Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 1.
Published in final edited form as: Psychol Addict Behav. 2022 May 5;37(2):199–208. doi: 10.1037/adb0000833

Measurement Invariance and Other Psychometric Properties of the Short Inventory of Problems (SIP-2R) Across Racial Groups in Adults Experiencing Homelessness and Alcohol Use Disorder

Silvi C Goldstein 1, Nichea S Spillane 1, Marie Tate 1, Lonnie Nelson 2, Susan E Collins 3,4
PMCID: PMC9636066  NIHMSID: NIHMS1791590  PMID: 35511527

Abstract

Objective:

People experiencing homelessness are disproportionately impacted by alcohol-related harm. Racially minoritized groups are disproportionately represented in the homeless population and are likewise disproportionately impacted by alcohol-related harm. Most alcohol outcome measures have not been adequately psychometrically studied in this marginalized population and across racial groups. This study documents psychometric properties, including measurement invariance, reliability, and convergent validity, of a measure of alcohol-related harm, the Short Inventory of Problems (SIP-2R), across Black, North American Indigenous (NAI), and White adults experiencing homelessness and alcohol use disorder (AUD).

Methods:

Adults experiencing homelessness and AUD who had participated in one of two randomized controlled trials of harm reduction treatment (N = 493; NAI = 205, Black = 125, and White = 163) were included in this psychometric study of the 15-item SIP-2R.

Results:

Multigroup confirmatory factor analysis (MGCFA) indicated that a model comprising one general alcohol-related harm factor overarching five factors, showed close fit and partial scalar invariance, χ2 (329, N = 493) = 624.902, p < .001, CFI = .966, RMSEA = .074, 90% CI [.066, .083], SRMR = .063, confirming acceptable measurement equivalence across racial groups. The SIP-2R showed internal consistency (α = .94, ω =.95) and convergent validity (i.e., positive correlation between the total SIP-2R score and the number of drinks consumed the heaviest drinking day, ρ (490)= .30, p < .001).

Conclusion:

This study provided support for the internal consistency, convergent validity, and cross-group measurement equivalence of the SIP-2R for NAI, Black, and White adults experiencing homelessness with AUD.

Keywords: North American Indigenous, Black/African American, Measurement invariance, Alcohol-related harm, Validity, Reliability

Introduction

People experiencing homelessness are disproportionately affected by medical, psychiatric, and substance use disorders (Fazel et al., 2014; Greenberg & Rosenheck, 2010; Krausz et al., 2013), among which alcohol use disorder (AUD) is one of the most prevalent and physically debilitating. A US population-based study indicated that about 40% of people experiencing homelessness meet criteria for AUD, which is 10 times more prevalent among those experiencing homelessness than in the general population (Fazel et al., 2008; Grant et al., 2004; North et al., 2010). Indeed, people experiencing homelessness are 6 to 10 times more likely to die of alcohol-attributable causes than the general population (Baggett et al., 2015; Hwang et al., 2009).

North American Indigenous (NAI) and Black individuals are overrepresented among those experiencing homelessness in urban settings (Culhane et al., 2019; Home, 2019; Tsai, 2018; Whitbeck et al., 2012), and NAI and Black men who have experienced homelessness are significantly more likely to meet lifetime criteria for AUD than those who have never been homeless (Bird et al., 2002; Whitbeck et al., 2012). Although the prevalence of alcohol use is lower among NAI and Black adults than among White adults in the general US population (SAMHSA, 2018; SAMHSA, 2019a), they experience a disproportionate level of alcohol-related harm (Collins, 2016; 2019; Zapolski et al., 2014). For example, NAI individuals’ experience of alcohol-related morbidity and mortality is 6.6 times higher than in the general population (Indian Health Services, 2019; Landen et al., 2014), with alcohol use reported as the largest contributing factor to increased mortality (Welty, 2003). Black adults who drink are at greater risk for alcohol-related harm including alcohol-related health sequelae, accidents, legal trouble, and social consequences compared to White adults (Mulia et al., 2009; Stahre & Simon, 2010; Wilson et al., 2014; Witbrodt et al., 2014). Researchers argue that unique stressors experienced by NAI and Black individuals living in the U.S. – including historical trauma, institutionalized racism, and discrimination – precipitate alcohol-related health and social inequities among these groups (Brave Heart et al, 2011; Gibbons et al., 2004; Gibbons et al., 2012; Gibbons et al., 2007; SAMHSA, 2019b; Whitesell et al., 2012; Wiechelt, 2012). Therefore, given that NAI and Black individuals are disproportionally impacted by alcohol-related harm, are overrepresented among those experiencing homelessness, and are significantly more likely to meet for AUD, it is important that measurement tools utilized to assess alcohol-related harm among this population that offer clinical and research utility are psychometrically valid across groups and allow for sound across group comparisons in order to assess alcohol-related health inequities.

Given the stark alcohol-related health and social inequities that exist across racial groups experiencing homelessness, it is of vital importance to ensure our measurement tools are reliable and valid and that that they measure the same constructs across different population subgroups. Although some commonly utilized alcohol measures have been validated for use in marginalized populations (Dillon et al., 2015; Goldstein et al., in press; Hirchak et al., 2021; Marra et al., 2014; Noel et al., 2010), many frequently used measurement tools have not been created nor psychometrically validated for use with individuals from racially diverse backgrounds. Importantly, measurement tools that are utilized to establish health inequities need to be empirically reliable, valid, and equivalent across diverse groups (Walls et al., 2019). Therefore, psychometric evaluation of measures of alcohol-related harm is warranted to ensure accurate documentation of individuals from racially diverse backgrounds in descriptive and correlational studies, as well as in measuring treatment efficacy and effectiveness.

While psychometric tests of overall reliability and validity are important, testing measurement invariance is an additional psychometric methodology that can be used to establish whether constructs are measured equivalently across groups (e.g., race). Testing measurement invariance allows for researchers to validate measures across diverse groups and to conduct substantive cross-group comparisons (Burlew et al., 2009; Meredith, 1993; Vandenberg & Lance, 2000; Widaman & Reise, 1997).

Derived from the 45-item DrInC (Miller et al., 1995), the Short Inventory of Problems (SIP-2R) is a commonly utilized 15-item, Likert-scale questionnaire that measures the extent of one’s experience with alcohol-related harm due to social, interpersonal, intrapersonal, physical, and impulse control issues. As an important distinction, other versions of the SIP (e.g., SIP-AD) — often derived from the related InDUC-2R (Miller et al 1995) — frame items to measure harm due to both alcohol and other drugs and sometimes include overlapping but different items from the SIP-2R (e.g., Allensworth‐Davies et al., 2012; Bender et al., 2007; Blanchard et al., 2003; Forcehimes et al., 2007; Gillespie et al., 2007; Hagman et al., 2009; Kiluk et al, 2013; Morse & Robertson, 2017; Tonigan & Miller, 2002)

Studies of the SIP-2R, have generally supported its reliability and validity (Feinn et al., 2003; Hirchak et al., 2021; Kenna et al., 2005; Kirouac & Witkiewitz, 2018; Marra et al, 2014). Psychometric research regarding factor structure, however, has been mixed (Feinn et al 2003; Kenna et al 2005; Kirouac & Witkiewitz, 2018; Marra et al 2014). Some confirmatory analyses have indicated support for a single-factor model represented a general alcohol-related harm construct (Alterman et al., 2009; Marra et al., 2014), whereas other studies have supported three- and five-factor models, representing more specific areas of alcohol-related harm (e.g., Kirouac & Witkiewitz 2018; Kenna et al., 2005)

Additionally, few studies to date have documented the psychometric properties of the SIP-2R in minoritized and marginalized populations who are disproportionately impacted by alcohol-related harm. Among this handful of studies, Hirchak and colleagues (2021) found the 15-item SIP-2R (Miller et al., 1995) to have internal consistency and convergent validity among urban American Indian adults who attended Alcoholics Anonymous. However, this study did not entail information about factor structure or measurement invariance across racial groups. Marra, Field, Caetano, and von Sternberg (2014) tested the factor structure and measurement invariance of the SIP-2R (Miller et al., 1995) among White, non-Spanish-speaking Latinx and Spanish-speaking Latinx patients who used alcohol. The authors tested the comparison of model fit between three competing SIP-2R models: first-order one-factor, first-order five-factor, and second-order five-factor (Marra et al., 2014). After finding strongest support for a first-order one-factor model for the SIP-2R, the authors found measurement invariance for the SIP-2R across Spanish and English speakers (Marra et al., 2014).

Although a prior study tested metric invariance of the SIP-AD across racial groups (Dillon et al., 2015), no study to date has assessed the measurement invariance of the SIP-2R across racial groups or among adults experiencing homelessness with AUD. Given the stark alcohol-related, racial health inequities that disproportionately impact people experiencing homelessness, it is important to expand this work further and test the reliability, validity, and measurement invariance of the SIP-2R in Black, NAI, and White people experiencing homelessness and AUD.

Building on this work, the current study aims to 1) investigate the factor structure using the three competing SIP-2R models for a community sample of NAI, Black, and White adults experiencing homelessness and AUD, 2) determine whether the SIP-2R fulfills criteria for measurement invariance, or measurement equivalence, across three racial groups disproportionately represented in the homeless population: NAI, Black, and White participants, and 3) assess the reliability and convergent validity of the SIP-2R.

Methods

Participants

Participants were drawn from the baselines of two larger randomized controlled trials of harm reduction treatment across six community-based settings in Seattle, WA, USA (Collins et al., 2019; Collins et al., 2021). The two studies were funded and implemented during an overlapping time period in the same city, drawing from the same population of people experiencing homelessness and AUD. The present, secondary study included 493 adults self-identifying as NAI, Black, and White (see Table 1 for sample description). Inclusion criteria for this secondary study included a) being at least 21 years of age, b) experiencing homelessness in the past 12 months, and c) meeting criteria for an AUD (i.e., DSM-IV-TR criteria for “alcohol dependence”). Inclusion and exclusion criteria for the parent studies are discussed elsewhere (Collins et al., 2019; Collins et al, 2021).

Table 1.

Sociodemographic information for NAI, Black, and White participants

Participants Demographics (N = 493) Frequency (%)
Sex assigned at birth Male 394 (79.9%)
Female 99 (20.1%)
Ethnicity
Not Hispanic/ Latinx 456 (93.1%)
Hispanic/ Latinx 34 (6.9%)
Race
Black/African American 205 (41.6%)
American Indian/ Alaska Native 125 (25.3%)
White/European American 163 (33.1%)
Highest level of education
7th Grade or less 22 (4.5%)
8th-12th Grade 229 (46.4%)
GED 79 (16.0%)
Vocational School 15 (3.0%)
Some College 117 (23.7%)
College Graduate 25 (5.1%)
Some Graduate School 3 (.6%)
Advanced Degree 2 (.4%)
Missing 1 (.2%)
Open in a new tab

Measures

Sociodemographic Questions were single items used to assess self-reported age, sex assigned at birth, race, ethnicity, education, and experience of homelessness in the past year.

The Short Inventory of Problems (SIP-2R) is a 15-item, Likert-scale questionnaire that measures firsthand alcohol-related harm encompassing physical, psychological, interpersonal, social, and impulse control issues. Instructions ask respondents to report how often an alcohol-related consequence has happened to them (e.g. “I have been unhappy because of my drinking”) on a 4-point scale (Never = 0, Daily or Almost Daily = 3; Miller et al., 1995a).

The Alcohol Quantity and Use Assessment (AQUA) measure comprises a series of open-ended questions asking participants about their quantity of alcohol use within a given timeframe, including alcohol type, unit size, and number of units consumed. This measure, used in conjunction with the Blood Alcohol Concentration Calculation System (Markham et al., 1993), yielded a single-item, self-report measure of alcohol quantity consumed on one’s peak drinking occasion in the past 2 weeks or 30 days, depending on the study (i.e., peak drinks) (Collins et al., 2015; Collins et al., 2012b; Collins et al., 2014; Larimer et al., 2009).

Procedures

All data included in this secondary study were collected during the baselines of two RCTs testing the efficacy of harm reduction treatments for AUDs among people experiencing homelessness (Collins et al., 2019; Collins et al., 2021; Collins et al., 2014). All research procedures used in the parent studies were approved by the University of Washington Institutional Review Board. Participants were approached by research staff in community-based settings that provide shelter, housing and/or other supportive services to people experiencing homelessness. Interested individuals who met RCT inclusion criteria provided written informed consent and engaged in a 45-minute baseline interview. These baseline data were used for the analyses described in the present, secondary study. More information about the parent study procedures can be found elsewhere (Collins et al., 2019; Collins et al., 2021; Collins et al., 2014). One of the studies was registered with clinicaltrials.gov (https://www.clinicaltrials.gov/ct2/show/NCT01932801). Code and materials are available upon request to the corresponding author.

Analytic Plan

All variables of interest were assessed to test assumptions of normality, homoscedasticity, and multicollinearity using SPSS 26.0. Internal consistency (i.e., Cronbach alpha coefficients) and convergent validity (i.e., correlations between SIP-2R and peak drinks) of the SIP-2R were likewise analyzed using SPSS 26.0.

To initially establish overall construct validity of the SIP-2R, a series of latent variable confirmatory factor analyses were conducted using weighted least squares (WLSMV) in RStudio 1.2 utilizing Lavaan (latent variable analysis) package (Rosseel et al, 2022). Lavaan utilizes delta-parameterization, and fixes one loading on each factor to one. WLSMV was used due to the categorical nature of item responses (i.e., SIP-2R). WLSMV has been shown to be less biased and more accurate than robust maximum likelihood in estimating factor loadings for categorical data (Li, 2016). Due to the categorical data, standardized estimates were interpreted (Byrne, 2005).

As in Marra et. al (2014), three competing factor structures were tested in the overall sample, including a) a first-order, single-factor model, b) a first-order, five-factor model with three items loading onto each factor, and c) a second-order, five-factor model. Model fit was assessed using the Comparative Fit Index (CFI; cut off ≥ .95), the Root-Mean Square Error of Approximation (RMSEA; adequate fit cut off < .08), the Standardized Root Mean Square Residual (SRMR; cut-off < .08), and the model chi-square significance test statistic for overall model fit (Hu & Bentler, 1999; MacCallum et al., 1996).

Next, a multiple group CFA (MGCFA) was conducted to test measurement invariance across the three most highly represented racial groups in the study (i.e., NAI, Black and White). The test was conducted based on a three-step approach for ordered categorical indicators (Kite et al., 2018; Putnick & Bornstein, 2016). For categorical variables, parameters include thresholds rather than intercepts, and scale factors are used in the estimation of residuals (Sass, 2011). The three-step approach entails testing increasingly strict definitions of invariance across groups: a) configural invariance (i.e., all groups share the same factor structure), b) metric invariance (i.e., factor loadings constrained to be equal across groups), and c) scalar invariance (i.e., factor loadings and item thresholds constrained to be equal across groups). If scalar invariance is achieved, we can assume item thresholds are the same across groups.

Taken together, measurement invariance indicates we can draw meaningful conclusions about the experience of alcohol-related harm across racial groups (Milfont & Fischer, 2010; Widaman & Reise, 1997). Fit indices were used to determine if metric and scalar invariance held using the following criteria: ΔCFI >.01, ΔRMSEA > .015, and ΔSRMR > .030 for metric and ΔSRMR > .015 for scalar invariance (Chen, 2007).

Lastly, a test of partial invariance was conducted, as not all items held across the racial groups. Relaxing constraints for specific parameters (i.e., thresholds) was determined by using the sequential search method (Yoon & Kim, 2014). This method consists of relaxing one constrained parameter with the largest modification index at a time and then reanalyze the model after each parameter has been relaxed (Yoon & Kim, 2014). The sequential search method has a smaller type one error rate than the nonsequential method where all problematic constraints are relaxed simultaneously (Yoon & Kim, 2014). Constraints on thresholds were lifted until the model fit was not significantly different from prior models.

Results

Establishing the Baseline Structural Model

Overall model fit

Goodness-of-fit indices for both the first-order, five-factor model, and the second-order, five factor model showed close fit (see Table 2). See Table 3 for factor loadings. All five factors were highly correlated (Pearson’s rs .52–.67; p < .001; see Table 4). We applied Byrne’s (2005) criteria to determine the best model for our data. First, as shown in Table 4, the second-order, five-factor model was a close-fitting model with very minimal differences on fit from the first-order, five-factor model. Further, the correlation among the five factors was strong, which provides support for an underlying factor. Finally, it is theoretically plausible and clinically supported in prior research that different facets of alcohol-related harm fit underneath it as a higher-order umbrella construct. We thus concluded that the second-order, five-factor model best represents the factor structure of the SIP-2R in this sample with adequate fit. The second-order, five-factor model also showed close fit when modeled for the NAI, Black and White groups, respectively (see Table 5).

Table 2.

Confirmatory factor analysis for model identification

Model χ2, (df) CFI RMSEA (90CI) SRMR
First-Order 1-Factor 646.116*, (90) 0.930 .112 (.104, .121) 0.066
First-Order 5-Factor 306.62*, (80) 0.971 .076 (.067, .085) 0.045
Second-Order 5-Factor 320.30*, (85) 0.970 .075 (.067, .084) 0.048
Open in a new tab

Note.

*

p < .001

Table 3.

Factor loadings by racial group for second-order five-factor model

Factor Item Factor Loading
AI Black White
Physical 2. Because of my drinking, I have not eaten properly. .75 .74 .74
7. My physical health has been harmed by my drinking. .75 .78 .83
9. My physical appearance has been harmed by my drinking. .83 .82 .81
Intrapersonal 1. I have been unhappy because of my drinking. .82 .76 .78
4. I have felt guilty or ashamed because of my drinking. .85 .83 .82
12. My drinking has gotten in the way of my growth as a person. .82 .91 .91
Interpersonal 10. My family has been hurt by my drinking. .79 .85 .78
11. A friendship or close relationship has been damaged by my drinking. .85 .82 .74
13. My drinking has damaged my social life, popularity, or reputation. .87 .97 .79
Social Responsibility 3. I have failed to do what is expected of me because of my drinking. .62 .79 .80
8. I have had money problems because of my drinking. .77 .85 .81
14. I have spent too much or lost a lot of money because of my drinking. .83 .84 .87
Impulse control 5. I have taken foolish risks when I have been drinking. .77 .88 .90
6. When drinking, I have done impulsive things that I regretted later. .76 .77 .77
15. I have had an accident while drinking or intoxicated. .75 .67 .74
Open in a new tab
Table 4.

Pearson’s Bivariate Correlations Among five-factors

Factor 1 2 3 4 5
1: Physical - .65* .52* .66* .61*
2: Intrapersonal - .59* .67* .55*
3: Interpersonal - .61* .57*
4: Social Responsibility - .62*
5: Impulse Control -
Open in a new tab

Note.

*

p < .001

Table 5.

Second-order 5-factor ordinal confirmatory factor analysis by racial group

Racial Group χ2, df CFI RMSEA (90CI) SRMR
NAI 145.326*, 85 0.973 .076 (.054, .096) 0.064
Black 201.981*, 85 0.971 .083 (.068, .098) 0.058
White 147.204*, 85 0.973 .067 (.049, .085) 0.067
Open in a new tab

Note.

*

p < .001

Assessing Measurement Invariance Across Groups

Using the second-order, five-factor model, we conducted a multigroup confirmatory factor analysis to test measurement invariance. As shown in Table 6, ΔCFI, ΔRMSEA, and ΔSRMR between the configural and metric models was at the prescribed level, indicating that invariance of factor loadings across racial groups is tenable. However, there was a significant CFI difference between metric and scalar models. This suggests greater lack of fit when constraining both the loadings and thresholds to be equal.

Table 6.

Multiple group confirmatory factor analysis for second-order 5-factor model

Model Invariance χ2 (df) CFI RMSEA (90CI) SRMR Model Comp Δχ2 (Δdf) ΔCFI ΔRMSEA ΔSRMR Decision
M1: Configural 493.482* (255) .972 .076 (.066, .086) .063 - - - - - -
M2: Metric 507.199* (283) .974 .070 (.060, .080) .076 M1 13.717* (28) .002 .006 .013 Accept
M3: Scalar 648.967* (331) .963 .077 (.068, .086) .063 M2 141.768* (48) .011 .007 .007 Reject (based on ΔCFI > .01,)
M3a: Partial Scalar 624.902* (329) .966 .074 (.066, .083) .063 M2a 117.703* (46) .008 .004 .001 Accept
Open in a new tab

Note. N = 493; group 1 (NAI) n = 125; group 2 (Black) n = 205; group 3 (White) n = 163. Fit indices were used to determine if metric and scalar invariance held with the following criteria: ΔCFI > .01, ΔRMSEA > .015, and ΔSRMR > .030 for metric and ΔSRMR > .015 for scalar invariance (Chen, 2007). For partial scalar invariance, Item Sip3 threshold one was relaxed to obtain scalar invariance.

*

p ≤ .001

Because loss of fit was noted in moving from metric to scalar invariance, we conducted a follow-up test of partial scalar invariance (Byrne et al., 1989). Specifically, we conducted a forward partial invariance test by looking at the modification index to determine which item thresholds to relax across groups (Yoon & Kim, 2014). It was determined that the first item threshold from Item 3 was not invariant across racial groups (i.e., moving from “never” to “once or a few times” in response to “I have failed to do what is expected of me because of my drinking”). Only 10.4% of NAI responded “never” to this question, in comparison to 18% of Black and 23.3% of White adults. In contrast, 18.8 % NAI responded “once or a few times,” in comparison to 21% Black and 25.8% of White adults. Otherwise, all other thresholds were invariant, which exceeds the criteria that half of items show invariance (Putnick & Bornstein, 2016), and the changes in fit indices between the metric and partial scalar models did not exceed suggested levels (Chen, 2007). Thus, we determined the SIP-2R is partially scalar invariant across racial groups. See Figure 1 for the final partial scalar model.

Figure 1. Second-order five-factor ordinal model reflecting partial scalar invariance.

Figure 1.

Open in a new tab

Note. Three threshold loadings are notated under each factor loading; Sip3 threshold one is underlined and bolded with an asterisk notated as T1 to indicate it as the relaxed constraint. The text box below details the SIP3 threshold one across racial groups.

Reliability and Validity

The overall scale indicated strong internal consistency of the items (α = .94, ω =.95). Reliability across the SIP-2R five factors and across racial group is shown in Table 7. Convergent validity was supported by positive correlations between the total SIP-2R score and the number of drinks consumed on a heavy drinking day the total sample ρ (490)= .30, p < .001.

Table 7.

Reliability (Cronbach’s alpha ordinal and McDonald’s omega) across factors and by racial group

Cronbach’s alpha McDonald’s omega
Full Sample NAI Black White Full Sample NAI Black White
Full Model .94 .94 .95 .94 .95 .96 .97 .95
F1: Physical .83 .82 .82 .83 .78 .78 .78 .79
F2: Intrapersonal .85 .84 .83 .87 .85 .85 .89 .83
F3: Interpersonal .86 .87 .91 .81 .82 .82 .87 .75
F4: Social Responsibility .82 .75 .86 .83 .80 .76 .82 .84
F5: Impulse Control .79 .79 .80 .80 .79 .76 .78 .85
Open in a new tab

Note. F1 = Factor 1; F2 = Factor 2; F3 = Factor 3, F4 = Factor 4; F5 = Factor 5

Discussion

Our findings extend the existing literature on the psychometrics of the SIP-2R by providing tests of measurement invariance, internal consistency, and convergent validity in a community sample comprising NAI, Black, and White adults experiencing homelessness and AUD. Findings indicated partial scalar measurement invariance across racial groups, which coincided with prior studies showing measurement invariance of the SIP across ethnicities (Marra et al., 2014). The internal consistency of the SIP-2R was strong, which corresponds to findings from previous studies (Alterman et al., 2009; Feinn et al., 2003; Hirchak et al., 2021; Kenna et al., 2005; Kirouac & Witkiewitz, 2018; Marra et al., 2014). Evidence for convergent validity was also consistent with past research showing other alcohol outcomes to be associated with SIP-2R scores (Feinn et al., 2003; Hirchak et al., 2021; Kirouac & Witkiewitz, 2018). Taken together, findings from this study are consistent with prior work and extend the existing literature to provide indications of this measure’s strong psychometric properties and measurement invariance in a diverse, nontreatment-seeking, community-based population of adults experiencing homelessness and AUD.

Results from the CFA indicated that both the first-order, five-factor and second-order, five-factor models were close-fitting; however, the second-order, five-factor model provided a better theoretical fit to the data, while maintaining statistical fit and parsimony. The second-order factor was able to account for the pattern of correlations among first-order factors, test how the second-order factor accounts for these correlations, and account for their unique variance (Byrne, 2005). Thus, this study’s findings support the SIP-2R as a measure of overall alcohol-related harm, comprising five distinct subtypes (i.e., physical, interpersonal, intrapersonal, social responsibility, and impulse control issues).

Our study meaningfully adds to the literature by including these tests of the second-order, five-factor model, which formed the original theoretical foundation of this measure’s development and clinical utility (Miller et al 1995). Our results align with those of the only other study testing the factor structure of the second-order, five-factor model (Marra et al., 2014). (Findings from Marra et al. [2014] showed the second-order, five-factor model was the best fitting, despite the fact that they chose to test measurement invariance across ethnicities with the single-factor model.) Of note, the second-order, five-factor model has been omitted from some studies that have tested the factor structure of the SIP-2R. Some researchers have tested the first-order, five-factor model but not the second-order, five-factor model (Kenna et al., 2005; Alterman et al., 2009), and others tested a new second-order, three-factor model but not the second-order, five-factor model (Kirouac & Witkiewitz, 2018). A few studies have focused on the SIP-2R as a single-factor scale, emphasizing the intrinsic clinical importance of a general alcohol-related harm construct (Kenna et al., 2005; Alterman et al., 2009; Marra et al. 2014). Given the importance of the second-order construct in the present study, our findings do not negate this prior work. Instead, our findings provide support for a general alcohol-related harm construct, while also providing more differentiation among subtypes of alcohol-related harm. Our findings also align with research on the SIP-AD, which has likewise shown support for the second-order, five-factor model (Kiluk et al, 2013).

Importantly, the SIP-2R evinced partial scalar invariance across racial groups. This finding indicates that scores on the SIP-2R can be compared across NAI, Black, and White adults experiencing homelessness with AUD (Putnick & Bornstein, 2016). In turn, there is little to no measurement bias when using the SIP-2R to compare alcohol-related harm across these groups (Milfont & Fischer, 2010). The partial scalar invariance stemmed from one significant difference in participant responses: Threshold one of item three (“I have failed to do what is expected of me because of my drinking”), such that fewer NAI responded to “never” than did to “once and a few times” compared to Black and White adults. Perhaps this finding reflects differing expectations across collectivist and individualistic cultures, more generally, and in the NAI community, more specifically. Many NAI cultures foster more interdependent, interconnected relationships and focus on the needs of others within one’s community (Beckstein, 2014). Thus, it is possible that NAI may likely feel more compelled to endorse that item given a heightened sense of responsibility and commitment to one’s community, and/or might strive harder to continue to fulfil their social obligations, even when experiencing the sequelae of severe AUD. Future studies are needed to better understand the source of this invariance and make changes to the measure or its use as needed.

Clinical Implications

Our results have important clinical implications. The second-order factor or construct that overarches and connects the five subfactors, justifies an overall summary score as a general indicator of the severity of alcohol-related harm. This overall score can then be used to provide a snapshot of clients’ current level of alcohol-related harm as well as to track treatment progress or natural recovery trajectories. Additionally, the five factors can provide clinicians with more clarity about their clients’ differing and perhaps disproportionate experience of specific subtypes of alcohol-related harm (i.e., physical, interpersonal, intrapersonal, social responsibility, and impulse control). Given that our sample comprises people experiencing severe and disproportionately high levels of alcohol-related harm (Collins, 2016; 2019; Jacobs-Wingo et al., 2016; Whitbeck et al., 2012; Zapolski et al., 2014), this structure provides both the level of detail and flexibility needed to document the overall impact of clients’ alcohol use, highlight different facets where harm is disproportionately experienced, and provide potential points for intervention.

Given its consistently strong psychometric properties, the SIP-2R is well-positioned to further the clinicians’ and researchers’ understanding of trajectories of alcohol-related harm in severely affected and diverse populations. For example, the SIP-2R has tremendous utility in harm reduction research and treatment, where tracking incremental reduction of alcohol-related harm is foundational to its underlying theory and practice (Collins et al., 2019; Collins et al., 2021). Considering more severely impacted populations and more realistic perspectives on recovery pathways, continuous measures of alcohol-related harm are more aligned with and provide a more differentiated measurement of the experience and potential resolution of AUD than quantity and frequency measures alone. Thus, psychometrically sound measures that are sensitive to incremental change in alcohol-related harm — versus more common but less sensitive, dichotomous measures of heavy drinking/non-heavy drinking or abstinence/use — are sorely needed. Furthermore, our study supports the utilization of the SIP-2R to assess reduction in alcohol-related harm for harm reduction treatment for NAI, Black, and White adults experiencing homelessness and AUDs.

Limitations and Future Directions

This study should be understood within the context of its limitations. First, data were collected from a marginalized population of urban-dwelling adults experiencing homelessness and moderate-to-severe AUD (i.e., DSM-IV-TR diagnosis of “alcohol dependence”). Although representative of people experiencing chronic homelessness and AUD in urban community settings, the sample was skewed heavily male-identifying compared to the general population (79.9% in this sample versus 50.8% in the general population; US Census Bureau, 2019). Therefore, generalizability of these specific findings may be limited to similar population bases. Future studies are needed to continue to test the SIP-2R’s measurement invariance, validity, and reliability across further racial, cultural, ethnic and gender groups to ensure it is psychometrically sound across various populations.

Second, there are limitations to the data analysis approach. We used a standard approach to MGCFA, including the use a stepwise selection strategy based on modification indices to achieve partial scalar invariance (Kite et al., 2018; Putnick & Bornstein, 2016). It should be noted that this method has been criticized, particularly if many modifications are made, there are large sample sizes, or many groups (Marsh et al, 2018). Fortunately, the present study was not impacted by the aforementioned issues. Further, we used traditional fit index cutoffs (Hu & Bentler, 1999; MacCallum et al 1996). Future research efforts may incorporate new and emerging methods, including dynamic fit index cutoffs, which are tailored to the data and model (McNeish & Wolf, 2021). Additionally, future research may seek to examine how incremental improvements of alcohol-related harms following therapeutic interventions may affect the factor structure. Future studies involving larger sample sizes and using calibration and validation samples may also establish model replicability.

Third, there are limitations of the measure and the data collected. As previously noted, literature on the SIP’s factor structure has been mixed. Future studies may also consider alternative measures of alcohol-related harm, including the PROMIS measure, which was rigorously developed among patient reported processes (Pilkonis et al, 2016). This secondary study comprised data from two studies assessing harm reduction treatment for AUD with people experiencing homelessness. Participants from both studies were recruited throughout the same timeframe, from similar but distinct community-based sites, and are not overlapping across the two studies. Furthermore, there was no specific recruitment specification as to why participants were entered into either study.

Conclusions

Despite its limitations, this secondary study provided evidence for the convergent validity, internal consistency, and measurement invariance of the SIP-2R in a sample of urban-dwelling NAI, Black, and White adults experiencing homelessness with AUD. The current research supports the use of the SIP-2R as a psychometrically sound and clinically useful measure of alcohol-related harm both overall and in its more specific facets and across racial groups in a marginalized population. This study provides evidence to suggest the SIP-2R is a psychometrically valid measurement tool for clinicians to use with patients across racial groups to assess alcohol-related harm. Furthermore, results indicate researchers may use this tool across diverse racial groups. Future research is needed to further investigate the psychometric properties of the SIP-2R across other racial and ethnic groups, as well as across sex assigned at birth and gender.

Public Health Statement:

This study indicates a widely used measure of alcohol-related harm, the Short Inventory of Problems (SIP-2R), is psychometrically sound for use among North American Indigenous and Black adults experiencing homelessness and Alcohol Use Disorder, which is imperative for reliably and validly reporting on alcohol-related health inequities and conducting cross-group comparisons.

Acknowledgements

We would like to acknowledge our large networks of support of staff, colleagues, and community partners who supported the parent studies that have provided data for this secondary analysis. In addition, thank you to Professor Joseph S. Rossi, Ph.D. for his teachings that assisted in the statistical interpretations of this manuscript. Most of all, we would like to thank the study participants for their role in this research and for helping us understand the importance of measuring and reducing alcohol-related harm along their recovery pathway.

Funding:

Data for this secondary analysis was collected in the context of research program grants from the National Institute on Alcohol Abuse and Alcoholism (R01AA022309 and R34AA022077) to Susan E. Collins. Work on this paper by Silvi C. Goldstein was supported by National Institute on Alcohol Abuse and Alcoholism Grant F31 AA029274. Code and materials are available upon request to the corresponding author. The current analysis has only been presented at the national APA 2020 conference as a part of a graduate student poster presentation. The parent study was published in the Lancet Psychiatry in 2021 (Collins et al, 2021) and a dual study paper involving harm-reduction goals and their correlations with the SIP outcome was published in Experimental and Clinical Psychology in 2021 (Fentress et al, 2021).

References

  1. Allensworth‐Davies D, Cheng DM, Smith PC, Samet JH, & Saitz R (2012). The short inventory of problems—modified for drug use (SIP‐DU): Validity in a primary care sample. The American Journal on Addictions, 21(3), 257–262. 10.1111/j.1521-0391.2012.00223.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alterman AI, Cacciola JS, Ivey MA, Habing B, & Lynch KG (2009). Reliability and validity of the alcohol short index of problems and a newly constructed drug short index of problems. Journal of Studies on Alcohol and Drugs, 70(2), 304–307. 10.15288/jsad.2009.70.304 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baggett TP, Chang Y, Singer DE, Porneala BC, Gaeta JM, O’Connell JJ, & Rigotti NA (2015). Tobacco, alcohol and drug-attributable deaths and their contribution to mortality disparities in a cohort of homeless adults in Boston. American Journal of Public Health, 105, 1189–1197. 10.2105/AJPH.2014.302248 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Beckstein A (2014). Native American subjective happiness: An overview. Indigenous Policy Journal, 25(2). [Google Scholar]
  5. Bender RE, Griffin ML, Gallop RJ, & Weiss RD (2007). Assessing negative consequences in patients with substance use and bipolar disorders: Psychometric properties of the short inventory of problems (SIP). American Journal on Addictions, 16(6), 503–509. 10.1080/10550490701641058 [DOI] [PubMed] [Google Scholar]
  6. Bird CE, Jinnett KJ, Burnam MA, Koegel P, Sullivan G, Wenzel SL, Ridgely MS, Morton SC, & Miu A (2002). Predictors of contact with public service sectors among homeless adults with and without alcohol and other drug disorders. Journal of Studies on Alcohol, 63(6), 716–725. 10.15288/jsa.2002.63.716 [DOI] [PubMed] [Google Scholar]
  7. Blanchard KA, Morgenstern J, Morgan TJ, Lobouvie EW, & Bux DA (2003). Assessing consequences of substance use: Psychometric properties of the inventory of drug use consequences. Psychology of Addictive Behaviors, 17(4), 328. 10.1037/0893-164X.17.4.328 [DOI] [PubMed] [Google Scholar]
  8. Blount-Hill KL, & St. John V (2017). Manufactured “mismatch” cultural incongruence and black experience in the academy. Race and Justice, 7(2), 110–126. [Google Scholar]
  9. Brave Heart MYH, Chase J, Elkins J, & Altschul DB (2011). Historical trauma among Indigenous Peoples of the Americas: concepts, research, and clinical considerations. Journal of psychoactive drugs, 43(4), 282–290. [DOI] [PubMed] [Google Scholar]
  10. Burlew AK, Feaster D, Brecht M-L, & Hubbard R (2009). Measurement and data analysis in research addressing health disparities in substance abuse. Journal of Substance Abuse Treatment, 36(1), 25–43. 10.1016/j.jsat.2008.04.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Byrne BM (2005). Factor analytic models: Viewing the structure of an assessment instrument from three perspectives. Journal of Personality Assessment, 85(1), 17–32. 10.1207/s15327752jpa8501_02 [DOI] [PubMed] [Google Scholar]
  12. Byrne BM, Shavelson RJ, & Muthén B (1989). Testing for the equivalence of factor covariance and mean structures: The issue of partial measurement invariance. Psychological Bulletin, 105(3), 456. 10.1037/0033-2909.105.3.456 [DOI] [Google Scholar]
  13. Chen FF (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 464–504. 10.1080/10705510701301834 [DOI] [Google Scholar]
  14. Collins SE (2016). Associations between socioeconomic factors and alcohol outcomes. Alcohol Research: Current Reviews, 38(1), 83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Collins SE, Clifasefi SL, Logan DE, Samples LS, Somers JM, & Marlatt GA (2012a). Current status, historical highlights, and basic principles of harm reduction. In Marlatt GA, Larimer ME, & Witkiewitz K (Eds.), Harm reduction: Pragmatic strategies for managing high-risk behaviors (2nd edition ed., pp. 3–35). The Guilford Press. [Google Scholar]
  16. Collins SE, Clifasefi SL, Nelson LA, Stanton J, Goldstein SC, Taylor EM, Hoffmann G, King VL, Hatsukami AS, & Cunningham ZL (2019). Randomized controlled trial of harm reduction treatment for alcohol (HART-A) for people experiencing homelessness and alcohol use disorder. International Journal of Drug Policy, 67, 24–33. 10.1016/j.drugpo.2019.01.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Collins SE, Duncan MH, Saxon AJ, Taylor EM, Mayberry N, Merrill JO, Hoffmann GE, Clifasefi SL, & Ries RK (2021). Combining behavioral harm-reduction treatment and extended-release naltrexone for people experiencing homelessness and alcohol use disorder in the USA: A randomised clinical trial. The Lancet Psychiatry, 8(4), 287–300. 10.1016/S2215-0366(20)30489-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Collins SE, Duncan MH, Smart BF, Saxon AJ, Malone D, Jackson T, & Ries R (2015). Extended-release naltrexone and harm reduction counseling for chronically homeless people with alcohol dependence. Substance Abuse, 36, 21–33. 10.1080/08897077.2014.904838 [DOI] [PubMed] [Google Scholar]
  19. Collins SE, Malone DK, Clifasefi SL, Ginzler JA, Garner MD, Burlingham B, Lonczak HS, Dana EA, Kirouac M, & Tanzer K (2012b). Project-based housing first for chronically homeless individuals with alcohol problems: Within-subjects analyses of 2-year alcohol trajectories. American Journal of Public Health, 102(3), 511–519. 10.2105/AJPH.2011.300403 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Collins SE, Saxon AJ, Duncan MH, Smart BF, Merrill JO, Malone DK, Jackson TR, Clifasefi SL, Joesch J, & Ries RK (2014). Harm reduction with pharmacotherapy for homeless people with alcohol dependence: Protocol for a randomized controlled trial. Contemporary Clinical Trials, 38(2), 221–234. 10.1016/j.cct.2014.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Culhane D, Henry M, Watt R, Mahathey A, Ouellette J, & Sitler A (2019). The 2019 annual homeless assessment report (AHAR) to congress, Part 1: Point-in-time estimates of homelessness. Washington, DC: US Department of Housing and Urban Development. https://www.hudexchange.info/resource/5948/2019-ahar-part-1-pit-estimates-of-homelessness-in-the-us/ [Google Scholar]
  22. Dillon FR, Whiteman K, & Duan R (2015). Measurement invariance of the short inventory of problems-revised across African-American and non-Latino White substance users. Journal of Ethnic And Cultural Diversity in Social Work, 24(2), 109–129. 10.1080/15313204.2014.977985 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Fazel S, Geddes JR, & Kushel M (2014). The health of homeless people in high-income countries: Descriptive epidemiology, health consequences, and clinical and policy recommendations. The Lancet, 384(9953), 1529–1540. 10.1016/S0140-6736(14)61132-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fazel S, Khosla V, Doll H, & Geddes J (2008). The prevalence of mental disorders among the homeless in western countries: Systematic review and meta-regression analysis. PLOS Medicine, 5, e225. 10.1371/journal.pmed.0050225 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Feinn R, Tennen H, & Kranzler HR (2003). Psychometric properties of the short index of problems as a measure of recent alcohol‐related problems. Alcoholism: Clinical and Experimental Research, 27(9), 1436–1441. 10.1097/01.ALC.0000087582.44674.AF [DOI] [PubMed] [Google Scholar]
  26. Fentress TS, Wald S, Brah A, Leemon G, Reyes R, Alkhamees F, … & Collins SE (2021). Dual study describing patient-driven harm reduction goal-setting among people experiencing homelessness and alcohol use disorder. Experimental and clinical psychopharmacology, 29(3), 261. [DOI] [PubMed] [Google Scholar]
  27. Forcehimes AA, Tonigan JS, Miller WR, Kenna GA, & Baer JS (2007). Psychometrics of the drinker inventory of consequences (DRINC). Addictive Behaviors, 32(8), 1699–1704. 10.1016/j.addbeh.2006.11.009 [DOI] [PubMed] [Google Scholar]
  28. Gibbons FX, Gerrard M, Cleveland MJ, Wills TA, & Brody G (2004). Perceived discrimination and substance use in african american parents and their children: A panel study. Journal of Personality and Social Psychology, 86(4), 517. 10.1037/0022-3514.86.4.517 [DOI] [PubMed] [Google Scholar]
  29. Gibbons FX, O’Hara RE, Stock ML, Gerrard M, Weng C-Y, & Wills TA (2012). The erosive effects of racism: Reduced self-control mediates the relation between perceived racial discrimination and substance use in African American adolescents. Journal of Personality and Social Psychology, 102(5), 1089–1104. 10.1037/a0027404 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Gibbons FX, Yeh H-C, Gerrard M, Cleveland MJ, Cutrona C, Simons RL, & Brody GH (2007). Early experience with racial discrimination and conduct disorder as predictors of subsequent drug use: A critical period hypothesis. Drug and Alcohol Dependence, 88, S27–S37. 10.1016/j.drugalcdep.2006.12.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Gillespie W, Holt JL, & Blackwell RL (2007). Measuring outcomes of alcohol, marijuana, and cocaine use among college students: A preliminary test of the shortened inventory of problems—alcohol and drugs (SIP-AD). Journal of Drug Issues, 37(3), 549–567. 10.1177/002204260703700304 [DOI] [Google Scholar]
  32. Goldstein SC, Spillane NS, Schick MR, & Rossi J (in press). Measurement invariance and application of an alcohol-related problem scale for American Indian adolescents. Assesssment [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Grant BF, Dawson DA, Stinson FS, Chou SP, Dufour MC, & Pickering RP (2004). The 12-month prevalence and trends in DSM-IV alcohol abuse and dependence: United states, 1991–1992 and 2001–2002. Drug and Alcohol Dependence, 74, 223–234. 10.1016/j.drugalcdep.2004.02.004 [DOI] [PubMed] [Google Scholar]
  34. Greenberg GA, & Rosenheck RA (2010). Correlates of past homelessness in the national epidemiological survey on alcohol and related conditions. Administration and Policy in Mental Health and Mental Health Services Research, 37(4), 357–366. 10.1007/s10488-009-0243-x [DOI] [PubMed] [Google Scholar]
  35. Hagman BT, Kuerbis AN, Morgenstern J, Bux DA, Parsons JT, & Heidinger BE (2009). An item response theory (IRT) analysis of the short inventory of problems-alcohol and drugs (SIP-AD) among non-treatment seeking men-who-have-sex-with-men: Evidence for a shortened 10-item SIP-AD. Addictive Behaviors, 34(11), 948–954. 10.1016/j.addbeh.2009.06.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Heck RH, Thomas SL, & Tabata LN (2013). Multilevel and longitudinal modeling with IBM SPSS. Routledge. [Google Scholar]
  37. Hirchak KA, Tonigan JS, Hernandez-Vallant A, Herron J, Cloud V, & Venner KL (2021). The validity of the short inventory of problems and drinking intensity among urban American Indian adults. Substance Use & Misuse, 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Home A (2019). Seattle/King Country point-in-time count of persons experiencing homelessness. http://allhomekc.org/wp-content/uploads/2019/05/2019-Report_KingCounty_FINAL.pdf
  39. Hu L. t., & Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. [Google Scholar]
  40. Hwang SW, Wilkins R, Tjepkema M, O’Campo P, & Dunn JR (2009). Mortality among residents of shelters, rooming houses and hotels in Canada: 11 year follow-up. British Medical Journal, 339, b4036. 10.1136/bmj.b4036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Indian Health Service. (2019). Fact sheet: Disparities. https://www.ihs.gov/newsroom/factsheets/disparities/
  42. Jacobs-Wingo JL, Espey DK, Groom AV, Phillips LE, Haverkamp DS, & Sandte L (2016). Causes and disparities in death rates among urban American Indian and Alaska Native populations, 1999–2009. American Journal of Public Health, 106(5), 906–914. 10.2105/AJPH.2015.303033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Kenna GA, Longabaugh R, Gogineni A, Woolard RH, Nirenberg TD, Becker B, Minugh PA, Carty K, Clifford PR, & Karolczuk K (2005). Can the short index of problems (SIP) be improved? Validity and reliability of the three-month SIP in an emergency department sample. Journal of Studies on Alcohol, 66(3), 433–437. 10.15288/jsa.2005.66.433 [DOI] [PubMed] [Google Scholar]
  44. Kiluk BD, Dreifuss JA, Weiss RD, Morgenstern J, & Carroll KM (2013). The short inventory of problems–revised (SIP-R): Psychometric properties within a large, diverse sample of substance use disorder treatment seekers. Psychology of Addictive Behaviors, 27(1), 307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Kirouac M, & Witkiewitz K (2018). Revisiting the Drinker Inventory of Consequences: An extensive evaluation of psychometric properties in two alcohol clinical trials. Psychology of Addictive Behaviors, 32(1), 52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Kite BA, Jorgensen TD, & Chen P-Y (2018). Random permutation testing applied to measurement invariance testing with ordered-categorical indicators. Structural Equation Modeling: A Multidisciplinary Journal, 25(4), 573–587. 10.1080/10705511.2017.1421467 [DOI] [Google Scholar]
  47. Krausz RM, Clarkson AF, Strehlau V, Torchalla I, Li K, & Schuetz CG (2013). Mental disorder, service use and barriers to care among 500 homeless people in 3 different urban settings. Social Psychiatry and Psychiatric Epidemiology, 48, 1235–1243. 10.1007/s00127-012-0649-8 [DOI] [PubMed] [Google Scholar]
  48. Landen M, Roeber J, Naimi T, Nielsen L, & Sewell M (2014). Alcohol-attributable mortality among American Indians and Alaska Natives in the United States, 1999–2009. American Journal of Public Health, 104(S3), S343–S349. 10.2105/AJPH.2013.301648 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Larimer ME, Malone DK, Garner MD, Atkins DC, Burlingham B, Lonczak HS, Tanzer K, Ginzler J, Clifasefi SL, & Hobson WG (2009). Health care and public service use and costs before and after provision of housing for chronically homeless persons with severe alcohol problems. JAMA – Journal of the American Medical Association, 301(13), 1349–1357. 10.1001/jama.2009.414 [DOI] [PubMed] [Google Scholar]
  50. MacCallum RC, Browne MW, & Sugawara HM (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods, 1(2), 130. 10.1037/1082-989X.1.2.130 [DOI] [Google Scholar]
  51. Markham MR, Miller WR, & Arciniega L (1993). Baccus 2.01: Computer software for quantifying alcohol consumption. Behavior Research Methods, Instruments, & Computers, 25(3), 420–421. [Google Scholar]
  52. Marra L, Field CA, Caetano R, & von Sternberg K (2014). Construct validity of the short inventory of problems among spanish speaking Hispanics. Addictive Behaviors, 39(1), 205–210. 10.1016/j.addbeh.2013.09.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Marsh HW, Guo J, Parker PD, Nagengast B, Asparouhov T, Muthén B, & Dicke T (2018). What to do when scalar invariance fails: The extended alignment method for multi-group factor analysis comparison of latent means across many groups. Psychological Methods, 23(3), 524. [DOI] [PubMed] [Google Scholar]
  54. McNeish D, & Wolf MG (2021). Dynamic fit index cutoffs for confirmatory factor analysis models. Psychological Methods. Advance online publication. 10.1037/met0000425 [DOI] [PubMed] [Google Scholar]
  55. Meredith W (1993). Measurement invariance, factor analysis and factorial invariance. Psychometrika, 58(4), 525–543. 10.1007/BF02294825 [DOI] [Google Scholar]
  56. Milfont TL, & Fischer R (2010). Testing measurement invariance across groups: Applications in cross-cultural research. International Journal of Psychological Research, 3(1), 111–130. [Google Scholar]
  57. Miller W, Tonigan JS, & Longabaugh R (1995a). The drinker inventory of consequences (DrInC): An instrument for assessing adverse consequences of alcohol abuse (vol. 4). Rockville, MD: National Institute on Alcohol Abuse and Alcoholism. [Google Scholar]
  58. Morse DT, & Robertson AA (2017). Psychometric properties of the Short Inventory of Problems (SIP) with adjudicated DUI intervention participants. Psychology of addictive behaviors, 31(1), 110. [DOI] [PubMed] [Google Scholar]
  59. Mulia N, Ye Y, Greenfield TK, & Zemore SE (2009). Disparities in alcohol‐related problems among White, Black, and Hispanic Americans. Alcoholism: Clinical and Experimental Research, 33(4), 654–662. 10.1111/j.1530-0277.2008.00880.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Muthén B (1978). Contributions to factor analysis of dichotomous variables. Psychometrika, 43(4), 551–560. [Google Scholar]
  61. Noel M, O’connor RM, Boudreau B, Mushquash CJ, Comeau MN, Stevens D, & Stewart SH (2010). The rutgers alcohol problem index (RAPI): A comparison of cut-points in First Nations Mi’kmaq and non-Aboriginal adolescents in rural Nova Scotia. International Journal of Mental Health and Addiction, 8(2), 336–350. 10.1007/s11469-009-9219-z [DOI] [Google Scholar]
  62. North CS, Eyrich-Garg KM, Pollio DE, & Thirthalli J (2010). A prospective study of substance use and housing stability in a homeless population. Social Psychiatry and Psychiatric Epidemiology, 45(11), 1055–1062. 10.1007/s00127-009-0144-z [DOI] [PubMed] [Google Scholar]
  63. Pilkonis PA, Yu L, Dodds NE, Johnston KL, Lawrence SM, & Daley DC (2016). Validation of the alcohol use item banks from the Patient-Reported Outcomes Measurement Information System (PROMIS®). Drug and alcohol dependence, 161, 316–322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Putnick DL, & Bornstein MH (2016). Measurement invariance conventions and reporting: The state of the art and future directions for psychological research. Developmental Review, 41, 71–90. 10.1016/j.dr.2016.06.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Rosseel Y, Oberski DL, Byrnes J, Vanbrabant L, Savalei V, Merkle E, … & Barendse M (2022). Lavaan: Latent variable analysis.
  66. Sass DA (2011). Testing measurement invariance and comparing latent factor means within a confirmatory factor analysis framework. Journal of Psychoeducational Assessment, 29(4), 347–363. 10.1177/0734282911406661 [DOI] [Google Scholar]
  67. Stahre M, & Simon M (2010). Alcohol-related deaths and hospitalizations by race, gender, and age in California. The Open Epidemiology Journal, 3(1), 3–15. 10.2174/1874297101003010003 [DOI] [Google Scholar]
  68. Substance Abuse and Mental Health Services Administration (SAMHSA). (2018). Results from the 2017 national survey on drug use and health: Detailed tables. https://www.samhsa.gov/data
  69. Substance Abuse and Mental Health Services Administration (SAMHSA). (2019a). Results from the 2018 national survey on drug use and health: Detailed tables. https://www.samhsa.gov/data/
  70. Substance Abuse and Mental Health Services Administration (SAMHSA). (2019b). Tip 61: Behavioral health services for American Indians and Alaska Natives. [PubMed] [Google Scholar]
  71. Tonigan JS, & Miller WR (2002). The inventory of drug use consequences (InDUC): Test-retest stability and sensitivity to detect change. Psychology of Addictive Behaviors, 16(2), 165. 10.1037/0893-164X.16.2.165 [DOI] [PubMed] [Google Scholar]
  72. Tsai J (2018). Lifetime and 1-year prevalence of homelessness in the us population: Results from the national epidemiologic survey on alcohol and related conditions-III. Journal of Public Health, 40(1), 65–74. 10.1093/pubmed/fdx034 [DOI] [PubMed] [Google Scholar]
  73. U.S. Census Bureau (2019). 2019: ACS demographic and housing estimates. https://www.census.gov/data.html
  74. Vandenberg RJ, & Lance CE (2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recommendations for organizational research. Organizational Research Methods, 3(1), 4–70. 10.1177/109442810031002 [DOI] [Google Scholar]
  75. Walls ML, Whitesell NR, Barlow A, & Sarche M (2019). Research with American Indian and Alaska Native populations: Measurement matters. Journal of Ethnicity in Substance Abuse, 18(1), 129–149. 10.1080/15332640.2017.1310640 [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Welty T (2003). The epidemiology of alcohol use and alcohol-related health problems among American Indians and Alaska Natives. In Alcohol use among American Indians and Alaska Natives: Multiple perspectives on a complex problem (pp. 49–70). National Institute on Alcohol Abuse and Alcoholism. [Google Scholar]
  77. Whitbeck LB, Crawford DM, & Hartshorn KJS (2012). Correlates of homeless episodes among Indigenous people. American Journal of Community Psychology, 49(1–2), 156–167. 10.1007/s10464-011-9446-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Whitesell NR, Beals J, Crow CB, Mitchell CM, & Novins DK (2012). Epidemiology and etiology of substance use among American Indians and Alaska Natives: Risk, protection, and implications for prevention. The American Journal of Drug and Alcohol Abuse, 38(5), 376–382. 10.3109/00952990.2012.694527 [DOI] [PMC free article] [PubMed] [Google Scholar]
  79. Widaman K, & Reise S (1997). Exploring the measurement invariance of psychological instruments: Applications in the substance use domain In Bryant KJ, Windle M, & West SG (Eds.), The science of prevention: Methodological advances from alcohol and substance abuse research (pp. 281–324). American Psychological Association. 10.1037/10222-009 [DOI] [Google Scholar]
  80. Wiechelt SA (2012). Shelly a. Wiechelt on Dr. Anonymous: “Treatment failure one step removed: The loss of researchers in the addictions” chasing the holy grail. Substance Use & Misuse, 47(13–14), 1590. 10.3109/10826084.2012.738901 [DOI] [PubMed] [Google Scholar]
  81. Wilson SR, Knowles SB, Huang Q, & Fink A (2014). The prevalence of harmful and hazardous alcohol consumption in older UD adults: Data from the 2005–2008 national health and nutrition examination survey (NHANES). Journal of General Internal Medicine, 29(2), 312–319. 10.1007/s11606-013-2577-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. Witbrodt J, Mulia N, Zemore SE, & Kerr WC (2014). Racial/ethnic disparities in alcohol‐related problems: Differences by gender and level of heavy drinking. Alcoholism: Clinical and Experimental Research, 38(6), 1662–1670. 10.1111/acer.12398 [DOI] [PMC free article] [PubMed] [Google Scholar]
  83. Yoon M, & Kim ES (2014). A comparison of sequential and nonsequential specification searches in testing factorial invariance. Behavior Research Methods, 46(4), 1199–1206. 10.3758/s13428-013-0430-2 [DOI] [PubMed] [Google Scholar]
  84. Zapolski TC, Pedersen SL, McCarthy DM, & Smith GT (2014). Less drinking, yet more problems: Understanding African American drinking and related problems. Psychological Bulletin, 140(1), 188–223. 10.1037/a0032113 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES