• In this article: What was the research problem and was it clearly defined? How was the problem statement explicated?

Culturally Competent Intervention with Families of Latino Youth at Risk for Drug Abuse

Dhira Crunkilton Juan J. Paz David P. Boyle

SUMMARY.A culturally specific intervention for families with chil- dren at risk for substance abuse was implemented byProyecto Juventud(Youth Project), a community-based family services program located in the rural southwestern United States. This intervention, incorporating culturally appropriate language, beliefs, values, and principles, provided substance abuse education to families. Participants were pretested at in- take and posttested at one year, utilizing one measure for knowledge of drugs and drug abuse, and a second measure for family functioning. Re- sults indicated that parents' scores relating to drug and drug abuse knowledge showed a trend of improvement on 16 out of 20 items. The results on family functioning were inconclusive. The study illustrated

SUBSTANCE ABUSING LATINOS

the importance of using a cultural approach with Latino families in the

area of substance abuse prevention.

KEYWORDS.Latino substance abuse prevention, family intervention, culturally specific intervention

INTRODUCTION

Research on Latino parental influences consistently points to the critical role that the attitudes and behaviors of parents play in predict- ing early use of substances by their children (Coombs, Paulson, & Richardson, 1991; Department of Health and Human Services, 1995; Hawkins, Catalano, & Miller, 1992; Hernandez & Lucero, 1996; Johnson & Johnson, 1999; Scheier, Botvin, & Miller, 1999; Sealy, 2000; Vega, Kourney, Zimmerman, & Gil, 1995). Researchers have emphasized that adolescent support networks may be more important in explaining drug use behavior than the individual characteristics of the adolescents themselves (Krohn & Thornberry, 1993; Wallace, 1999). Latino parental influence and quality of the parent-child rela- tionship were critical predictors in the initiation and experimentation of alcohol, tobacco, and other drug use among high-risk youth (Coombs et al., 1991; Hawkins et al., 1992).

Recent studies (Scheier et al., 1999; Sealy, 2000) further empha- sized the key importance of the parent-child relationship in substance abuse prevention. Sealy's evidence revealed that the higher the level of parental support, as defined as praise, encouragement, and physical affection, the less often the adolescent drank alcohol. Scheier et al. found that the beneficial life experiences of Hispanic youth such as "family communication," as well as an internal health focus of con- trol, buffered the effects of negative life events on alcohol use. John- son and Johnson (1999) also reported on the buffering effect of the Hispanic family on the initiation of alcohol use by children and ado- lescents. In a press release summarizing a critical juncture in the accu- mulation of research pointing to the importance of parent-child communication in light of increasing rates of Latino youth drug use, Chavez (1999) stated:

Crunkilton, Paz, and Boyle 115

As the Substance Abuse and Mental Health Services Administra- tion National Household Survey on Drug Abuse has shown, heavy alcohol use among Hispanic/Latino youth has doubled in the last few years. And following all trends of youth, there has been an in- crease of illicit drug use among Hispanic/Latino youth during the 1990s . . . Research shows that parents are consistently identified by young people as the most important trustworthy sources of in- formation and guidance on substance abuse and other serious is- sues . . . We know that parents, grandparents, foster parents, and other caring adults can play a major role in helping young people avoid substance abuse . . . the key is talking with our children often and early. (p. 1)

Latino parents' alcohol and drug use behaviors have also been stud- ied as predictors of their children's alcohol and drug use behaviors. Geroerer and De La Rosa (1993) reported a relationship between La- tino parents' illicit drug use and children's use. Barrera Jr. and Li (1993) noted the presence of alcoholism/problem drinking in Latino parents tended to correlate with elevated drinking among their adoles- cent children. Studies on the relationship between substance abuse of parents and subsequent alcohol problems in their children were re- viewed by J. L. Johnson and Leff (1999), and additional research (Barrera Jr. & Li, 1995; Barrera Jr., Li, & Chassin, 1998; Gilbert, Solis, & Mora, 1992) suggested that ethnicity and alcoholism in par- ents were factors influencing the stress of adolescents.

In addition to issues of poor family relationships and parental drug use, some Latino youth are subjected to other factors which may place him or her at risk for drug abuse. Additional risk factors include the in- fluence of friends and social networks, acculturation stressors such as culture shock, intergenerational and acculturation gaps, language barri- ers, and identity conflicts (Rhodes & Jason, 1990; Szapocznik et al., 1989). Vega et al. (1995) discussed the issue that some Hispanic adoles- cents feel themselves caught between family and cultural values and the dominant society's expectations. Some adolescents may turn to sub- stance abuse as one of many coping mechanisms.

Figueroa-Moseley (1998) reported that the degree of acculturation of Latino adolescents was associated with the likelihood of substance abuse; however, individuals who were bicultural or had medium accul- turation were at the greatest risk for substance abuse. Another study (Perez & Padilla, 2000) concluded that within a few generations, most Hispanic adolescents exhibited a predominant American culture orien-

116 SUBSTANCE ABUSING LATINOS

tation while maintaining some traces of their Hispanic cultural orienta- tion. Although home culture orientation diminished across generations, it did not disappear completely; many adolescents retained allegiance to their Hispanic familial values. These findings suggest that the Latino child's cultural orientation or degree of acculturation is important to as- sess when considering substance abuse prevention.

The main idea underlying a culturally sensitive intervention is that adolescents from different ethnic backgrounds are exposed to a differ- ent set of risk factors, requiring culturally relevant methodologies and programs. Culturally based approaches in relation to Latino family is- sues including substance abuse have been frequently addressed in the literature (Austin & Pollard, 1993; Delgado, 1998a; Flores & Carey, 2000; Gilbert et al., 1992; Glick & Moore, 1990; Kar, 1999; Krestan, 2000; Maddahian, Newcomb, & Bentler, 1988; Mayers, Kail, & Watts, 1993; Sotomayor, 1991; Szapocznik, Amaro, & U.S. Dept. of Health and Human Services, 1995; Vega & Gil, 1998; Vega et al., 1995). In their study of cultural conflicts and problem behaviors of Latino adoles- cents in home and school environments, Vega et al. reported a connec- tion between the parent-adolescent relationship and alcohol use. Their findings alerted professionals to the necessity of designing interven- tions to include family and culture. Researchers (Acosta & Hamel, 1995; Delgado, 1995, 1996) have recommended that practitioners also recognize natural support systems as an important ingredient of His- panic culture and integrate the natural support system into all aspects of alcohol and other drug services practice with this population.

Another issue to consider in planning a comprehensive intervention is Latino family functioning. Sokol-Katz and Ulbrich (1992) were inter- ested in family functioning in terms of the structure of the family. They examined whether family structure, as defined by which or both parents were present, was related to alcohol and drug use among Mexican, Puerto Rican, and Cuban American adolescents. Mexican adolescents living in female-headed households had higher rates of drinking, drug use, and overall risk-taking behaviors (RTBs) than those living with both parents. By comparison, Puerto Rican adolescents living in fe- male-headed households had higher rates of overall RTBs than those living with both parents. In the Cuban adolescent group, family struc- ture was unrelated to RTBs. Figueroa-Moseley (1998) also studied the relationship of Hispanic family structure and the likelihood of sub- stance abuse in children; however, these findings were inconclusive.

The current study did not examine family structure, but did consider detailed patterns in family functioning. The necessity to include factors

Crunkilton, Paz, and Boyle 117

of both family functioning and culture in prevention models for Latino youth has been emphasized in previous studies (Rhodes & Jason, 1990; Szapocznik et al., 1989). Santisteban, Szapocznick, and Rio (1993) pre- sented an overview of conceptual models and practical approaches in working with Latino youth already identified as alcohol and other drug (AOD) users, as well as those considered at high risk for future AOD use, and discussed the role of the family in the behavior problem syndrome. Researchers have made literature available for scientists, decision mak- ers, and service providers to develop a deeper understanding of the role families and culture play in the prevention of alcohol and other drug abuse in Hispanic/Latino American populations (Department of Health and Human Services, 1995). Issues in preventing alcohol and other drug abuse among Hispanic/Latino families, as well as issues of behavior problems among Hispanic/Latino youth have been explored (Department of Health and Human Services). Family-based intervention models have been implemented and tested in high-density Hispanic/Latino communi- ties throughout the United States and Puerto Rico (Delgado, 1998b; De- partment of Health and Human Services, 1995).

An example of a comprehensive urban prevention program includ- ing variables of family and culture wasLa Familia(Hernandez & Lucero, 1996).La Familia's community-based alcohol, tobacco, and other drug (ATOD) prevention program targeted Latino families with high-risk youth age from age 6 to 11.La Familiaattempted to reduce identified risk factors while building on culturally relevant protective factors. During a two year period, the program enrolled over 200 youth and their families through existing community net- works and aggressive outreach; Hernandez and Lucero reported a 92% retention rate and over 80% attendance per session. As a result of the program, families became more willing to discuss ATOD is- sues openly.

Based on the similar vital features of family, culture, and commu- nity, the current study proposed a family intervention as part of a comprehensive rural community-based program calledProyecto Juventud(Youth Project).Proyecto Juventudwas aimed at meeting the prevention needs of rural Latino youth demonstrating high-risk characteristics for substance abuse. LikeLa Familia, Proyecto Juventudproposed a specific intervention to increase family partici- pation, which would hopefully lead to open discussion of ATOD is- sues.

118 SUBSTANCE ABUSING LATINOS

METHOD

Design and Sample

This study incorporated a pre-experimental pretest posttest group de- sign. The family intervention was part of a larger program, which in- cluded an intervention for youth. Participants in the family intervention were selected when their child was randomly chosen for participation in the youth intervention. The sampling frame for the youth intervention was created by several community agencies cognizant of children at risk of substance abuse, in a rural area of the southwest United States. After children were randomly selected from the list, parents were con- tacted. The nature of the program was explained to the parents, and they were asked to sign a consent form in either English or Spanish, in which they gave consent for their child's participation. Parents who gave con- sent for their child to participate created the selection pool for this study. Forty-nine parents representing 49 different families made up the sam- ple by self-selection. Parents were assured that their participation was confidential.

This study tested two hypotheses:

Families in the community who participate in the culturally com- petent interventionProyecto Juventudwill improve their scores on family functioning as measured by the FACES III.

Families in the community who participate in the culturally com- petent interventionProyecto Juventudwill improve their scores on measures of drug knowledge.

The independent variable is the "service offered to families who par- ticipate in a culturally competent intervention." The two dependent variables are "scores on family functioning" and "scores on drug use knowledge."

Measures

The Family Profile Questionnaire was designed by the program eval- uator to generate both descriptive and experimental data. The level of acculturation was measured by the Acculturation Rating Scale for Mex- ican Americans-ASRMA II (Cuellar, Arnold, & Maldonado, 1995). The ARSMA II, totaling 48 items, consisted of a Mexican Orientation Scale and an Anglo Orientation Scale. ARSMA II has two subscales,

Crunkilton, Paz, and Boyle 119

Anglo Orientation (AOS) and Mexican Orientation (MOS). Both subscales have good internal reliabilities, with Cronbach's alpha = .86 and .88 for the AOS and MOS, respectively. ARSMA II appears to have good face validity, and has been demonstrated to have strong construct validity (Cuellar, Arnold, & Maldonado).

Two instruments, the Family Adaptability and Cohesion Evaluation Scale (FACES-III), and a drug knowledge questionnaire were used to generate the experimental data for testing the two hypotheses.

The Family Adaptability and Cohesion Evaluation Scale, or the FACES III (Olsen, 1986) has become the benchmark measuring instru- ment for family functioning (McClanahan, 1998). The history preced- ing the development of FACES III goes back to the Timberlawn Institute in Dallas, Texas, which was the first organization to produce such research on variables of family functioning (Lewis, Beavers, Gosset, & Phillips, 1976). The institute initiated development of rating scales to measure dimensions of family functioning such as structural organization, power, individuation, expression, and how members deal with separation (Lewis et al.).

FACES III (Olsen, 1986) measures three family qualities: cohesion, adaptability, and family type.Cohesion,measured by 10 questions us- ing a Likert-type scale, deals with the parents' perception of how cohe- sive they and their children are, as a group.Adaptability, also measured by 10 items using a Likert-type scale, deals with the parent's perception of how adaptable they and their children are, as a group. These two scores are combined and averaged, to calculate aFamily type, which in- dicates the degree of healthy balance.

FACES III has total instrument alpha of .68, which is considered fair internal consistency (Olsen as cited in Fischer & Corcoran, 1994). The subscales' internal consistency scores are .77 for cohesion, and .62 for adaptability. Five week test-retest correlation was reported to be .83 for cohesion and .80 for adaptability. FACES III appears to have good face validity, and a number of studies have shown that the previous version, FACES II, had fair known-groups validity. FACES III has been tested on 2.453 adults and 412 adolescents (Fischer & Corcoran).

The drug knowledge questionnaire consisted of 20 "true or false" questions referring to a range of substances, including cigarettes (nico- tine), alcohol, marijuana, inhalants, barbiturates, and cocaine. The test appears to have good face validity. An example of an item about ciga- rettes was "Smoking cigarettes is harmful to the unborn child." An ex- ample related to cocaine was "Regular cocaine inhalation rarely causes addiction."

120 SUBSTANCE ABUSING LATINOS

Data Collection

The supervising agency forProyecto Juventudemployed bi-lingual social workers to collect data. Workers were trained in the basic tech- niques of administering questionnaires, including instruction in confi- dentiality, maintaining neutrality, preserving the integrity of the data, and the time frame for the study. The pretest Family Profile Question- naire was administered to the parents during the first family camp. The questionnaires were again administered to the parents at the final family camp approximately 12 months later, for the posttest.

Intervention

The intervention was carried out by a team of social workers who were assisted by volunteers via two main strategies. One strategy was family camps (campamentos) and the other was family support groups (plticas familiares). The intervention, including culturally appropri- ate themes, focused on activities designed to involve parents in learn- ing about substance abuse prevention, to support family functioning, and to increase the families' protective factors. The educational compo- nents of the entire intervention included: (a) tobacco use, (b) HIV/AIDS, (c) alcohol, (d) other drugs, (e) drug related domestic violence, (f) par- ent/child relationships and (g) teacher/parent relationships.

Family Camps

Tenfamily campsin the year were held at various sites. The length of eachfamily campvaried according to the site. Indoorscampswere held in the local community; these lasted a few hours in the evenings or on a weekend day. Outdoorfamily camps,requiring travel and camping for a duration of two to four days, were held at southwest state and national parks, such as the Grand Canyon, Mogollon Rim, Canyon de Chelly, Sedona, and Catalina.Family campswere designed to provide the youth and their parents the opportunity to get away from their usual home set- ting which allowed for an arena facilitating more open discussion. For the camping trips, all families were invited to attend and supplied camp- ing equipment. Families who chose to attend then traveled by car cara- van, led by staff, to a family camping site which had been pre-arranged by the staff. After arriving at the site, families and staff proceeded with camping activities: setting up the camp, preparing meals, and participat- ing in recreational activities such as hiking, swimming, games, and

Crunkilton, Paz, and Boyle 121

nightly campfires. In addition to these activities, interactions were structured to give both parents and their children opportunities to discuss issues related to alcohol, tobacco, and other drugs (ATOD) and to work on the families' issues such as decision making and problem solving. Another focus was on how families could develop proactive healthy lifestyles to enhance their protective factors. Often, these discussions occurred during campfire talks, as well as part of skits that were per- formed.

Family Support Groups

Held every three to four weeks in their community, family support groups met for a total of 13 meetings in the year. The support groups lasted 3.5 hours. Churches and homes were utilized for locations of the support groups. It was believed that support groups held in a setting that is familiar to families would promote participation. The topics included basic information about ATOD use and abuse, as well as problem solv- ing techniques and mutual support.

DATA ANALYSIS AND RESULTS

Data collected by the evaluation team was processed using the Statis- tical Package for the Social Sciences (SPSS, Version 10.0). Descriptive statistics were calculated to determine demographics, drug use, and ac- culturation. An independent sample t-test was conducted to determine whether there were significant differences between the mean scores for the pretests and the posttests of the FACES III. Descriptive statistics were also compiled for the pretest and posttest of the drug knowledge inventory.

Participant Characteristics

A total of 49 parents were interviewed at pretest. Each of the parents had one or more children participating in the project. The average par- ent age was 38.7 with a range of 27 to 65 years of age. The breakdown of ethnicity was as follows: 39 Mexican Americans or 79%; 7 African Americans, or 14%; 2 Native Americans, or 5%; and 1 White, or 2%. The mean number of children per household under age 18 was 3.5. The mean parent education was a ninth grade education. All the respondents

122 SUBSTANCE ABUSING LATINOS

were female. Over half, 53%, were married, 14.3% were divorced, 16.3% were separated, 14.3% were single, and 2% were widowed.

Acculturation

The ARSMA II was used to assess orientation to Mexican or Ameri- can cultures. Item scores range from one to five, with one being a more Mexican orientation and five being a more Anglo orientation. Results indicated that parents who participated had a mean score of 1.9, indicat- ing a strong Mexican orientation.

Self-Reported Drug Use

Response to 12 questions assessed the parents' self-reported use of ATOD (N = 49). Table 1 displays the actual percentages of affirmative responses on each question.

Almost 57% of the parents reported that they had tried alcohol, while 26.1% reported use in the last month. Almost 44% of parents re- ported trying cigarettes, and 28.3% reported using cigarettes in the last month. While almost 11% acknowledged that they had tried mari- juana, 2.2% admitted that they had used marijuana in the last month. When asked about cocaine and other drugs, 2.2% of the parents re- ported that they had tried each. For the past month use, no one reported using cocaine, but 2.2% responded that they had used "other drugs." Considering inhalants, parents said they had neither tried nor used this substance.

TABLE 1. Percent of Parents Reporting Having Tried Substance, and Report- ing Last Month Use (N = 49)

Substance Percent of parents who

Percent of parents who reported last month drug use

28.3 26.1 2.2 0.0 0.0 2.2

had ever tried

cigarettes 43.5 alcohol 56.6 marijuana 10.9 inhalants 0.0 cocaine 2.2 other drugs 2.2

substance

Crunkilton, Paz, and Boyle 123

Identification of Other Drug Users

Responses to six questions assessed the parents' perceptions of their friends and families ATOD use (N = 49). Table 2 depicts the actual per- centages of affirmative responses on each question. Parents reported high percentages of their friends and family used alcohol, 73.3% and 71.7% respectively. As far as marijuana use, parents reported that 13.3% of their friends and 17.4% of their family used this substance. In addition, parents reported that 4.3% of their friends used other drugs.

Approximately 20 parents and 68 youth participated in the family camps for a total of 88 unduplicated participants. The average atten- dance for each family camp was 13 persons. An average of 9 parents at- tended each family support group. A core group of 18 parents became regular participants at these family support groups. An average parent attended slightly over three support groups with a total of 14.6 hours at- tended by each parent.

Test of hypothesis one.Mean scores, for both the pretest and the posttests, were compiled for each of the sub-scales of FACES III. The results of independent sample t-tests, using an alpha level of .05 for a one tail test, indicated no differences in pretest and posttest mean scores forCohesion(see Table 3),Adaptability(see Table 4), andFamily type(see Table 5).

TABLE 2. Percent of Parents Reporting They Have Friends and Family Who Use Alcohol or Drugs (N = 49)

Substance Alcohol use by friends 73.3 Marijuana use by friends 13.3 Other drug use by friends 4.3 Marijuana use by family 17.4 Alcohol use by family 71.7 Other drug use by family N/A

Percentage

124 SUBSTANCE ABUSING LATINOS

Test of Hypothesis Two

For the drug knowledge questionnaire, the average percentage of cor- rect responses at pretest and posttest was calculated for each question, and then the percent of change was calculated (see Table 6). Comparing pretest and posttest, the percentage of correct responses increased on 16 out of 20 questions. The overall average percentage of correct responses was 58.8 (N = 49) on the pretest and 68.2 (N = 32) on the posttest. This reflected a 9.4 average point improvement of scores.

DISCUSSION

The parents had a mean score of 1.9 giving them a strong Mexican orientation. This finding was consistent with our expectation that the first generation holds to their culture of origin. This finding is signifi- cant because it indicated that we had a group of "Mexican thinking" par- ents whose concept of drug use did not really include drug abuse. In other words, the parents did not have concepts for drug abuse, and thus, were not really equipped to deal with their children's drug abuse issues. This fact of lack of awareness, or lack of knowledge of the concepts of abuse, not only indicated a high risk population, but the necessity to vary the intervention according to the acculturation level. This study at- tempted to do just that, adjusting the intervention to the acculturation level of the parents. Basic features of the program included the staff be- ing bi-lingual, and all parent questionnaires were available in Spanish as well as English.

TABLE 3. Cohesion Scores for Parents Before and After Intervention Programs (N = 32)

Pretest Posttest Difference 3.75 4.16 -.41 t = -.81, df = 31, p = .21

The cohesion scores are categorized as follows: 1 and 2 = disengaged; 3 and 4 = separated; 5 and 6 = connected; 7 and 8 = very connected.

Crunkilton, Paz, and Boyle 125

TABLE 4. Adaptability Scores for Parents Before and After Intervention Programs (N = 32)

Pretest Posttest Difference 5.1 4.5 0.6 t = 1.09, df = 31, p = .42

The adaptability scores are categorized as follows: 1 and 2 = rigid; 3 and 4 = structured; 5 and 6 = flexible; 7 and 8 = very flexible.

TABLE 5. Family Type Scores for Parents Before and After Intervention Programs (N = 32)

Pretest Posttest Difference 4.44 4.36 .08 t = .18, df = 31, p = .43

The family type scores are categorized as follows: 1 and 2 = extreme; 3 and 4 = mid-range; 5 and 6 = moderately balanced; 7 and 8 = balanced.

Although a research hypothesis concerning parent drug use was not included in the study, the findings were noteworthy. Parent drug use ap- peared to be somewhat low; this may be due to the all female sample that might tend to have lower usage levels than a male or a mixed group. Despite assurance of confidentiality, the factor of "social desirability" may have resulted in participants reporting lower use than the actual facts. Nonetheless, since a relationship between parents' and children's drug use has been reported in the literature, the current findings about the drug usage of parents serves to alert professionals to the potential risk for children's drug use.

There was no evidence to support the hypothesis that the FACES III scores on cohesiveness, adaptability, or family type improved from pre- test to posttest. It can be said that overall, the parents were functioning in the mid-range category. The mid-range category is the approximate

126 SUBSTANCE ABUSING LATINOS

TABLE 6. Percent of Correct Responses on Pretest and Posttest, and Pretest to Posttest Change, for Parents' Responses to Drug Knowledge Questions

#

1

2 3

4

5

6

7 8

9 10 11 12

13 14 15 16 17 18 19 20

Drug Knowledge Question

Frequent drinkers need alcohol to feel good.

Pot helps people remember things.

The effects of pot on growing persons is unknown.

Frequent use of inhalants may cause brain damage.

Barbiturate users can stop without side effects.

When the effects of cocaine are over, the user feels happy.

Pot may make the user feel afraid.

Cocaine does not harm the body of the user.

Smoking crack is safer than sniffing cocaine.

Anxiety reaction refers to the anxiety sensation from smoking pot.

Smoking crack will not produce symptoms of cocaine use.

Alcohol related car accidents are the number one cause of teenage deaths.

Regular cocaine use rarely causes addiction.

Smoking cigarettes is harmful to the unborn child.

Regular use of uppers may cause harm.

Smoking pot improves driving ability.

It is easy to overdose on barbiturates when mixing with alcohol.

Cocaine may produce symptoms of paranoia.

Pot smoke is less cancer causing than cigarette smoke.

Sharing drug needles increases the risk of contracting AIDS.

Percentage correct on

Percentage correct on posttest (N = 32)

63.6

72.7 27.3

100.0

61.3

60.6

30.3 78.8

76.2 50.0 57.6 93.9

78.8 97.0 87.5 72.7 78.8 63.6 21.9 90.9

Pretest to posttest change*

-2.4

5.3 -2.2

6.4

2.6

8.4

1.4 8.6

12.4 13.8 21.4 10.9

3.2 -0.8 0.3 1.8 8.6 8.3 -4.2 58.2

pretest (N 66.0

67.4 29.5

93.6

58.7

52.2

28.9 70.2

63.8 36.2 36.2 83.0

46.8 97.8 87.2 74.2 70.2 55.3 26.1 32.7

= 49)

*Negative numbers indicated that percentage of correct responses went down at posttest, while positive numbers indicated that percentage of correct responses increased.

Crunkilton, Paz, and Boyle 127

mid-point between the two extremes on the continuum of family types, and is typical of a desirable score range for a family.

There were a few factors that could account for the lack of differ- ences from pretest to posttest on FACES III scores. Due to self-selec- tion of the sample, perhaps only relatively well functioning families entered into the study. Also, since pretest scores of these participants were in a desirable range, a major change in posttest scores was not likely for the group. A future study with families known to have less than desirable functioning might better test whether the intervention used affected family functioning. Another related explanation of lack of change is that the intervention may not have been long enough in dura- tion to effect a change in family functioning. Additionally, the statistical test used may not have been adequate in detecting a difference. It would have been preferable to match parents on their pretest and posttest scores. Paired difference scores of individual parents were unavailable for analysis due to lack of a coding system in the data collection phase. Had scores been matched for a stronger analysis, and still pre-post com- parisons failed to reveal significant differences, individual family change could have been investigated as a possible avenue to more effec- tively apply future interventions. A final reason that a change did not emerge from the analysis could be that the instruments, translated from English to Spanish, simply were not meaningful to the respondents or did not measure what they intended to measure. Concepts of family functioning translated to Spanish may lose the original meaning by which they were developed and tested in English. The FACES III has been used previously in Spanish and Puerto Rican populations in the eastern United States and Mexican American populations in Los An- geles (J. J. Paz, personal communication, February 9, 2004). Data on utilization of the translated FACES III, as well as information on its dis- criminate validity, would support assessment of its validity with Latino populations.

The Census Bureau Office of Management and Budget defined His- panic or Latino as "a person of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin regardless of race" (Grieco & Cassidy, 2001). While most participants in the current study identified themselves as Mexican Americans (79%), families from Cuba or South America who are newly arriving in the United States and/or are not yet highly acculturated may have similarities to the study sample.

The data revealed preliminary evidence that a culturally competent approach improved drug knowledge of Latino parents. While the 9.4

128 SUBSTANCE ABUSING LATINOS

average improvement of scores was indicative of a positive trend, this evidence needs to be considered with caution. As in English, Spanish drug terminology varies with locality, and by county/culture of origin. There may be concepts imbedded in the English instrument that when translated to the Spanish version, do not accurately reflect what the in- strument was intended to measure.

However, if drug knowledge of parents can improve by any degree, there is potential for parents to influence and improve their children's drug knowledge and awareness. This potential for influence may be an important key for prevention of drug abuse. TheProyecto Juventudproject is currently being replicated with urban Latino adolescents in Tucson, Arizona. The program known asCorazon de Aztlan(funded by the Center for Substance Abuse Prevention) is under the auspices ofChicanos Por La Causaand has attained similar outcomes (J. J. Paz, personal communication, February 9, 2004). Model program descrip- tions and outcomes are available as resources for social workers inter- ested in learning about and/or implementing a prevention program with Latino families and youth at risk for substance abuse and other related high-risk behaviors (Substance Abuse and Mental Health Services Ad- ministration National Registry of Effective Programs, 2002). As Chavez (1999) noted, parents are vital sources of guidance in the Latino community, and it is imperative that parents use their influence to coun- teract the likelihood that their children will engage in substance abuse. This study illustrates the importance of addressing cultural issues as so- cial workers attempt to effect change in the Latino family.