Article Critique for the Examination of Illicit Drug Use Frequency Using Multiple Drug Assessment Methods in Mothers Referred to Treatment by Child Protective Services

Article Critique Proposal for Examination of Illicit Drug Use Frequency Using Multiple Drug Assessment Methods in Mothers Referred to Treatment by Child Protective Services

Sherrie Miranda

Liberty University

Article Critique for the Examination of Illicit Drug Use Frequency Using Multiple Drug Assessment Methods in Mothers Referred to Treatment by Child Protective Services

The article critiqued in this paper is titled “Examination of Illicit Drug Use Frequency Using Multiple Drug Assessment Methods in Mothers Referred to Treatment by Child Protective Services”.  This study article was published in 2014 in the Journal of Family Violence.  It’s available in the eighth issue of volume twenty nine.  Brad Donohue, Jason M. Holland, Karen Lopez, Jessica Urgelles, and Daniel N. Allen conducted the study and co-authored the article.  The article explained the reasons, methods, and results of a study that measured and compared the accuracy of drug-use determination techniques in mothers who had been referred to treatment by Child Protection Services. Although this study has noble and practical aims, it tries to do too much with too little.  

Critical Summary

Identifying Information

The overall purpose of this study was to determine if there is a reliable means for assessing drug use patterns in mothers who had been referred to treatment by Child Protection Services.  It utilized surveys from mothers, their significant others, case workers, and urinalysis tests to gather the data for this investigation.  This study sought to answer three main questions.  The first question was, “How close will self-reports and correlate reports be to baseline urinalysis findings?”  (Donohue, Holland, Lopez, Urgelles, & Allen, 2014, p. 912). To this inquiry, researchers hypothesized that mothers who had received positive urinalysis results would report higher use than those who had received negative results (Donohue et al., 2014).  Additionally, mothers with positive urinalysis results would report higher drug use than their correlates (Donohue et al., 2014).  Secondly, the researchers desired to find how closely the mothers’ self-reports would match their correlates’ reports (Donohue et al., 2014).  The investigators presumed that the mothers’ self-reports would be higher than the correlates’ reports because the mothers had first-hand knowledge of their own drug use (Donohue et al., 2014).  The examiners’ third query was entailed identifying “unique predictors of clients’ and their significant others’ drug use reports?” (p. 912). They intuited that mothers who exhibited defensive behavior, faced negative consequences, or encountered greater scrutiny would report less recent drug use; moreover, they expected spouses and romantic partners, and significant others living outside of the mothers’ homes to report less drug use (Donohue et al., 2014).    

Methods

This was a quantitative study.  Seventy seven mothers participated in this study.  They all had to meet the following criteria:

(1) They had to have been investigated by CPS for child neglect within four months of referral; (2) use of illicit drugs within four months of referral; (3) diagnosed with Substance Abuse or Dependence according to the results of a structured clinical interview (Structured Clinical Interview for DSM-IV; First, Spitzer, Gibbon, and Williams 1996); (4) residing with, or scheduled to reside with, the child referred for neglect; (5) not referred for child sexual abuse; and (6) domestic violence was not the primary reason for referral. (Donohue, Holland, Lopez, Urgelles, & Allen, 2014, p. 912).

The secondary participants in this study were called significant others.  These were people that the mother identified as persons that would help in their treatment.  This opened up the study for biasness because those identified may or may not have truly been involved in assisting the client in treatment.  Additionally, it is not far-fetched to assume that they could provide answers to reports that they felt would place the mother in a favorable light.  The third group of participants was case workers of the mothers.  

The study was initiated by first conducting telephone interviews with mothers who had been referred to treatment and conducting similar interviews with their case workers.  Those who met the study’s criteria were invited to an in-person assessment. This assessment was composed of urinalysis testing, demographic questions, timeline follow-back questions and defensive responding scale measurement (Donohue et al., 2014).  Some of the ethical concerns met surrounding this assessment included informed consent of the urinalysis tests along with confidentiality certificates (Donohue et al., 2014).  Data from the mothers’ caseworkers were received upon referral. In reports submitted at this time, caseworkers wrote of instances that they knew their clients had used within the four months prior to the date of referral according to positive urinalysis or hair follicle tests.    

Interpretation of Findings

The predictions of the investigators were evaluated by utilizing Statistical Package for the Social Sciences (SPSS) software programming.  This enabled them to model several elements of this study using number of days of hard drug or marijuana use as dependent variables.  Because the reports were reflective but conducted at the same point in time, the data was treated the same as if it had been a longitudinal study.  The investigation revealed that clients and their significant others tended to report lower substance use during times closest to the date of treatment referral compared to their caseworkers.  The first question was, “How close will self-reports and correlate reports be to baseline urinalysis findings?”  (Donohue, Holland, Lopez, Urgelles, & Allen, 2014, p. 912). To this inquiry, researchers hypothesized that mothers who had received positive urinalysis results would report higher use than those who had received negative results (Donohue et al., 2014).  Additionally, mothers with positive urinalysis results would report higher drug use than their correlates (Donohue et al., 2014).  The findings were in conjunction with the researchers predictions.  Secondly, the researchers desired to find how closely the mothers’ self-reports would match their correlates’ reports (Donohue et al., 2014).  The investigators presumed that the mothers’ self-reports would be higher than the correlates’ reports because the mothers had first-hand knowledge of their own drug use (Donohue et al., 2014).  In general, the findings corroborated with the researchers hypotheses.  However, there was one notable finding that the researchers were compelled to expound upon.  That is, although African American mothers had reported similar trends with regards to hard drug use as Caucasian mothers, their reported marijuana use was significantly higher than that of their Caucasian counterparts even up to one month prior to the investigation.  The investigators identified the following limitations: 1) small sample; 2) data gathered at a single point in time; and 3) no controlled portion of the sample to effectively make useful comparisons.  Because of the limitations of the study, it would not be advisable to conclude that the findings could be applied to other populations.  

Personal Reflection

Strengths/Weaknesses

According to Donohue et al. (2014), there is no comprehensive means for examining “illicit drug use in mothers referred to treatment by Child Protective Services” (p. 911).  They pointed out that although urinalysis testing is accurate and relatively inexpensive, it cannot be used to determine drug use patterns over long periods of time (one month or more).  Hair follicle testing does not have this limitation. However, it is prohibitively expensive.  As a result, a combination of methods should be utilized to accurately assess drug use behaviors of mothers referred to treatment by Child Protective Services.  This is a strong argument for the overall aim of this study.  The investigators reported that they used the timeline follow-back (TLFB) method in conjunction with urinalysis testing in an effort to observe concordance between mothers’ self-reports, collateral reports, and urinalysis findings.  They stated that the TLFB method is the most “psychometrically validated self- and collateral report method of illicit drug use frequency” (p. 912).  This statement and along with what seems to be a naïve assumption that the reader understands the TLFB method is a weakness of this study.  It’s heavily used in this study to gather much needed data, yet the researchers inadequately described it or the reason why it should be trusted. This point is made obvious as one reads the articles entitled, “Reliability of the Timeline Followback for Cocaine, Cannabis, and Cigarette Use” and “Assessing Drug Use During Follow-up: Direct Comparison of Candidate Outcome Definitions in Pooled Analyses of Addiction Treatment Studies”.  Robinson, Sobell, Sobell, & Leo (2014) provided a much more clear explanation of the TLFB method.  They also demonstrated why this method is a valid and reliable means for collecting self-reported substance use patterns.  One of the most significant points gathered is that TLFB methods are shown to be reliable because test-retest values are very similar to each other.  That is, when clients are asked to provide a history of their substance use over a certain 30-day period, the results tend to agree when asked about that same period of time’s substance use if the test is conducted again seven to fourteen days after the original set of questions were given.   Although the subject of this critique and the article just discussed provide raving reviews for the TLFB method, Korte et al. (2011) study showed that TLFB was not as reliable as others studies had indicated.  Thus, there is evidence on both sides of the fence with regards to the TLFB method.  Therefore, the subject of this critique’s significant use of TLFB methods to draw conclusions about the substance use patterns is weak.  On another note, Donohue et al. (2014) purported since parental substance abuse is closely associated with child maltreatment there is a need to identify predictors of different drug use patterns. There can be no doubt that trying to determine predictors of drug use patterns is of paramount importance for those who are in contact with substance abusing mothers.  This aim can be viewed as a strength of this study.  However, it is extremely important to realize that the one of the most evident predictors of drug use patterns in the population of interest is employment satisfaction.  This very important factor was not considered in this study.  Nevertheless, Donohue, Plant, Barchard, & Gillis (2017) concluded “These findings suggest that when mothers are involved in child protective services their risk of perpetrating child maltreatment may be reduced when they are assisted in gainful employment that is personally satisfying” (p. 168).