Interagency collaboration and receipt of substance abuse treatment services for child welfare-involved caregivers

Journal of Substance Abuse Treatment 79 (2017) 20–28

Contents lists available at ScienceDirect

Journal of Substance Abuse Treatment

Interagency collaboration and receipt of substance abuse treatment services for child welfare-involved caregivers☆ Amy S. He Graduate School of Social Work, University of Denver, Craig Hall, 2148 S. High St., Denver, CO, USA

a r t i c l e

i n f o

Article history: Received 10 January 2017 Received in revised form 5 May 2017 Accepted 9 May 2017 Keywords: Caregiver Child welfare Collaboration Service delivery

a b s t r a c t Caregivers dealing with problematic substance use pose persistent challenges for families involved with the child welfare (CW) system. Research has indicated that receipt of substance use disorder (SUD) services help improve family outcomes. However, there are many challenging stages of intervention in the SUD treatment process, including detection, assessment, referral, entry, and completion. Considerable work is needed to illuminate factors that strengthen the delivery of SUD-related services at various points in the treatment services continuum. Although a growing body of work has focused on individual-level correlates, few studies have examined organizational factors that potentially affect the delivery of SUD-related services. This study sought to further understanding of the relationship between CW organizational factors (interagency collaboration and organizational resources) and delivery of SUD-related services in a nationally representative sample of CW-involved caregivers. In this study sample, engagement in collaboration through a memorandum of understanding (MOU) and co-location supported caregiver receipt of a referral to SUD services. Caregivers were more likely to receive a formal assessment for SUD problems when their CW agencies reported the availability of a standardized SUD assessment tool. Also, having arrangements with SUD agencies so that CW-involved families had priority status to enter treatment was pertinent to caregiver receipt of SUD treatment services. These results provide evidence that engagement in collaboration activities and greater organizational resources can increase an organization's capacity to deliver services to clients. © 2017 Elsevier Inc. All rights reserved.

1. Introduction Caregivers dealing with problematic substance use pose persistent challenges for families involved with the child welfare (CW) system (U.S. Government Accountability Office, 1998; Wulczyn, Ernst, & Fisher, 2011; Young, Boles, & Otero, 2007). This includes associations with negative outcomes such as lower rates of reunification, higher rates of out-of-home placement, and reentry to CW (Barth, Gibbons, & Guo, 2006; Brook & McDonald, 2007; Choi & Ryan, 2006; Grella, Hser, & Huang, 2006; U.S. Department of Health and Human Services, 1999; Vanderploeg et al., 2007). Additionally, stipulations from the 1997 Adoption and Safe Families Act (ASFA), which require permanency hearings to take place within 12 months of a child being placed in foster care, place parents with substance use disorders (SUD) at greater risk for losing their parental rights (Green, Furrer, Worcel, Burrus, & Finigan, 2007). An estimated of up to two thirds CW-involved caregivers

☆ All contributing authors have no relevant financial interests pertaining to this manuscript and certify that there are no conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript. E-mail address: [email protected].

http://dx.doi.org/10.1016/j.jsat.2017.05.006 0740-5472/© 2017 Elsevier Inc. All rights reserved.

are affected by SUDs (U.S. Government Accountability Office, 1998; Wulczyn et al., 2011; Young et al., 2007), making delivery of SUD-related services crucial to improving child and family well-being outcomes. Moreover, research has indicated that receipt of SUD services such as treatment services help improve family outcomes, including children spending fewer days in foster care and a greater likelihood of family reunification (Green, Furrer, et al., 2007; Green, Rockhill, & Furrer, 2007; Ryan, Marsh, Testa, & Louderman, 2006). However, research has found many challenging stages of intervention in the SUD treatment process, including detection, assessment, referral, entry, and completion (Arria & Thoreson, 2007; Belenko et al., 2017; Brady & Ashley, 2005; Redko, Rapp, & Carlson, 2006; Xu, Rapp, Wang, & Carlson, 2008). Indeed, many different behaviors, actions, and circumstances at various levels (e.g., individual, staff, and agency characteristics) influence both the delivery of and engagement in these services for CW-involved families. Considerable work is needed to illuminate factors that strengthen the delivery of SUD-related services at various points in the treatment services continuum, especially for families receiving services across service systems (Belenko et al., 2017). Although a growing body of work has focused on individuallevel correlates (e.g., child, family, and caseworker; Choi & Ryan, 2006; Chuang, Wells, Bellettiere, & Cross, 2013; Grella, Needell, Shi, & Hser,

A.S. He / Journal of Substance Abuse Treatment 79 (2017) 20–28

2009; Ryan, Choi, Hong, Hernandez, & Larrison, 2008), few studies have examined organizational factors that potentially affect the delivery of SUD-related services. This work sought to further understanding of the relationship between CW organizational contextual factors and delivery of SUD-related services in a nationally representative sample of CW-involved caregivers. In particular, drawing from interorganizational theories of collaboration, this study examined the role of interagency collaboration between CW and drug and alcohol service [DAS] agencies and the availability of SUD resources in CW agencies in influencing the delivery of SUD-related services to this population. 1.1. Interorganizational theories of collaboration Alter and Hage's (1993) interorganizational collaboration theory expanded on several theories that provide context for the connection between engagement in collaboration activities and organizational environmental factors in the delivery of services to high-risk populations. One of the theories they used was rational choice theory (Alter & Hage, 1993; Postmus & Hahn, 2007), which posits that organizations that provide services to the same clientele (particularly those with high needs) collaborate to increase service capacity and meet client needs. Additionally, according to resource-based theories (Dyer & Singh, 1998; Pfeffer & Salancik, 1978), organizations also collaborate with partners when that partner can provide or contribute to resources or capacities that the organization does not possess. Therefore, CW agencies may be engaging in collaborative efforts with DAS providers to improve delivery and connection to SUD-related services for caregivers dealing with SUD problems (He, 2015; Pfeffer & Salancik, 1978). Moreover, because CW agencies traditionally provide nonspecialized services, collaborating with specialized services providers is vital to providing expedited services (e.g., SUD-related treatment services) to vulnerable families. The few studies that have examined organizational factors such as collaboration in delivering SUD-related services to caregivers have been mostly atheoretical. However, these studies indicated that collaboration between CW agencies and DAS providers was especially effective in increasing caregiver receipt of formal assessment by DAS specialists, referrals to SUD services, and access to SUD treatment services (Aarons, Hurlburt, & Horwitz, 2011; Green, Furrer, et al., 2007; Ryan et al., 2008; Traube, He, Zhu, Scalise, & Richardson, 2015; Wells & Chuang, 2012). Although collaboration with DAS providers has been found to promote access and referral to SUD treatment services, there is a lack of understanding regarding which aspects of collaboration or what collaboration strategy contributes to strengthening delivery of SUD-related services. 1.2. Organizational resources and SUD services Emergent research has indicated that availability of organizational resources, such as the use of a standardized SUD assessment instrument, may reduce subjectivity and better equip CW caseworkers with the tools necessary for accurate identification and assessment of SUD problems (Chuang et al., 2013; Feit, Fisher, Cummings, & Peery, 2015). In turn, having adequate resources to screen for SUD needs can support assessment of and referral to SUD services and facilitate entry into SUD treatment for CW-involved caregivers at high risk of SUD problems (Traube et al., 2015). On the other hand, even if SUD needs have been identified, additional organizational resource constraints such as limited availability of SUD service agencies and long wait lists have posed challenges in delivering SUD-related services to a wide range of individuals, including CW-involved caregivers (Arria & Thoreson, 2007; Brucker, 2010; Small, Curran, & Booth, 2010). Additionally, organizational resource constrictions including worker burden related to high caseloads can also affect caseworker service delivery (Burton, 2010; Ford, Cerasoli, Higgins, & Decesare, 2011).

21

1.3. Caregiver characteristics and SUD services Extensive research has explored individual-level caregiver factors associated with the use of various health services, primarily demographic characteristics such as age, gender, race and ethnicity, and education (Andersen, 1995; Kimerling & Baumrind, 2005; Small, 2015; Stein, Andersen, & Gelberg, 2007). For example, a review of the literature found that women with SUDs were less likely than men to enter treatment (Greenfield et al., 2007). Associations have also been found between caregivers' age, race and ethnicity, education, and poverty and receipt of SUD treatment (Grella et al., 2006; Small, 2015). Moreover, CW-involved caregivers dealing with SUD-related problems participate in treatment services at a lower rate than those without such involvement (Choi & Ryan, 2006; Ryan et al., 2008). Hence, accounting for caregivers' individual characteristics, the goal of this paper was to examine the relationship between organizational factors of interagency collaboration and organizational resources and the delivery of SUD-related services (assessment, referral, and treatment) to CW-involved caregivers at high risk of SUD problems. 2. Methods 2.1. Study design The current study used data from the second cohort of families from the National Survey of Child and Adolescent Well-Being (NSCAW II). NSCAW II is the only national longitudinal study of families investigated by U.S. child protective services agencies. Researchers employed a complex sampling approach involving two stages of stratification. The primary sampling units were county CW agencies (81 located in 30 states) and the secondary sampling units were children (and their families) randomly chosen from a list of all children investigated by sampled CW agencies between February 2008 and April 2009. Children were given a unique identification number which was used to connect them to corresponding data collected from: (a) the CW agency in which they were involved with; (b) the investigative caseworkers who were assigned to the children's cases; and (c) the children's primary caregivers. The NSCAW II dataset contained analysis weights that accounted for the complex sampling design (e.g., issues of nested data, missingness) and the estimation of differential probabilities of inclusion in the sample. Additionally, the NSCAW study used two-level probability weights (adjusted for stratification by state), with person-level weights at level 1 and agency weights at level 2. More comprehensive information about NSCAW II's study design can be found elsewhere (Dowd et al., 2012). For this current study, NSCAW II baseline data were provided by local agency directors, investigative caseworkers, and primary caregivers. Local CW agency directors provided information on CW organizational contexts and represented counties participating in the NSCAW II. When an agency director had oversight over multiple participating counties, he/she completed separate interviews for each county. Investigative caseworkers were required to access written case records for the families (Dowd et al., 2010) and asked to provide information on the services received by families. Data were also collected from interviews with primary caregivers identified as the person most involved with the child on a day-to-day basis. 2.2. Analytic sample The initial NSCAW II sample consisted of 5873 families, of which 5091 cases included investigative caseworker interviews; data were collected for one unique child and one unique primary caregiver for each family. The study sample was restricted to primary caregivers who met criteria for harmful use of or dependence on alcohol or drugs on two validated instruments (caregiver self-report, as described in the variables section) and those reported by CW caseworkers as

22

A.S. He / Journal of Substance Abuse Treatment 79 (2017) 20–28

needing SUD services. Both of these reporting sources have been used as criteria for identifying SUD problems using the NSCAW II data (Bunger, Chuang, & McBeath, 2012; Chuang et al., 2013). Given the challenges of underreporting (Grekin et al., 2010; Marsh, Smith, & Bruni, 2011) and underidentification of SUDs in this population (Chuang et al., 2013; Schroeder, Lemieux, & Pogue, 2008), these inclusion criteria were intended to capture caregivers at high risk off SUD problems and the delivery of services to said population. Application of these inclusion criteria resulted in a sample of 1651 caregivers (see Table 1). Preliminary analyses also indicated that 83% of the primary caregivers in this study sample were biological parents. 2.3. Variables 2.3.1. SUD problems Caregivers with high risk of SUD problems were identified based on caregiver and caseworker reports of SUD needs and problems. Harmful use of or dependence on alcohol or drugs was identified based on each primary caregiver's self-reported responses on two validated instruments: the Alcohol Use Disorders Identification Test (AUDIT) and the Drug Abuse Screening Test (DAST-20). The AUDIT was developed by the World Health Organization and assesses the amount and frequency of alcohol consumption (three items) and adverse consequences associated with drinking (seven items; Babor, Higgins-Biddle, Saunders, & Monteiro, 2001). Psychometric properties of the AUDIT have been well established across a wide range of samples and settings when using a cutoff score of 5 to detect dependence (Reinert & Allen, 2007; Rumpf, Hapke, Meyer, & John, 2002). Reliability has been found to range from 0.75 to 0.97, sensitivity from 0.73 to 0.94, and specificity from 0.79 to 0.94 (Berner, Kriston, Bentele, & Härter, 2007; de Meneses-Gaya, Zuardi, Loureiro, & Crippa, 2009; Reinert & Allen, 2007). The DAST-20 consists of 20 dichotomous items designed to capture problematic substance use and has been broadly used in clinical screening and treatment and evaluation research (Skinner & Goldberg, 1986). Internal consistency reliability of the DAST-20 has been good when using a cutoff score of 6 (internal consistency ranging from 0.74 to 0.92, sensitivity of 0.74, specificity of 0.83, and a hit rate of 0.81; Cocco & Carey, 1998; Yudko, Lozhkina, & Fouts, 2007). Due to the long-recognized issue of underidentification and underreporting of SUD needs in families involved in CW (Chuang et al., 2013; Dore, Doris, & Wright, 1995; Schroeder et al., 2008), caseworker reports of SUD needs was also used to better determine each caregiver's risk of SUD problems in this population. In the NSCAW II data, caseworkers reported whether caregivers needed alcohol (yes or no) or drug (yes or no) services. Additionally, preliminary analyses indicated that there were no differences in study outcomes (receipt of SUD assessment, referral to services, or receipt of services) between

Table 1 Caregivers at high risk of substance use disorder problems. Unweighted n Caregiver report of SUD problems Alcohol (AUDIT) Drug (DAST-20) Total Caseworker report of SUD problems Alcohol Drug Total Caregivers at high risk of SUD problems Caregiver receipt of SUD services Received assessment Received referral Received services

Weighted %

356 196 501

9.09 2.66 10.57

371 1220 1374 1651

4.75 12.47 14.46 23.92

989 1125 691

46.76 52.34 72.32

Note. Report of caregiver receipt of SUD services limited to those referred to services. AUDIT = Alcohol Use Disorders Identification Test; DAST = Drug Abuse Screening Test; SUD = substance use disorder.

caregivers who met criteria for AUDIT/DAST versus those identified by caseworkers as needing SUD services. Hence, to better capture need for SUD services, caregivers identified as having a high risk of SUD problems consisted of individuals who met criteria for either alcohol (AUDIT) or drug dependence (DAST-20), and/ or were reported by caseworkers as needing alcohol or drug services. 2.3.2. Caregiver characteristics Caregiver characteristics previously identified to be associated with delivery and receipt of SUD-related services were included as control variables. This included age (categorized as 18–26, 27–33, 34–43, and 44 or older), gender, race and ethnicity (White, Black, Hispanic, and other), and education (less than high school, high school, and more than high school). Family financial hardship was a dichotomous variable measured by asking caregivers if they were “struggling to get by” (yes or no). 2.3.3. Collaboration CW agency directors reported engagement in the following four collaboration strategies: (a) memorandum of understanding (MOU); (b) interdisciplinary/cross-training; (c) co-location of SUD staff members in CW offices; and (d) shared funding or resources. To determine the role that a specific collaboration strategy may have on the outcomes of interest, collaboration strategies were operationalized as four nonmutually exclusive dichotomous variables (yes or no). Detailed information on the conceptualization and measurement of collaboration, especially as it relates to the NSCAW II data, is available elsewhere (blinded for review). 2.3.4. Organizational resources Variables measuring SUD-related organizational resources were: availability of SUD specialists for child abuse and neglect investigations (rarely or sometimes vs. always), availability of SUD assessment tools for CW workers (yes or no), and priority status for CW-involved caregivers with SUD-service agencies (yes or no). Because caseworker caseload issues have been found to influence service delivery (Burton, 2010; Ford et al., 2011; North, Pollio, Perron, Eyrich, & Spitznagel, 2005), a variable indicating if a CW agency experienced an increase in their worker caseload (yes or no) in the past 12 months was also included. All organizational contextual variables were reported by the local CW agency director. 2.3.5. Receipt of SUD services Receipt of services was conceptualized as the following: receipt of SUD assessment, referral made to SUD services, and receipt of SUD treatment services, as reported by caseworkers. Specifically, workers assigned to each case indicated whether: (a) the caregiver received a formal assessment for alcohol and drug problems (yes or no); (b) CW staff members had referred the caregiver to services for alcohol or drug problems (inpatient, detox, intensive day treatment, outpatient, 12-step program, or other; yes or no); and (c) the caregiver received services for alcohol or drug problems as a result of this referral (yes or no). 2.4. Analyses NSCAW data have a hierarchical structure in which children, caregivers, and caseworkers are nested within CW agencies. Preliminary analyses exploring fully unconditional random-effects models found significant variation across CW agencies in caregivers' likelihood of receipt SUD-related services, with an intraclass correlation coefficient of 9.7% for SUD assessment, 8.6% for referral, and 4.5% for treatment services. However, statistical power considerations prevented the use of multilevel analyses to model interorganizational differences in caregivers' odds of receiving SUD-related services. Specifically, the final analytic sample contained a low number of level 2 units (81 CW agencies),

A.S. He / Journal of Substance Abuse Treatment 79 (2017) 20–28

with group membership as low as 11 caregivers per agency (Maas & Hox, 2004; Moineddin, Matheson, & Glazier, 2007; Raudenbush & Bryk, 2002). Therefore, analyses for all models in the study used single-level equations that contained a post hoc adjustment to standard errors and accommodated for the clustering nature of the data (e.g. caregivers within CW agencies) (de Leeuw & Meijer, 2008). Analyses were conducted using the svy module in Stata 13.0, which allowed for analyses that accounted for the data's complex survey design, accommodated probability weights and stratification, and adjusted for the correlation in outcomes across caregivers nested within the same CW agency. Similar analytic strategies have been used elsewhere (Bunger et al., 2012; Chuang et al., 2013). Additionally, client-level weights were used throughout the analyses to correct for the effects of unequal probabilities of selection and provide for unbiased parameter estimation, producing national estimates of CW-involved caregivers in the United States (Biemer, Christ, & Wiesen, 2009). Univariate analyses were conducted to assess the availability of SUD assessment and treatment services resources and determine the distribution of collaborative activities among CW agencies and DAS providers. Descriptive bivariate analyses were conducted to examine the relationship between organizational contextual variables (collaboration activities and SUD-related resources) and service receipt outcomes. Adjusting for weighted nature of the data, bivariate analyses consisting of categorical tests of independence for complex survey designs were used, with results reported as F-statistics using a second-order correction (Rao & Scott, 1984). Weighted logistic regression models were fitted to examine the relationship between organizational contextual variables (collaboration strategies and organization resources) and receipt of SUD-related services (assessment, referral, and treatment services). A value of p b 0.05 or an odds ratio confidence interval (CI) that did not include 1 was viewed to be statistically significant for the final models. Furthermore, correlation analysis was conducted to explore potential multicollinearity issues between organizational contextual independent variables. No unusually high correlations were discovered, with the highest correlation between the individual variables of 0.39. Approval for the current study was obtained from the institutional review board of [blinded for review].

23

had priority status for services at SUD treatment agencies. Additionally, 42% of the CW agencies reported experiencing an increase in caseload. 3.3. Weighted logistic regression Table 3 presents results from the logistic regression analysis that examined the extent to which different collaboration strategies and organizational resource variables were associated with caregivers' odds of receiving SUD services. Accounting for other covariates, caregivers whose agencies engaged in the collaboration strategies of a MOU (OR = 1.75, 95% CI = 1.01, 3.03) or co-location (OR = 2.08, 95% CI = 1.02, 4.25) had significantly higher odds of being referred to SUD services compared to agencies that did not engage in such collaboration activities. Additionally, caregivers whose agencies engaged in the collaboration strategies of shared funding had significantly higher odds of receiving SUD treatment services (OR = 5.15, 95% CI = 1.17, 15.46). On the other hand, engagement in the collaboration activity of co-location was associated with lower odds of caregiver receipt of SUD treatment services (OR = 0.34, 95% CI = 0.15, 0.76). In terms of other organizational resources, caregivers whose agencies reported having the availability of SUD assessment tools had increased odds of receiving a formal SUD assessment (OR = 2.28, 95% CI = 1.40, 3.71). Caregivers whose agencies reported having SUD priority service status for CW-involved families also had significantly higher odds of receiving SUD treatment services (OR = 5.85, 95% CI = 2.16, 15.84). Several control variables were also significantly associated with caregiver receipt of SUD services. Older caregivers had decreased odds of receiving a formal assessment for SUD problems compared to caregivers who were 18 to 26 years old (age group 27–33: OR = 0.36, 95% CI = 0.22, 0.61; age group 34–43: OR = 0.20, 95% CI = 0.10, 0.45). Caregivers experiencing extreme financial hardship had lower odds of being referred to SUD services (OR = 0.49, 95% CI = 0.26, 0.91), and Black caregivers had lower odds receiving SUD treatment services (OR = 0.18, 95% CI = 0.07, 0.45). Female caregivers had significantly higher odds of being referred to SUD services (OR = 3.44, 95% CI = 1.63, 7.29). 4. Discussion

3. Results 3.1. Sample and SUD problems Of the primary caregivers in the NSCAW II study, 11% met the clinical cutoff for a substance use problem and 15% were identified by a CW worker as needing SUD services. This resulted in a study sample of 1651 primary caregivers identified as having a high risk of SUD problems. Of this group, 47% received a formal assessment for an alcohol or drug problem and 52% were referred to services for an alcohol or drug problem; of those referred to services, 72% received alcohol or drug services (see Table 1). 3.2. Descriptives Table 2 presents sample and organizational contextual characteristics and bivariate relationships with SUD-related service variables. The majority of caregivers were female (89%) and the mean age was 36.38. Almost half of the sample was White (non-Hispanic), 19% was Black, and 26% was Hispanic. Most respondents had an education level of high school or beyond (70%) and about 57% reported experiencing financial hardship. The most common collaboration strategy was having a MOU (66%), followed by cross-training (45%), shared funding (23%), and co-location (21%). Almost half of CW agencies reported that SUD specialists were available to accompany CW caseworkers on child abuse and neglect investigations, 38% had SUD assessment tools available to CW workers, and 71% reported that their CW-involved families

In this nationally representative sample of CW-involved caregivers, 1 in 5 of the sample either self-identified or was identified by caseworkers as having a high risk of SUD problems, consistent with prior research various estimates of the prevalence of SUD problems in this population (U.S. Government Accountability Office, 1998; Wulczyn et al., 2011; Young et al., 2007). This high prevalence also emphasizes the need for research on factors that influence the delivery of SUD-related services to this population. Although there is a strong body of literature examining individual characteristics, there is a paucity of research concerning the role of CW organizational factors in service delivery as it relates to various aspects of the SUD treatment process. Hence, this study represents one of the only studies to examine organizational contextual factors of collaboration and resources and their role in the receipt of SUD assessments, referrals, and treatment services for CWinvolved caregivers. Overall, study findings support interorganizational theories of collaboration and resource dependence theories, which posit that engagement in collaboration strategies and greater organizational resources can increase an organization's ability to delivery services to clients. 4.1. Collaboration Extant research indicates that participating in multiple types of collaborative activities is a promising strategy for improving family outcomes (Wells & Chuang, 2012). However, ambiguity regarding whether certain strategies of collaboration may have a greater effect on supporting the delivery of specific treatment services along the

24

A.S. He / Journal of Substance Abuse Treatment 79 (2017) 20–28

Table 2 Characteristics and bivariate analysis of child welfare agencies and caregivers (N = 1651). Assessed

Caregivers Gender Female Male Agea 18–16 27–33 34–43 44+ Race and ethnicity White Black Hispanic Other Education Less than high school High school More than high school Financial hardship Yes No Collaboration strategies Memorandum of understanding Yes No Cross-training Yes No Co-location Yes No Shared funding Yes No Organization resources Availability of SUD specialist Always Sometimes or rarely Availability of SUD assessment tool Yes No Priority status for CW families Yes No Increased caseload Yes No

Referred

Total %

No %

Yes %

0.89 0.11

0.54 0.62

0.46 0.38

0.19 0.29 0.29 0.23

0.38 0.63 0.71 0.42

0.62 0.37 0.29 0.58

0.47 0.19 0.26 0.08

0.61 0.52 0.46 0.49

0.39 0.48 0.54 0.51

0.30 0.43 0.27

0.51 0.51 0.65

0.49 0.50 0.35

0.57 0.43

0.59 0.51

0.41 0.49

0.66 0.34

0.53 0.58

0.47 0.42

0.45 0.55

0.52 0.57

0.48 0.43

0.21 0.79

0.46 0.57

0.54 0.43

0.23 0.77

0.49 0.56

0.51 0.44

0.48 0.52

0.49 0.60

0.51 0.40

0.38 0.62

0.41 0.62

0.59 0.38

0.71 0.29

0.54 0.56

0.46 0.44

0.42 0.58

0.58 0.51

0.42 0.49

F (df)

p

0.71 (1, 71)

0.40

6.32 (2.61, 185.39)

0.83 (2.10, 149.36)

1.70 (1.99, 141.03)

1.97 (1, 71)

0.33 (1, 70)

0.61 (1, 70)

3.09 (1, 70)

1.10 (1, 70)

2.76 (1, 70)

14.95 (1, 69)

0.09 (1, 70)

1.08 (1, 70)

No %

Yes %

0.46 0.73

0.54 0.27

0.44 0.48 0.67 0.36

0.56 0.52 0.33 0.64

0.54 0.39 0.43 0.61

0.46 0.61 0.57 0.39

0.49 0.46 0.56

0.51 0.54 0.44

0.59 0.51

0.41 0.49

0.47 0.54

0.53 0.46

0.45 0.53

0.55 0.47

0.42 0.52

0.58 0.48

0.53 0.48

0.48 0.52

0.44 0.55

0.46 0.45

0.46 0.52

0.54 0.48

0.46 0.56

0.54 0.44

0.54 0.44

0.46 0.56

b0.001

Received F (df)

p

15.72 (1, 71)

b0.001

4.02 (2.64, 187.11)

0.44

1.29 (2.38, 169.03)

0.19

1.13 (1.96, 139.21)

0.16

6.13 (1, 71)

0.57

1.22 (1, 70)

0.44

1.61 (1, 70)

0.08

1.53 (1, 70)

0.30

0.37 (1, 70)

0.10

3.91 (1, 70)

b0.001

0.85 (1, 69)

0.76

2.54 (1, 70)

0.30

2.16 (1, 70)

No %

Yes %

0.28 0.22

0.72 0.78

0.26 0.31 0.40 0.18

0.74 0.69 0.60 0.82

0.20 0.55 0.21 0.26

0.80 0.45 0.79 0.74

0.35 0.15 0.37

0.65 0.85 0.63

0.35 0.23

0.65 0.77

0.28 0.26

0.72 0.74

0.29 0.26

0.71 0.74

0.31 0.26

0.69 0.74

0.21 0.30

0.79 0.70

0.24 0.32

0.76 0.68

0.23 0.31

0.77 0.69

0.24 0.40

0.76 0.60

0.19 0.37

0.81 0.63

0.01

0.28

0.33

0.02

0.27

0.21

0.22

0.54

0.05

0.36

0.12

0.15

F (df)

p

0.44 (1, 70)

0.51

1.40 (2.79, 194.96)

0.25

3.70 (2.76, 193.38)

0.02

3.50 (1.88, 131.58)

0.04

1.98 (1, 70)

0.16

0.08 (1, 70)

0.78

0.05 (1, 70)

0.82

0.36 (1, 70)

0.55

0.29 (1, 70)

0.29

0.51 (1, 70)

0.48

0.63 (1, 69)

0.43

2.24 (1, 70)

0.14

3.38 (1, 70)

0.07

Note. Total column features weighted column percentage; all other percentages are weighted row percentages. Only referred individuals were asked whether they received services. CW = child welfare; SUD = substance use disorder. a M = 36.38, SD = 12.24.

SUD treatment continuum warrants further examination. Findings from this study helped to shed light on this matter, especially as it relates to SUD service delivery outcomes. Indeed, in this study sample, engagement in collaboration through a MOU and co-location supported caregiver receipt of a referral to SUD services. One possible explanation is that among CW agencies that engaged in the collaboration strategy of co-location, the co-located SUD staff members may have supported their CW counterparts by locating agencies that provide needed SUD services, contributing to the subsequent referral of high-risk caregivers to these services. Additionally, because almost 90% of the study sample were women (most of whom were biological mothers) and studies that have shown that mothers involved in the CW system often encounter greater difficulties related to their SUD needs (e.g., accessing and using services; Choi & Ryan, 2006; Greenfield et al., 2007; Grella et al., 2009), co-located staff members may have been particularly helpful in finding and referring these caregivers to SUD agencies that provide

specialized services to women and families. Moreover, the finding that women had greater odds of being referred to SUD services is evidence that co-location of staff members can potentially bolster SUD service delivery to women caregivers involved in the CW system. Co-location of SUD staff members and the support it provided to CW workers may also help explain the null finding between increased caseloads and the likelihood of caregivers being referral to SUD services. Although caseload burden for CW workers has been found to impede service delivery (Burton, 2010; Ford et al., 2011) and despite the fact that 40% of CWagencies in this study reported increases in caseloads, the presence of co-located staff members may have ameliorated effects of caseload burden and supported efforts to provide referrals to SUD services for caregivers at high risk of SUD problems. Additionally, collaboration through sharing of funds was also found to support caregiver receipt of SUD treatment services. Horwath and Morrison's (2007) conceptualization of collaboration, in which they

A.S. He / Journal of Substance Abuse Treatment 79 (2017) 20–28

25

Table 3 Multivariable logistic regression model: collaboration strategies and SUD service receipt. Assessed

Caregivers Female Age 27–33 34–43 44–87 Race and ethnicity Black Hispanic Other Education High school More than high school Financial hardship Collaboration strategies Memorandum of understanding Cross-training Co-location Shared funding Organization resources Availability of SUD specialist Availability of SUD assessment tool Priority status for CW families Increased caseload

Referred

Received

OR

95% CI

OR

95% CI

OR

95% CI

1.46

0.68, 3.12

3.44⁎⁎⁎

1.63, 7.29

1.35

0.40, 4.62

0.36⁎⁎⁎ 0.20⁎⁎⁎ 0.83

0.22, 0.61 0.10, 0.45 0.40, 1.73

0.91 0.46 1.87

0.50, 1.65 0.23, 0.91 0.82, 4.27

1.80 0.49 1.46

0.83, 3.93 0.20, 1.21 0.63, 3.39

1.26 1.63 1.63

0.64, 2.48 0.84, 3.18 0.51, 5.23

1.61 1.77 0.69

0.79, 3.28 0.84, 3.74 0.21, 2.26

0.18⁎⁎⁎ 0.86 1.19

0.07, 0.45 0.34, 2.14 0.59, 2.41

1.07 0.79 0.81

0.55, 2.09 0.35, 1.79 0.51, 1.29

1.00 0.74 0.49⁎

0.62, 1.63 0.35, 1.56 0.26, 0.91

2.39⁎ 0.80 0.58

1.09, 5.19 0.38, 1.68 0.27, 1.28

1.24 1.06 1.36 1.10

0.69, 2.27 0.61, 1.87 0.79, 2.33 0.61, 2.02

1.75⁎ 1.05 2.08⁎ 0.59

1.01, 3.03 0.60, 1.84 1.02, 4.25 0.31, 1.12

0.36 0.37 0.34⁎⁎ 5.15⁎⁎⁎

0.13, 1.02 0.14, 1.00 0.15, 0.76 1.71, 15.46

1.21 2.28⁎⁎⁎ 0.71 0.68

0.70, 2.11 1.40, 3.71 0.38, 1.30 0.44, 1.06

1.31 0.89 1.00 0.62

0.79, 2.20 0.49, 1.62 0.56, 1.79 0.37, 1.05

1.24 1.80 5.85⁎⁎⁎ 2.11

0.52, 2.96 0.76, 4.23 2.16, 15.84 0.21, 8.38

Note. Referent categories were 18–26 for age, White for race and ethnicity, and less than high school for education. CW = child welfare; SUD = substance use disorder. ⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.

posit that collaborative activities fall along a collaboration continuum of low to high levels, helps provide insight on this matter. They propose that whereas information sharing is considered to be a low level of collaboration, joint budgeting or sharing of funds is viewed as falling on the higher end of the collaboration continuum. Based on this framework, CW agencies in this study collaborating through shared funding activities with DAS providers may have been engaging in higher and stronger levels of collaboration, which in turn aided the delivery of SUD services not just at the referral stage but through the receipt of SUD treatment services. Research has indicated that the step between referral and entry to and receipt of treatment services is the largest gap in the SUD treatment process (Substance Abuse and Mental Health Services Administration, 2009; Traube et al., 2015). Thus, greater levels of collaboration, as represented by shared funding resources, may address the interruption between referral to and receipt of treatment services. On the other hand, engagement in collaboration activities in this study was not significantly associated with caregiver receipt of a formal assessment for SUD problems. One possible explanation is that among caregivers being investigated for issues of child abuse and neglect, there may already be very clear evidence of serious SUD problems (e.g., known history of SUD problems or maltreatment allegation due to SUD problems). Rather than sending these particular caregivers to be assessed for SUD problems, CW caseworkers may have bypassed the SUD assessment process and referred them directly to SUD treatment services. Finally, given that co-location of staff members supported receipt of SUD service referral in another model, the negative association between co-location and receipt of SUD services was unexpected and warrants careful consideration. One reason may be that because these caregivers were already referred to SUD services and given their extreme work demands, CW workers may at times be overly reliant on their co-located SUD counterparts to procure treatment services for caregivers. Moreover, CW workers may not have made more intentional efforts to get caregivers into treatment, instead entrusting this responsibility to co-located SUD staff members. This suggests that when engaging in

collaboration activities such as co-location of staff members, consideration of and guidelines for the delineation of staff roles and responsibilities, especially as it relates to follow-up of services referrals, may be crucial for supporting the receipt of SUD treatment services by CW-involved caregivers. 4.2. Organizational resources In this study, caregivers were more likely to receive a formal assessment for SUD problems when their CW agencies reported the availability of a standardized SUD assessment tool. Providing a SUD assessment tool is particularly important for CW workers because they are not SUD specialists. Access to a SUD assessment tool could have therefore provided guidance to CW workers in gathering relevant information regarding indicators of SUD problems and need for services. The availability of a SUD assessment tool may have increased the ability of CW workers to flag potential SUD problems among caregivers and contributed to subsequent case planning to ensure that these caregivers received a formal assessment for SUD problems. Previous research has called for expanded understanding of when the use of a standardized SUD assessment tool may be most useful as it relates to outcomes for CW-involved families (Chuang et al., 2013). This study finding suggests that availability of a standardized SUD assessment tool in CW agencies may support the delivery of services at the front end of the SUD treatment continuum, namely the receipt of a formal SUD assessment. Indeed, receipt of a formal assessment could expedite delivery of SUDrelated services and aid appropriate clinical decision making and case planning for CW-involved families. Study findings also indicate that having arrangements with SUD agencies so that CW-involved families had priority status to enter treatment was pertinent to caregiver receipt of SUD treatment services. As previously discussed, entry into and completion of SUD treatment is crucial for CW-involved families that are working against certain time limitations that act as barriers to positive outcomes (Young et al., 2007). For example, the long-term nature of SUD treatment, incidence

26

A.S. He / Journal of Substance Abuse Treatment 79 (2017) 20–28

of relapse, and ASFA's 12-month time frame to meet permanency goals increase the risk of losing parental rights (Green, Furrer, et al., 2007). Study findings suggest that for CW-involved families, having priority status for SUD treatment may be a viable option to increase the odds that caregivers will receive SUD treatment services. Indeed, more than two thirds of CW agencies reported having a priority status arrangement with DAS providers, indicating that this may be a growing trend in supporting service delivery to families dealing with issues of child abuse and neglect and SUD problems. 4.3. Caregiver characteristics Finally, accounting for organizational contextual factors, study results reflect the ongoing need to improve the delivery of SUD services to minority groups (such as Blacks) and families dealing extreme financial needs in the CW population. These findings align with existing studies that found that minority caregivers were less likely to complete treatment than White caregivers (Choi & Ryan, 2006) and that issues of poverty greatly influence receipt of SUD treatment services (Grella et al., 2006; Small, 2015). 4.4. Limitations Results should be considered relative to certain study limitations. First, although the study examined engagement in collaboration strategies and availability of organizational resources, it did not assess the quality of the collaboration or use of resources (e.g., SUD assessment tools) by CW caseworkers. In-depth assessment of engagement in other collaboration strategies (e.g., data sharing) and use or knowledge of SUD-related resources among CW caseworkers would have furthered the understanding of the role of organizational contextual factors in the delivery of SUD-related services. Second, because this study only used baseline data to measure receipt of SUD services, it was not able to determine the delivery of these services at a later point; this is especially relevant to capturing service delivery to caregivers receiving ongoing services from their CW welfare agency. Third, this study relied on caregiver and caseworker report of SUD problems, both of which pose unique challenges to the identification of SUD needs in the CW population. Even though caregivers used validated instruments (AUDIT and DAST-20) to report substance use problems and were assured confidentiality, issues of social desirability and fear of repercussions related to their CW case may have resulted in underreporting of SUD needs (Yudko et al., 2007). CW workers who are not specialized SUD service providers to properly identify SUD needs may have also contributed to a lack of accuracy on the true prevalence of SUD needs in this population (Hohman, Finnegan, & Clapp, 2008; Hughes & Rycus, 2006). Given the study's inclusion of both caregiver and caseworker reports of SUD needs, this was not viewed as a major limitation. Finally, this study did not account for additional caregiver characteristics such as engagement, e.g., receptivity to services and behaviors such as not keeping appointments (Ferguson, 2009; Yatchmenoff, 2005), or mental health status, both of which have been found to challenge service delivery (Altman, 2008; Grella et al., 2006; Grella et al., 2009). However, regarding mental health status, post hoc analyses indicated that adding a measure of caregiver depression did not significantly affect the models in this study. 5. Implications Despite these limitations, current study results present implications for research, practice, and policy. Future research should more fully explore the relationship between organizational contextualization factors and caregiver receipt of SUD services. For example, further research is needed to examine why engagement in certain collaboration strategies or the availability of certain organizational resources have a greater

effect on caregiver receipt of SUD services than other factors. Moreover, given results relating to co-location of staff members, additional work is needed to examine the role of this collaboration strategy (e.g., relationship and interaction between co-located SUD staff members and CW workers) in service delivery to caregivers dually served by CW and DAS providers. Future research is also warranted to explore practice and policies that shore up the development of the organizational resource of having priority SUD treatment status for CW-involved families, especially as this was found to have significant positive impact on caregiver's receipt of SUD treatment services. Previous studies have also found that treatment completion is one of the most challenging steps along the SUD service treatment continuum (Belenko et al., 2017; Choi & Ryan, 2006; Traube et al., 2015). For CW-involved families dealing with SUD problems, successful SUD treatment completion is an integral part of family reunification requirements. Additionally, families in the CW system face many stressors (not necessarily found in the general population) that impact treatment completion, including a greater representation of women, issues of domestic violence, mental health needs, and unemployment (Dolan, Smith, Casanueva, & Ringeisen, 2011; Grella et al., 2009). Future studies should therefore explore how organizational contextual factors might influence long-term outcomes for CW-involved families, including SUD treatment completion and family reunification or recidivism. Given the promising role of collaboration in supporting the delivery of SUD services, CW agencies should consider engaging in and developing other collaborative strategies that increase service delivery capacity. This includes developing joint committees and task groups to ensure that caregivers receive uninterrupted SUD services, from assessment to referral to receipt to completion of treatment services. Task groups could also explore the implementation of evaluation frameworks that help to identify potential unmet needs during key points of transition within the SUD treatment continuum (e.g. from referral to entry into treatment) (Belenko et al., 2017). Moreover, ensuring the proper training and use of SUD assessment tools by CW caseworkers may help in the identification of SUD needs, increasing the likelihood that these caregivers will receive timely and appropriate services. Additionally, there has been a recent call from federal agencies to reexamine the effect of high caseloads on service delivery and work burnout in the CW field. Engagement in and development of collaboration strategies directed toward reducing service burden for CW workers may help to address issues of burnout and effective service delivery even in the face of CW workers struggling with the demands of high caseloads. Finally, given the high prevalence of SUD needs in the CW population, policies that promote and support collaboration between CW agencies and DAS service providers are needed to increase and strengthen collaboration between these two entities.

References Aarons, G. A., Hurlburt, M., & Horwitz, S. M. (2011). Advancing a conceptual model of evidence-based practice implementation in public service sectors. Administration and Policy in Mental Health and Mental Health Services Research, 38, 4–23. http://dx.doi. org/10.1007/s10488-010-0327-7. Adoption and Safe Families Act of 1997 (1997). Pub. L. No. 105-89. 42. U.S.C. 1305. Alter, C., & Hage, J. (1993). Organizations working together. Newbury Park, CA: Sage. Altman, J. C. (2008). Engaging families in child welfare services: Worker versus client perspectives. Child Welfare, 87, 31–63. Andersen, R. M. (1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36, 1–10. http://dx.doi.org/10.2307/ 2137284. Arria, A. M., & Thoreson, A. (2007). Integration of child welfare and drug treatment services in Baltimore city and Prince George's County: An evaluation of the implementation of Maryland's House Bill 7. Baltimore, MD: Alcohol and Drug Abuse Administration. Babor, T. F., Higgins-Biddle, J. C., Saunders, J. B., & Monteiro, M. G. (2001). The Alcohol Use Disorders Identification Test (AUDIT): Guidelines for use in primary care (2nd ed.). Geneva, Switzerland: World Health Organization. Barth, R. P., Gibbons, C., & Guo, S. (2006). Substance abuse treatment and the recurrence of maltreatment among caregivers with children living at home: A propensity score analysis. Journal of Substance Abuse Treatment, 30, 93–104. http://dx.doi.org/10. 1016/j.jsat.2005.10.008.

A.S. He / Journal of Substance Abuse Treatment 79 (2017) 20–28 Berner, M. M., Kriston, L., Bentele, M., & Härter, M. (2007). The Alcohol Use Disorders Identification Test for detecting at-risk drinking: A systematic review and meta-analysis. Journal of Studies on Alcohol and Drugs, 68, 461–473. http://dx.doi.org/10.15288/jsad. 2007.68.461. Biemer, P. P., Christ, S. L., & Wiesen, C. A. (2009). A general approach for estimating scale score reliability for panel survey data. Psychological Methods, 14, 400–412. http://dx. doi.org/10.1037/a0016618. Belenko, S., Knight, D., Wasserman, G. A., Dennis, M. L., Wiley, T., Taxman, F. S., ... Sales, J. (2017). The Juvenile Justice Behavioral Health Services Cascade: A new framework for measuring unmet substance use treatment services needs among adolescent offenders. Journal of Substance Abuse Treatment, 74, 80–91. Brady, T. M., & Ashley, O. S. (2005). Women in substance abuse treatment: Results from the Alcohol and Drug Services Study (ADSS). Washington, DC: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Brook, J., & McDonald, T. P. (2007). Evaluating the effects of comprehensive substance abuse intervention on successful reunification. Research on Social Work Practice, 17, 664–673. http://dx.doi.org/10.1177/1049731507300148. Brucker, D. (2010). Exploring the relationship between access and retention among substance abuse treatment admissions. Journal of Drug Issues, 40, 553–576. http://dx.doi. org/10.1177/002204261004000302. Bunger, A. C., Chuang, E., & McBeath, B. (2012). Facilitating mental health service use for caregivers: Referral strategies among child welfare caseworkers. Children and Youth Services Review, 34, 696–703. http://dx.doi.org/10.1016/j.childyouth.2011.12.014. Burton, J. (2010). WHO healthy workplace framework and model: Background and supporting literature and practice. Geneva, Switzerland: World Health Organization. Choi, S., & Ryan, J. P. (2006). Completing substance abuse treatment in child welfare: The role of co-occurring problems and primary drug of choice. Child Maltreatment, 11, 313–325. http://dx.doi.org/10.1177/1077559506292607. Chuang, E., Wells, R., Bellettiere, J., & Cross, T. P. (2013). Identifying the substance abuse treatment needs of caregivers involved with child welfare. Journal of Substance Abuse Treatment, 45, 118–125. http://dx.doi.org/10.1016/j.jsat.2013.01.007. Cocco, K. M., & Carey, K. B. (1998). Psychometric properties of the Drug Abuse Screening Test in psychiatric outpatients. Psychological Assessment, 10, 408–414. http://dx.doi. org/10.1037/1040-3590.10.4.408. de Leeuw, J., & Meijer, E. (Eds.). (2008). Handbook of multilevel analysis. New York, NY: Springer. de Meneses-Gaya, C., Zuardi, A. W., Loureiro, S. R., & Crippa, J. A. S. (2009). Alcohol Use Disorders Identification Test (AUDIT): An updated systematic review of psychometric properties. Psychology & Neuroscience, 2, 83–97. http://dx.doi.org/10.3922/j.psns.2009. 1.12. Dolan, M., Smith, K., Casanueva, C., & Ringeisen, H. (2011). NSCAW II baseline report: Introduction to NSCAW II: Final report (OPRE Report No. 2011-27a). Washington, DC: U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research and Evaluation. Dore, M. M., Doris, J. M., & Wright, P. (1995). Identifying substance abuse in maltreating families: A child welfare challenge. Child Abuse & Neglect, 19, 531–543. http://dx. doi.org/10.1016/0145-2134(95)00013-X. Dowd, K., Dolan, M., Wallin, J., Miller, K. A., Biemer, P., Aragon-Logan, E., et al. (2010). National Survey of child and adolescent well-being II: Data file user's manual restricted release version. Ithaca, New York: National Data Archive on Child Abuse and Neglect. Dowd, K., Dolan, M., Wallin, J., Miller, K., Biemer, J., Aragon-Logan, E., & Smith, K. (2012). National survey of child and adolescent well-being (NSACW): NSCAW II combined waves 1–2 data file user's manual, restricted release version. Ithaca, NY: Cornell University, National Data Archive on Child Abuse and Neglect. Dyer, J. H., & Singh, H. (1998). The relational view: Cooperative strategy and sources of interorganizational competitive advantage. Academy of Management Review, 23, 660–679. http://dx.doi.org/10.5465/AMR.1998.1255632. Feit, M. D., Fisher, C., Cummings, J., & Peery, A. (2015). Substance use and abuse: Screening tools and assessment instruments. In J. S. Wodarski, M. J. Holosko, & M. D. Feit (Eds.), Evidence-informed assessment and practice in child welfare (pp. 123–133). Cham, Switzerland: Springer. Ferguson, H. (2009). Performing child protection: Home visiting, movement and the struggle to reach the abused child. Child & Family Social Work, 14, 471–480. http:// dx.doi.org/10.1111/j.1365-2206.2009.00630.x. Ford, M. T., Cerasoli, C. P., Higgins, J. A., & Decesare, A. L. (2011). Relationships between psychological, physical, and behavioural health and work performance: A review and meta-analysis. Work and Stress, 25, 185–204. http://dx.doi.org/10.1080/ 02678373.2011.609035. Green, B. L., Furrer, C., Worcel, S., Burrus, S., & Finigan, M. W. (2007a). How effective are family treatment drug courts? Outcomes from a four-site national study. Child Maltreatment, 12, 43–59. http://dx.doi.org/10.1177/1077559506296317. Green, B. L., Rockhill, A., & Furrer, C. (2007b). Does substance abuse treatment make a difference for child welfare case outcomes? A statewide longitudinal analysis. Children and Youth Services Review, 29, 460–473. http://dx.doi.org/10.1016/j.childyouth.2006.08.006. Greenfield, S. F., Brooks, A. J., Gordon, S. M., Green, C. A., Kropp, F., McHugh, R. K., ... Miele, G. M. (2007). Substance abuse treatment entry, retention, and outcome in women: A review of the literature. Drug and Alcohol Dependence, 86, 1–21. http://dx.doi.org/10. 1016/j.drugalcdep.2006.05.012. Grekin, E. R., Svikis, D. S., Lam, P., Connors, V., LeBreton, J., Streiner, D., ... Ondersma, S. J. (2010). Drug use during pregnancy: Validating the Drug Abuse Screening Test against physiological measures. Psychology of Addictive Behaviors, 24, 719–723. http://dx.doi. org/10.1037/a0021741. Grella, C. E., Hser, Y. -I., & Huang, Y. C. (2006). Mothers in substance abuse treatment: Differences in characteristics based on involvement with child welfare services. Child Abuse & Neglect, 30, 55–73. http://dx.doi.org/10.1016/j.chiabu.2005.07.005.

27

Grella, C. E., Needell, B., Shi, Y., & Hser, Y. -I. (2009). Do drug treatment services predict reunification outcomes of mothers and their children in child welfare? Journal of Substance Abuse Treatment, 36, 278–293. http://dx.doi.org/10.1016/j. jsat.2008.06.010. He, A. S. (2015). Examining intensity and types of interagency collaboration between child welfare and drug and alcohol service providers. Child Abuse & Neglect, 46, 190–197. http://dx.doi.org/10.1016/j.chiabu.2015.07.004. Hohman, M., Finnegan, D., & Clapp, J. D. (2008). A concurrent validation study of the Alcohol and Other Drug Identification (AODI) scale. Journal of Social Work Practice in the Addictions, 8, 367–379. http://dx.doi.org/10.1080/15332560802224527. Horwath, J., & Morrison, T. (2007). Collaboration, integration and change in children's services: Critical issues and key ingredients. Child Abuse & Neglect, 31, 55–69. http://dx. doi.org/10.1016/j.chiabu.2006.01.007. Hughes, R. C., & Rycus, J. S. (2006). Issues in risk assessment in child protective services. Journal of Public Child Welfare, 1, 85–116. http://dx.doi.org/10.1300/J479v01n01_05. Kimerling, R., & Baumrind, N. (2005). Access to specialty mental health services among women in California. Psychiatric Services, 56, 729–734. http://dx.doi.org/10.1176/ appi.ps.56.6.729. Maas, C. J. M., & Hox, J. J. (2004). Robustness issues in multilevel regression analysis. Statistica Neerlandica, 58, 127–137. http://dx.doi.org/10.1046/j.0039-0402.2003. 00252.x. Marsh, J. C., Smith, B. D., & Bruni, M. (2011). Integrated substance abuse and child welfare services for women: A progress review. Children and Youth Services Review, 33, 466–472. http://dx.doi.org/10.1016/j.childyouth.2010.06.017. Moineddin, R., Matheson, F. I., & Glazier, R. H. (2007). A simulation study of sample size for multilevel logistic regression models. BMC Medical Research Methodology, 7, 34. http://dx.doi.org/10.1186/1471-2288-7-34. North, C. S., Pollio, D. E., Perron, B., Eyrich, K. M., & Spitznagel, E. L. (2005). The role of organizational characteristics in determining patterns of utilization of services for substance abuse, mental health, and shelter by homeless people. Journal of Drug Issues, 35, 575–591. http://dx.doi.org/10.1177/002204260503500309. Pfeffer, J., & Salancik, G. R. (1978). The external control of organizations: A resource dependence perspective. New York, NY: Harper & Row. Postmus, J., & Hahn, S. A. (2007). The collaboration between welfare and advocacy organizations: Learning from the experiences of domestic violence survivors. Families in Society, 88, 475–484. http://dx.doi.org/10.1606/1044-3894.3658. Rao, J. N. K., & Scott, A. J. (1984). On chi-squared tests for multiway contingency tables with cell proportions estimated from survey data. Annals of Statistics, 12, 46–60. http://dx.doi. org/10.1214/aos/1176346391. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods. Thousand Oaks, CA: Sage. Redko, C., Rapp, R. C., & Carlson, R. G. (2006). Waiting time as a barrier to treatment entry: Perceptions of substance users. Journal of Drug Issues, 36, 831–852. http://dx.doi.org/ 10.1177/002204260603600404. Reinert, D. F., & Allen, J. P. (2007). The Alcohol Use Disorders Identification Test: An update of research findings. Alcoholism: Clinical and Experimental Research, 31, 185–199. http://dx.doi.org/10.1111/j.1530-0277.2006.00295.x. Rumpf, H. -J., Hapke, U., Meyer, C., & John, U. (2002). Screening for alcohol use disorders and at-risk drinking in the general population: Psychometric performance of three questionnaires. Alcohol and Alcoholism, 37, 261–268. http://dx.doi.org/10.1093/ alcalc/37.3.261. Ryan, J. P., Marsh, J. C., Testa, M. F., & Louderman, R. (2006). Integrating substance abuse treatment and child welfare services: Findings from the Illinois alcohol and other drug abuse waiver demonstration. Social Work Research, 30(2), 95–107. Ryan, J. P., Choi, S., Hong, J. S., Hernandez, P., & Larrison, C. R. (2008). Recovery coaches and substance exposed births: An experiment in child welfare. Child Abuse & Neglect, 32, 1072–1079. http://dx.doi.org/10.1016/j.chiabu.2007.12.011. Schroeder, J., Lemieux, C., & Pogue, R. (2008). The collision of the Adoption and Safe Families Act and substance abuse: Research-based education and training priorities for child welfare professionals. Journal of Teaching in Social Work, 28, 227–246. http:// dx.doi.org/10.1080/08841230802179316. Skinner, H. A., & Goldberg, A. E. (1986). Evidence for a drug dependence syndrome among narcotic users. British Journal of Addiction, 81, 479–484. http://dx.doi.org/10.1111/j. 1360-0443.1986.tb00359.x. Small, J., Curran, G. M., & Booth, B. (2010). Barriers and facilitators for alcohol treatment for women: Are there more or less for rural women? Journal of Substance Abuse Treatment, 39, 1–13. http://dx.doi.org/10.1016/j.jsat.2010.03.002. Small, L. F. F. (2015). Co-morbidities among persons with substance abuse problems: Factors influencing the receipt of treatment. Journal of Drug Issues Advance online publication 10.1177/0022042615619642. Stein, J. A., Andersen, R., & Gelberg, L. (2007). Applying the Gelberg-Andersen behavioral model for vulnerable populations to health services utilization in homeless women. Journal of Health Psychology, 12, 791–804. http://dx.doi.org/10.1177/ 1359105307080612. Substance Abuse and Mental Health Services Administration (2009). Treatment Episode Data Set (TEDS): 2009 discharges from substance abuse treatment services (DASIS Series S-46, DHHS Publication No. SMA 09-4378). Rockville, MD: Author. Traube, D. E., He, A. S., Zhu, L., Scalise, C., & Richardson, T. (2015). Predictors of substance abuse assessment and treatment completion for parents involved with child welfare: One state's experience in matching across systems. Child Welfare, 94, 45–66. U.S. Department of Health and Human Services (1999). Blending perspectives and building common ground: A report to congress on substance abuse and child protection. Washington, DC: U.S. Government Printing Office. U.S. Government Accountability Office (1998). Foster care: Agencies face challenges securing stable homes for children of substance abusers (HEHS-98-182). Washington, DC: U.S. Government Printing Office.

28

A.S. He / Journal of Substance Abuse Treatment 79 (2017) 20–28

Vanderploeg, J. J., Connell, C. M., Caron, C., Saunders, L., Katz, K. H., & Tebes, J. K. (2007). The impact of parental alcohol or drug removals on foster care placement experiences: A matched comparison group study. Child Maltreatment, 12, 125–136. http:// dx.doi.org/10.1177/1077559507299292. Wells, R., & Chuang, E. (2012). Does formal integration between child welfare and behavioral health agencies result in improved placement stability for adolescents engaged with both systems? Child Welfare, 91, 79–100. Wulczyn, F., Ernst, M., & Fisher, P. (2011). Who are the infants in out-of-home care? An epidemiological and developmental snapshot. Chicago, IL: Chapin Hall at the University of Chicago. Xu, J., Rapp, R. C., Wang, J., & Carlson, R. G. (2008). The multidimensional structure of external barriers to substance abuse treatment and its invariance across

gender, ethnicity, and age. Substance Abuse, 29, 43–54. http://dx.doi.org/10.1300/ J465v29n01_06. Yatchmenoff, D. K. (2005). Measuring client engagement from the client's perspective in nonvoluntary child protective services. Research on Social Work Practice, 15, 84–96. http://dx.doi.org/10.1177/1049731504271605. Young, N. K., Boles, S. M., & Otero, C. (2007). Parental substance use disorders and child maltreatment: Overlap, gaps, and opportunities. Child Maltreatment, 12, 137–149. http://dx.doi.org/10.1177/1077559507300322. Yudko, E., Lozhkina, O., & Fouts, A. (2007). A comprehensive review of the psychometric properties of the Drug Abuse Screening Test. Journal of Substance Abuse Treatment, 32, 189–198. http://dx.doi.org/10.1016/j.jsat.2006.08.002.