:1–116. 5. Stotts A, Green C, Northrup T, et al. Feasibility and efficacy of an intervention to reduce secondhand smoke exposure among infants discharged from a neonatal intensive care unit. J Perinatol 2013;33: 811–16. 6. Klosky JL, Tyc VL, Lawford J, et al. Predictors of non-participation in a randomized intervention trial to reduce environmental tobacco smoke (ETS) exposure in pediatric cancer patients. Pediatr Blood Cancer 2009;52:644–9. 7. Wilson SR, Yamada EG, Sudhakar R, et al. A controlled trial of an environmental tobacco smoke reduction intervention in low-income children with asthma. Chest 2001;120:1709–22. 8. El-Khorazaty MN, Johnson AA, Kiely M, et al. Recruitment and retention of low-income minority women in a behavioral intervention to reduce smoking, depression, and intimate partner violence during pregnancy. BMC Public Health 2007;7:233. 9. Stotts A, Evans P, Green C, et al. Secondhand smoke risk in infants discharged from a NICU: potential for significant health disparities? Nicotine Tob Res 2011;13:1015–22. 10. Coyne IT. Parental participation in care: a critical review of the literature. J Adv Nurs 1995;21:716–22. 11. McGrath JM. Developmentally supportive caregiving and technology in the NICU: isolation or merger of intervention strategies? J Perinat Neonatal Nurs 2000;14:78–91.
DOI: 10.3109/14767058.2014.947950
Downloaded by [Deakin University Library] at 08:14 06 November 2015
12. Churcher E, Egan M, Walop W, et al. Fine motor development of high-risk infants at 3, 6, 12 and 24 months. Phys Occup Ther Pediatr 1993;13:19–37. 13. Meck NE, Fowler SA, Claflin K, Rasmussen LB. Mothers’ perceptions of their NICU experience 1 and 7 months after discharge. J Early Interv 1995;19:288–301. 14. Nehra V, Pici M, Visintainer P, Kase JS. Indicators of compliance for developmental follow-up of infants discharged from a regional NICU. J Perinat Med 2009;37:677–81. 15. Hovell MF, Meltzer SB, Zakarian JM, et al. Reduction of environmental tobacco smoke exposure among asthmatic children: a controlled trial. Chest 1994;106:440–6. 16. Lewis-Esquerre JM, Colby SM, Tevyaw TO, et al. Validation of the timeline follow-back in the assessment of adolescent smoking. Drug Alcohol Depend 2005;79:33–43. 17. Cohen S, Mermelstein R, Kamarck T, Hoberman HM. Measuring the functional components of social support. In: Sarason IG, Sarason BR, eds. Social support: theory, research, and applications 1985:73–94. 18. Miles MS, Funk SG, Carlson J. Parental Stressor Scale: neonatal intensive care unit. Nurs Res 1993;42:148–52.
Intervention attend
1213
19. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav 1983;24:385–96. 20. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001;16: 606–13. 21. Cancer Prevention Research Center (CPRC). Smoking: processes of change (short form). 2011. Available from: http://www.uri.edu/ research/cprc/Measures/Smoking01.htm [last accessed 22 Jul 2011]. 22. Allison PD. Logistic regression using the SAS system: theory and application. Cary (NC): SAS Institute Inc.; 1999. 23. Hosmer DW, Lemeshow S. Applied logistic regression, 2nd edn. New York: John Wiley & Sons; 2000. 24. Stotts AL, Northrup TF, Schmitz JM, et al. Baby’s Breath II protocol development and design: a secondhand smoke exposure prevention program targeting infants discharged from a neonatal intensive care unit. Contemp Clin Trials 2013;35: 97–105. 25. Woodal A, Morgan C, Sloan C, Howard L. Barriers to participation in mental health research: are there specific gender, ethnicity and age related barriers? BMC Psychiatry 2010;10:103.
