HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 63:98-106 (2008)
© 2008 The Gerontological Society of America

Longitudinal Course of Substance Treatment Benefits in Older Male Veteran At-Risk Drinkers

Faika Zanjani, Shahrzad Mavandadi, Tom TenHave, Ira Katz, Nalla B. Durai, Dean Krahn, Maria Llorente, JoAnn Kirchner, Edwin Olsen, William Van Stone, Susan Cooley and David W. Oslin

1 Graduate Center for Gerontology, University of Kentucky, Lexington.
Departments of 2 Psychiatry (Section of Geriatric Psychiatry)
3 Biostatistics, University of Pennsylvania, Philadelphia.
4 Philadelphia Veterans Affairs Medical Center (VAMC), VISN 4 Mental Illness, Research, Education and Clinical Center, Pennsylvania.
5 College of Medicine, Psychiatry, University of Illinois, Chicago.
6 Chicago VA Medical Center, Illinois.
7 William S. Middleton Memorial Veterans Hospital, Madison, Wisconsin.
8 Department of Psychiatry, University of Wisconsin School of Medicine and Population Health, Madison.
9 Department of Psychiatry, University of Miami, Florida.
10 University of Arkansas for Medical Sciences, Little Rock.
11 Miami VAMC, Florida.
12 Office of Mental Health Services, U.S. Department of Veterans Affairs, Washington, DC.
13 Office of Geriatrics and Extended Care, U.S. Department of Veterans Affairs, West Palm Beach, Florida.
14 Philadelphia Center of Excellence for Substance Abuse Treatment and Evaluation (CESATE), Pennsylvania.
15 Department of Psychiatry, Center for Studies on Addiction, University of Pennsylvania, Philadelphia.

Address correspondence to Faika Zanjani, PhD, University of Kentucky–Gerontology, 306B Wethington Health Science Bldg., 900 South Limestone, Lexington, KY 41094. E-mail: f.zanjani{at}uky.edu


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Background. This investigation aims to determine the 12-month drinking trajectory of older at-risk drinkers in treatment. Furthermore, the drinking trajectory between at-risk drinkers who had met the threshold suggestive of alcohol dependence (problem at-risk drinkers) and those who did not meet this threshold (nonproblematic at-risk drinkers) were compared.

Methods. This investigation is a component of the PRISM-E (Primary Care Research in Substance Abuse and Mental Health for the Elderly) Study, a multisite randomized trial comparing service use, outcomes, and cost between Integrated (IC) versus Enhanced Specialty Referral (ESR) care models for older (65+ years) adults with depression, anxiety, and/or at-risk alcohol consumption. This investigation focuses only on at-risk drinkers, generally defined as exceeding recommended drinking limits, which in the case of older adults has been classified as consuming more than one drink per day. Two hundred fifty-eight randomized older at-risk drinkers were examined, of whom 56% were problem drinkers identified through the Short Michigan Alcohol Screening Test-Geriatric version.

Results. Over time, all at-risk drinkers showed a significant reduction in drinking. Problem drinkers showed reductions in average weekly consumption and number of occurrences of binge drinking at 3, 6, and 12 months, whereas nonproblematic drinkers showed significant reductions in average weekly consumption at 3, 6, and 12 months and number of occurrences of binge drinking at only 6 months. IC treatment assignment led to higher engagement in treatment, which led to better binge drinking outcomes for problem drinkers. Despite significant reductions in drinking, approximately 29% of participants displayed at-risk drinking at the end of the study.

Conclusions. Results suggest that older at-risk drinkers, both problem and nonproblematic, show a considerable decrease in drinking, with slightly greater improvement evidenced in problem drinkers and higher engagement in treatment seen in those assigned to IC.

Key Words: Alcohol • At-risk drinking • Problem drinkers

UNTREATED at-risk drinking in elderly persons [defined as exceeding recommended drinking limits, for older adults, more than one drink per day (1,2)] is a health care burden that increases medical complexity and costs for both patients and society as a whole (3). The U.S. Census Bureau estimates that, by the year 2020, 18% of the population will be older than 65 years compared to the current 12%, as a consequence of baby boomers successfully reaching old age (4). With the percentage of older persons nearly doubling over the next 15 years, there will be an estimated 2-fold increase in the prevalence of older adults in need of substance treatment, with excessive alcohol consumption being one of the most common substance concerns (5). Furthermore, for older adults, at-risk drinking is an important health consideration, because it has been known to interfere with the treatment process for existing medical conditions (6), can lead to new physical and emotional difficulties (7,8), and can impair quality of life in old age (9).

To efficiently manage at-risk drinking among the growing elderly population, evidence-based and more accessible alcohol prevention and treatment programs need to be developed (10). With respect to the clinical needs of older Americans, the Primary Care Research in Substance Abuse and Mental Health for the Elderly (PRISM-E) Study was primarily developed to improve access to treatments for mental health and/or substance abuse in older adults (11). The PRISM-E models of clinical care take a different approach from that of customary models of care in that: (i) specialized referrals based on excessive substance use and mental health needs are offered and (ii) preventive approaches to alcohol abuse and/or dependence are implemented by targeting at-risk drinkers for treatment. Earlier results from the PRISM-E trial have demonstrated no differences in drinking outcomes between the two treatment arms (Integrated [IC] vs Enhanced Specialty Referral [ESR]) (12) and higher participant attendance in the IC treatment arm (13) at 6 months after baseline.

We extend the results of the initial studies in two ways. First, we are interested in the sustained effects in drinking over a 12-month period; second, we are interested in the differentiating treatment effects among participants who scored at levels suggesting alcohol abuse and/or dependence (problem at-risk drinkers) and those who did not meet this threshold (nonproblematic at-risk drinkers). The identification of problem drinking can serve as an early indicator of alcohol abuse and/or dependence through the use of a consequence-based classification criterion (14). Consequently, meeting at-risk drinking criteria combined with problem drinking represents a more severe spectrum of drinking that can be indicative of greater harmful health consequences (15,16). We hypothesized that: (i) drinking trajectories would differentiate over 12 months between the two treatment arms, (ii) problem drinkers would have worse drinking outcomes than nonproblematic drinkers, and (iii) level of engagement in treatment would affect drinking status. These hypotheses were examined using three (Chicago, Madison, and Philadelphia Veterans Affairs Medical Centers [VAMCs]) of the 10 original sites participating in the randomized multisite study, given that only these sites collected 12-month data.


   MATERIALS AND METHODS
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
The PRISM-E Study is a multisite randomized trial comparing service use, outcomes, and costs in integrated and referral models of mental health and substance abuse care for older persons with depression, anxiety, and/or at-risk alcohol consumption. The study is aimed at improving access to treatments for mental health and/or substance abuse in older adults. Participants received either IC or ESR care. Screening was conducted in each primary care clinic at each site to identify patients with possible depression, anxiety, and/or at-risk drinking. Participants who screened positive and met eligibility criteria on the baseline assessment were randomly assigned to one of the two treatment arms. Randomized participants were then asked to attend three follow-up interviews (at 3, 6, and 12 months) and to keep any scheduled treatment appointments. A more detailed description of the study methods is provided in earlier articles (11–13).

Drinking Classifications
At-risk drinking [defined as exceeding recommended drinking limits; specifically for older adults, consumption of more than one drink per day (1,2)] at screening was operationalized as the consumption of more than seven drinks in a week or more than four drinks in a day (binge drinking), more than twice in the past 3 months. To qualify for randomization into the study, at baseline individuals needed to report drinking > 13 drinks per week for men and 11 drinks per week for women (1.5 times the National Institute on Alcohol Abuse and Alcoholism [NIAAA] recommended drinking level for older adults) or having four or more drinks (binge drinking) on four or more occasions during the previous 3 months (1). A higher threshold than NIAAA guidelines was used to take into account that individuals underreport their drinking (17), to remain within moderate drinking guidelines (18), and to attempt to remain consistent with other primary care intervention studies of older at-risk drinkers (19). The gender differences used for defining at-risk drinking recognize the well-established differences in absorption and metabolism between men and women (1).

Problem drinking has been defined as a more harmful and hazardous consumption of alcohol, compared to at-risk drinkers (1). Problem drinking status (PDS) has previously been determined in older adults using the Short Michigan Alcohol Screening Test-Geriatric Version (SMAST-G), which has been evidenced to have an 85% sensitivity and 97% specificity compared with the Diagnostic and Statistical Manual of Mental Disorders (DSM; 3rd Edition, Revised) diagnoses of alcohol abuse or dependence (14). In this study, the SMAST-G was used to distinguish at-risk drinking behaviors as problem drinking (SMAST-G ≥3) or as nonproblematic drinking (SMAST-G < 3); the selected criteria were based on previous research (14–16).

Sample
The sample included persons 65 years old or older who had a primary care appointment during the study period (March 2000 through August 2001). Eligible participants, randomly selected from these clinic appointments, were given a complete description of the study, and written informed consent was obtained for persons who agreed to participate. Consenting patients who completed the screening and the baseline interview and who met eligibility criteria were randomized to one of two treatment models. For the purposes of this article, only three (Chicago, Madison, and Philadelphia VAMCs) of the 10 original sites participating in the randomized multisite study were included, given that only these sites had funds to collect 12-month data, had a sufficient number of at-risk drinkers, and randomly assigned individuals to the care model. Fifty-one percent of the total multisite sample of randomized at-risk drinkers (N = 258) are represented in this analysis. Participants were then further divided in two groups: problem at-risk drinkers (N = 111) and nonproblematic at-risk drinkers (N = 147), exclusively for analysis purposes. Participants represented in the current analysis had significantly better physical functioning scores (p =.007) and fewer drinks per week (p =.008), with no differences in age, number of binges, mental functioning, PDS, and treatment group assignment distribution at baseline, compared to the original PRISM-E at-risk drinking sample.

Assessments
Baseline and follow up instruments used in the current analysis included a demographic survey and several instruments to assess mental and substance abuse symptoms. The demographic survey included questions concerning gender, age, date of birth, marital status, educational level, financial status, social exposure, and chronic diseases. The Mini International Neuropsychiatric Interview (MINI) (20) modules for mania, psychosis, panic disorder, generalized anxiety disorder, and depression were used to determine target symptomology, and exclusion was determined by meeting criteria for psychosis and mania. An alcohol quantity–frequency scale (21) and the MINI alcohol module were used to estimate drinking levels. The Medical Outcomes Study 36-Item Form (SF-36) was used to measure both physical and mental health global functioning (lower scores indicate greater impairment) (22). Tracking records, maintained by research staff, were used to estimate individual attendance rates at treatment care appointments (a.k.a. treatment engagement).

Treatment Models
Both the ESR and IC arms were required to be functioning for at least 6 months prior to initiating participant enrollment, to be accessible after research protocol termination, and to ensure that treatment appointments occurred within 4 weeks following the primary care provider (PCP) visit. IC included: (i) mental health and/or substance abuse (MH/SA) services colocated within primary care; (ii) verbal and/or written communication about the evaluation and treatment plans between the MH/SA clinician and PCP; and (iii) availability of brief alcohol interventions (BAI) designed for at-risk drinking (23). ESR care included: (i) MH/SA evaluation and treatment occurring in a separate location by licensed mental health or substance abuse professionals; (ii) coordinated follow-up contacts with the primary care clinic if the participant missed the first scheduled visit; and (iii) assistance with transportation. As part of the research protocol, both models of treatment required at the minimum one treatment session, with no maximum or minimum limit on the number of sessions, over a 6-month study period. After the 6-month study period, participants interested in further treatment voluntarily attended treatment services; participants were not required to attend treatment after 6 months.

Study Outcomes
Treatment engagement was evaluated from research staff records of appointment attendance by using both a dichotomous scale (0 = no treatment engagement, 1 = treatment engagement) and a continuous scale representing the total number of treatment sessions attended. Drinking indicators were quantity and frequency of alcohol use during a 7-day window prior to each assessment (continuous scale), the number of binge drinking episodes in the 3-month period before each assessment (continuous scale), and at-risk drinking status (dichotomous scale: 0 = not meeting randomization at-risk drinking criteria, 1 = meeting randomization at-risk drinking criteria). In addition, functional ability was characterized by the mental component score (MCS) and physical component score (PCS) of the SF-36, using a continuous scale. All outcomes, with the exception of treatment engagement, were measured at baseline and at 3, 6, and 12 months postbaseline.

Analyses
The analyses focused on several different types of effects: (i) baseline PDS effects on postbaseline treatment engagement, drinking outcomes, and MCS/PCS; (ii) randomized IC versus ESR care effects (treatment care assignment effects, Tx) on postbaseline treatment engagement, drinking outcomes, and MCS/PCS; (iii) the interaction between treatment care effects and PDS effects on postbaseline treatment engagement, drinking outcomes, and MCS/PCS; and (iv) the interaction between baseline problem drinking and time-varying treatment engagement effects (adjusted for treatment care effects) on postbaseline drinking outcomes and MCS/PCS.

The results for (i), (ii), and (iii) above, unadjusted comparisons of sample characteristics and group differences were based on t tests and analyses of variance for continuous outcomes and chi-squares and logistic regression tests for dichotomous outcomes. Only complete cases (those with complete available data) were analyzed for these results.

For the multivariate results based on (iv) above, longitudinal random effects linear modeling procedures were used to estimate effects of treatment engagement and problem drinking on postbaseline drinking outcomes and MCS/PCS. Treatment care (IC vs ESR), visit (6 and 12 months vs 3 months), and PDS (problem vs nonproblematic) were treated as dummy variables, with referral, 3 months, and nonproblematic drinking serving as the reference groups, respectively. Treatment engagement during the interval between research visits was coded as a dummy variable (engaged vs nonengaged with nonengaged serving as the reference level). In this higher-order model we did not include baseline data because of the lack of variability in a primary factor, treatment engagement (e.g., treatment engagement was a nonentity-variable at baseline and was monitored only after random assignment to treatment care). Furthermore, to adjust for unmeasured confounding of the treatment engagement–outcome relationship, we used a random effects instrumental variable approach [Small and colleagues (24)]. In this procedure, the randomized assignment to IC or ESR care served as the instrumental variable to control for unmeasured confounders, and thus was not entered directly into the equation for the outcome models. Rather, treatment engagement was regressed on randomized assignment using a logistic model. The residuals from this logistic model were included as a term in the outcome models to adjust for any unmeasured confounding of the effect of treatment engagement on the outcome. Instead of presuming no unmeasured confounders, this approach assumes that any intent-to-treat effects of IC versus ESR care were due to patients engaging in IC. That is, there was no other path for IC to impact the outcome such as raising the sensitivity of practice staff or providers in caring for depressed or anxious patients. When it was necessary to reduce non-normality of continuous outcomes such as drinking, log10 transformations were conducted for analysis. Each longitudinal, random-effects model was adjusted for patient and practice clustering. Models included main effects and all possible two- and three-way interactions, depending on the number of factors under consideration. Time was treated as a categorical variable, with three dummy variables (i.e., 0/1) representing visits 3, 6, and 12. Baseline served as the reference time point. The intercept in these models represent the average level of the outcome (log transformed) when all other variables in the model were equal to 0. All analyses were performed in SAS version 9.0 (SAS Institute, Cary, NC). Longitudinal, random effects models were analyzed using SAS PROC MIXED.


   RESULTS
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Sample Characteristics
Table 1 describes sample characteristics for the total sample and t test (continuous variables)–chi-square (categorical data) comparisons of PDS (problem drinkers vs nonproblematic drinkers). Nonproblematic drinkers (ND) were more likely to be white (p <.001) and married (p =.01), when compared to problem drinkers (PD). Conversely, PD were more likely to report being financially limited (p =.003), having a dual diagnosis of at-risk drinking and depression (p <.001), being a smoker (p =.005), having a physical and mental functional impairment (p =.01; p <.001), experiencing alcohol abuse and/or dependent behavioral-like patterns (p <.001), and having higher levels of chronic disease (p =.001) and alcohol binges (p =.02). Gender was not analyzed because 100% of the randomized VAMC participants were men. There were no additional significant demographic differences by treatment care assignment (Tx) or interaction effects differentiating sample demographics between problem drinkers.


View this table:
[in this window]
[in a new window]

  		Table 1. Sample Demographics at Baseline for Specified Sites: Chicago, Madison, and Philadelphia, comparing Problem Drinkers (PD) and Nonproblematic Drinkers (ND).

 
Follow-Up Rates
Approximately 80% (total sample range: 85%–79%) of randomized participants were able to attend follow-up research appointments, with no differences in participation rates based on treatment care assignment. However, there was a PDS effect at 6 months (p =.04) and 12 months (p =.04), indicating that at those time points PD (77% vs 86%; 77% vs 88%) were less likely to complete their research interviews, as compared to ND.

Table 2 depicts treatment engagement rated over 12 months, using a logistic regression approach. Overall results showed that participants in IC treatment had higher levels of treatment engagement with no PDS effect. The treatment effect on treatment engagement existed at 3 months (p =.04) and 6 months (p =.03), but not at 12 months. Participants assigned to IC were more likely to attend at least one visit from baseline to 3 months and from 3 months to 6 months.


View this table:
[in this window]
[in a new window]

  		Table 2. Integrated/Referral Assignment (Tx) and PDS Effects on Percentage of Treatment Engagement Over 12 Months.

 
Table 3 depicts mean changes in drinking and functioning scores (MCS, PCS) from baseline to the 12-month follow-up using an analysis–of-variance approach. There was a PDS effect for binges (p =.03), indicating that PD showed greater decline in binges (mean range = –10, –18) as compared to ND (mean range = –3, –7). There was also an interaction effect for PCS indicating that problem drinkers in IC care (p =.02) showed nominal PCS improvement (mean improvement =.97), whereas ND in ESR showed nominal PCS improvement (mean improvement =.78). Changes in number of drinks and MCS did not differ between groups. It is important to note that, despite significant reductions in drinking, mean drinking and binge rates for both nonproblematic and problem drinkers at 12 months were above NIAAA recommended levels for older adults, and the binge levels exceeded the randomization criterion levels used in this study.


View this table:
[in this window]
[in a new window]

  		Table 3. Integrated/Referral Assignment (Tx) and PDS Effects on 12-Month Mean Change in Drinking and Functioning.

 
Table 4 depicts the proportion of at-risk drinkers (based on randomization criteria) at each research stage using a logistic regression approach. There was an interaction effect (p =.03) for at-risk drinking at 3 months indicating that, at this time point, fewer problem drinkers than NP were drinking excessively in IC (22% vs 34%), whereas, conversely, more problem drinkers than ND were at-risk in ESR (43% vs 25%). It is important to note that approximately 24%–38% of the sample still met the study's at-risk alcohol consumption criteria at 12 months.


View this table:
[in this window]
[in a new window]

  		Table 4. Integrated/Referral Assignment (Tx) and PDS Effects on the Proportion of At-Risk Drinkers at Each Research Stage.

 
Multivariate Results
Longitudinal random effects models examining time, treatment, and PDS effect on drinks per week, binge drinking, MCS, and PCS, indicated trends for reduction in drinking over time irrelevant of treatment care assignment. ND showed significant reduction in drinks per week at 3, 6, and 12 months and binge drinking at 6 months (p <.05); problem drinkers showed reduction in both drinking and binge drinking at 3, 6, and 12 months (p <.01). Figures 1 and 2 depict change in drinking and binge drinking from baseline to 12 months. No significant treatment care assignment effects or changes in MCS and PCS were indicated.


Figure 01
View larger version (25K):
[in this window]
[in a new window]

  		Figure 1. Changes in drinks per week at baseline and 3, 6, and 12 months comparing problem and nonproblematic in integrated care (IC) and enhanced specialty referral care (ESR) treatment assignments. ND = nonproblematic drinkers; PD = problem drinkers

 

Figure 02
View larger version (22K):
[in this window]
[in a new window]

  		Figure 2. Changes in binge drinking at baseline and 3, 6, and 12 months comparing problem and nonproblematic drinkers in integrated care (IC) and enhanced specialty referral (ESR) care treatment assignments. ND = nonproblematic drinkers; PD = problem drinkers

 
Table 5 displays the results from the longitudinal random effects models assessing the effects of baseline PDS and postbaseline time-varying treatment engagement (continuous level), accounting for treatment care assignment, on postbaseline drinking outcomes and MCS and PCS. Time-varying number of drinks per week was treated as the first outcome. The resulting model revealed that, relative to3-month drinking levels, baseline ND who did not engage in any treatment visits from baseline to 3 months, showed significant declines in number of drinks at 12 months (p =.04). However, this effect was different for PD, who had less of a decline in number of drinks from 3 to 12 months compared to ND (p =.03).


View this table:
[in this window]
[in a new window]

  		Table 5. Multivariate Analysis of Alcohol Outcomes Over 12 Months as a Function of Problematic Drinking Status (PDS), Treatment Group*, and Treatment Engagement.

 
Next, time-varying number of binge episodes per month was considered as the outcome. Findings again demonstrated that ND who did not engage in treatment visits from baseline to 3 months had a significant decrease in binges from 3 to 12 months (p =.01). This effect, however, varied by PDS and treatment engagement; PD who engaged in treatment visits between 6 and 12 months had a significantly greater decline in binge drinking levels from 3 to 12 months relative to participants who were ND and participants who did not engage in any treatment (p <.001).

The final two sets of analyses examined changes in MCS and PCS scores over time. There were no significant changes in either score over time. Furthermore, neither treatment engagement nor PDS emerged as a significant predictor of change in mental or physical health component scores over time.


   DISCUSSION
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
Findings indicated that older adults showed significant reduction in drinking through 12 months; at the end of the study, approximately two-thirds of the sample no longer met the study's at-risk drinking criteria. Further analysis indicated differentiated patterns of drinking reduction among problem drinkers and nonproblematic drinkers. Although problem drinkers showed heavier binge patterns at the start of the study, problem drinkers appeared to equally benefit, if not more, from the interventions aimed at reducing alcohol consumption. Whereas there were no effects indicating better outcomes for either treatment model, there was evidence of higher treatment engagement in the IC model, as similarly evidenced in earlier works (12,13). However, participation in treatment improved binge drinking reductions for problem drinkers, but treatment engagement appeared to have minimal effect on reductions in the number of drinks per week, regardless of treatment condition. Furthermore, there was no evidence that the implemented treatment models were successful at improving physical or mental functioning.

The current study sought to extend this work by examining the extent to which group differences in severity affected outcomes among at-risk older drinkers. Generally at baseline, problem drinkers in this investigation showed higher binge levels, without any actual difference in the number of drinks per week. This finding implies that, beyond logical differences in a higher level of alcohol-related severity, there was a difference in the level of alcohol consumption. Specifically, in addressing potential group variations in PDS on outcomes, this research investigation was able to show positive 12-month trajectories in both problem and nonproblematic older at-risk drinkers in treatment, with comparably higher and consistent levels of reduction in binge drinking among problem drinkers.

Both treatment models were successful at reducing drinking in both problem and nonproblematic at-risk drinkers. However, it is important to note that, despite significant reductions in drinking, mean drinking and binge rates for both nonproblematic and problem drinkers at 12 months were above healthy levels, and approximately one-third of the sample still met the study's at-risk alcohol consumption criteria. This finding may indicate the need to adapt treatment for those not reducing their alcohol use, who struggle to maintain initial improvements, or who have difficulty maintaining and achieving healthy drinking levels. In this study there is considerable evidence that veterans who simply enroll in a substance abuse treatment program, regardless of whether it is within integrated or referral care, show improvement in drinking. This finding may allude to the idea that physicians generally paying attention to the drinking patterns of their older patients, or the simple aspect of monitoring drinking, can significantly improve drinking patterns to relatively healthier levels. Furthermore, those at-risk drinkers who were assigned to an IC treatment program become more engaged in treatment than do those randomized to an ESR group. Results do not show that such engagement generally results in positive outcomes, but there was some evidence that it can help improve certain outcomes for a more severe segment of the treatment population (e.g., binge levels in problem drinkers).

Despite seeing improvements in drinking, we did not see improvements in mental and physical functioning. However, based on the current treatment design, we cannot determine whether the absence of substance treatment would naturally result in a decline of mental and physical functioning in older at-risk drinkers. In such a circumstance, these treatment models may have prevented inevitable decline in mental and physical functioning, but such an assumption needs to be explored empirically. Although this treatment trial was able to show improvements in drinking patterns, improvements in functioning may be a consequence of drinking patterns unseen in the time frame that was investigated. However, immediate benefits to functioning may be seen in substance treatments that incorporate management of mental and physical functioning components.

This investigation represents a new research phase in geriatric alcohol treatment. Formerly, the need for geriatric alcohol screenings and treatment was established and promoted by the Substance Abuse and Mental Health Services Administration (SAMSHA) (1) and the U.S. Preventive Task Force (9). When alcohol consumption concerns were recognized in the elderly population, the logistics and the efficacy and feasibility of alcohol treatment were examined. Thus, a range of alcohol-related treatment research has been considered: integrative care models (25,26), brief interventions (19,27–32), treatment engagement and adherence (33–35), psychosocial interventions (36–38), and pharmacotherapy (39–41). The treatment presented in this investigation represents a combination of IC model using brief-intervention alcohol treatment for older adults. Although this combination of treatment for older adults is not commonly found in the literature, there is an indication that integrative care treatment models have reduced alcohol consumption (25,26) among adult populations of all ages with substance use disorders showing significant improvement in access and abstinence rates. In contrast, brief intervention treatment has been shown to reduce drinking specifically among older adults (19,27,28,30–32). Therefore, the results of this study indicate that an IC model using brief-intervention alcohol treatment can be successful at reducing drinking among older adults. With regard to change in functioning, Saxon and colleagues (26) also observed an absence of PCS change, but unlike our results theirs did include a slight MCS change as a result of substance treatment. This difference in findings may have been the result of their more heterogeneous sample of all aged adults characterized with any substance use disorder. Other reports that have indicated psychosocial improvements consisted of younger and/or mixed-aged samples, sometimes with more severe drinking symptomology (e.g., abuse and/or dependence), using pharmacotherapy (42,43), payee programs (44), integrated mental health and substance abuse programs (45), and cognitive behavioral therapy (46) treatment methods. Therefore, more treatment research is needed that can show improvements not only in alcohol consumption, but also in functioning, specifically among older adult at-risk drinkers.

When considering the results of this investigation, it is important to consider generalizability. Participants in this investigation only represent patients from the VAMCs. Thus, the findings of this study may have limited generalization beyond older men or patients in VAMC settings. It is also important to acknowledge the absence of a control group. The aim of this trial was to compare the effectiveness of two intervention treatments; therefore, implications can not be compared to that of a treatment as usual group for at-risk drinkers. Future research will be necessary to compare these results with those of a treatment-as-usual group and among nonveteran older adult populations. Finally, it is important to consider the attrition effects in any longitudinal investigation. Although there were high rates of continued research participation across the phases, those represented in the research trial are most likely to value and be committed to the objectives of the research. Also, the observed attrition at follow-up visits may not follow the missing at random assumptions of the longitudinal random effects models. That is, participants who did drop out may have decided to do so because of unobserved drinking behavior at the time of dropout. The longitudinal methods we have used assume that such dependence does not exist; however dropout decisions could have been dependent on observed drinking behavior prior to dropout. Hence, the reported results may be biased due to individual factors causing attrition.

Despite these limitations, there were substantial reductions in alcohol use associated with participation in the trial. These reductions are likely to translate into meaningful improvement in the lives of these participants through the prevention of potential negative consequences associated with at-risk drinking. The results of this study further support the importance of identifying and treating older at-risk drinkers. We feel that this study supports the notion that alcohol consumption should be a topic of discussion during clinical visits of older people. Because improvement in drinking was seen even among older at-risk drinkers who did not engage in treatment, comparisons between the effectiveness of lower threshold treatments that simply identify and monitor alcohol consumption with those of more intensive treatments need to be made. Future studies in this area need to focus on empirically testing treatment algorithms that evaluate PDS to develop the optimal mechanisms for treating at-risk drinking and maintaining healthy drinking levels in older drinkers. Furthermore, this investigation only showed changes in drinking, with minimal changes in mental and physical function. Future studies need to assess how reductions in drinking can be associated with improvements in other behavioral, medical, and psychosocial domains and identify mechanisms for improving mental and physical functioning.


   Acknowledgments
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 
PRISM-E is a collaborative research study funded by the Substance Abuse and Mental Health Services Administration (SAMHSA), including its three centers: the Center for Mental Health Services (CMHS), the Center for Substance Abuse Treatment (CSAT), and the Center for Substance Abuse and Prevention (CSAP). The Department of Veterans Affairs (VA), the Health Resources and Services Administration (HRSA), and the Centers for Medicare and Medicaid Services (CMS) provided additional support and funding. The development of the manuscript was supported by a training grant from the National Institute of Mental Health (NIMH; 5 T32 MH19931-08A1) awarded to David Oslin.


   Footnotes
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 
Decision Editor: Darryl Wieland, PhD, MPH

Received November 1, 2006

Accepted April 12, 2007


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Blow FC. Substance abuse among older adults: Treatment Improvement Protocol (TIP) series. Washington DC: U.S. Department of Health and Human Services; 2004.
  2. Dufour MC, Archer L, Gordis E. Alcohol and the elderly. Clin Geriatr Med. 1992;8:127-141.[Medline]
  3. Dyson J. At-risk drinking and older people. Nurs Older People. 2006;18:32-35.[Medline]
  4. Spencer G. Projections of the Population of the United States, by Age, Sex, and Race: 1988 to 2080, vol Series P-25. Washington, DC: U.S. Department of Commerce; 1989.
  5. Gfroerer J, Penne M, Pemberton M, Folsom R. Substance abuse treatment need among older adults in 2020: the impact of the aging baby-boom cohort. Drug Alcohol Depend. 2003;69:127-135.[Medline]
  6. Onder G, Landi F, Vedova CD, et al. Moderate alcohol consumption and adverse drug reactions among older adults. Pharmacoepidemiol Drug Saf. 2002;11:385-392.[Medline]
  7. Weintraub E, Weintraub D, Dixon L, et al. Geriatric patients on a substance abuse consultation service. Am J Geriatr Psychiatry. 2002;10:337-342.[Medline]
  8. Hurt RD, Finlayson RE, Morse RM, Davis IJ. Alcoholism in elderly persons: medical aspects of prognosis of 216 inpatients. Mayo Clin Proc. 1988;63:753-760.[Medline]
  9. U.S. Preventive Services Task Force. Screening and behavioral counseling interventions in primary care to reduce alcohol misuse: recommendation statement. Ann Intern Med. 2004;140:554-556.[Abstract/Free Full Text]
  10. Schuckit M, Pastor P. The elderly as a unique population. Alcoholism Clin Exp Res. 1978;2:31-38.[Medline]
  11. Levkoff S, Chen H, Coakley E, et al. Design and sample characteristics of the PRISM-E multisite randomized trial to improve behavioral health care for the elderly. J Aging Health. 2004;16:3-27.[Abstract/Free Full Text]
  12. Oslin D, Grantham S, Coakley E, et al. Managing at-risk alcohol use in older primary care patients: results of a randomized trial of referral care vs. integrated care. Psychiatr Serv. 2006;57:954-958.[Abstract/Free Full Text]
  13. Bartels SJ, Coakley EH, Zubritsky C, et al. Improving access to geriatric mental health services: a randomized trial comparing treatment engagement with integrated versus enhanced referral care for depression, anxiety, and at-risk alcohol use. Am J Psychiatry. 2004;161:1455-1462.[Abstract/Free Full Text]
  14. Blow FC, Gillespie BW, Barry KL, Mudd SA, Hill EM. Brief screening for alcohol problems in elderly populations using the Short Michigan Alcoholism Screening Test-Geriatric Version (SMAST-G). Alcohol Clin Exp Res. 1998;22:131A.
  15. Blow F. Short Michigan Alcoholism Screening Test-Geriatric Version (SMAST-G). Ann Arbor, MI: University of Michigan Alcohol Research Center; 1991.
  16. Barry KL, Blow FC. Screening and assessment of alcohol problems in older adults. In: Lichtenburg PA, ed. Handbook of Assessment in Clinical Gerontology. New York: John Wiley & Sons; 1999:243–269.
  17. Searles JS, Helzer JE, Rose GL, Badger GJ. Concurrent and retrospective reports of alcohol consumption across 30, 90 and 366 days: interactive voice response compared with the timeline follow back. J Stud Alcohol. 2002;63:352-362.[Medline]
  18. Masters JA. Moderate alcohol consumption and unappreciated risk for alcohol-related harm among ethnically diverse, urban-dwelling elders. Geriatr Nurs. 2003;24:155-161.[Medline]
  19. Fleming MF, Manwell LB, Barry KL, Adams W, Stauffacher EA. Brief physician advice for alcohol problems in older adults: a randomized community-based trial. J Fam Pract. 1999;48:378-384.[Medline]
  20. Sheehan DV, Lecrubier Y, Sheehan K. The Mini-International Neuropsychiatric Interview (M.I.N.I): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. 1998;59:(Suppl 20): 22-33.[Medline]
  21. Werch C. Quantity-frequency and diary measures of alcohol consumption for elderly drinkers. Int J Addict. 1989;24:859-865.[Medline]
  22. McHorney CA, Ware JE, Lu JF, Sherbourne CD. The MOS 36-item short-from health survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care. 1994;32:40-66.[Medline]
  23. Barry KL, Oslin DW, Blow FC. Prevention and Management of Alcohol Problems in Older Adults. New York: Springer; 2001.
  24. Small D, Ten Have T, Cheng J. Random effects logistic models for analyzing efficacy of a longitudinal randomized treatment with non-adherence. Stat Med. 2005;25:1981-2007.
  25. Weisner C, Mertens J, Parthasarathy S, Moore C, Lu Y. Integrating primary medical care with addiction treatment: a randomized controlled trial. JAMA. 2001;286:1715-1723.[Abstract/Free Full Text]
  26. Saxon AJ, Malte CA, Sloan KL. Randomized trial of onsite versus referral primary medical care for veterans in addictions treatment. Med Care. 2006;44:334-342.[Medline]
  27. Oslin DW, Sayers S, Ross J, et al. Disease management for depression and at-risk drinking via telephone in an older population of veterans. Psychosom Med. 2003;65:931-937.[Abstract/Free Full Text]
  28. Fleming M, Manwell LB. Brief intervention in primary care settings. A primary treatment method for at-risk, problem, and dependent drinkers. Alcohol Res Health. 1999;23:128-137.[Medline]
  29. Fleming MF, Barry KL, Manwell LB, Johnson K, London R. Brief physician advice for problem alcohol drinkers. A randomized controlled trial in community-based primary care practices. JAMA. 1997;227:1039-1045.
  30. Blow FC, Barry KL. Older patients with at-risk and problem drinking patterns: new developments in brief interventions. J Geriatr Psychiatry Neurol. 2000;13:115-123.[Medline]
  31. Gordon AJ, Conigliaro J, Maisto SA, McNeil M, Kraemer KL, Kelley ME. Comparison of consumption effects of brief interventions for hazardous drinking elderly. Subst Use Misuse. 2003;38:1017-1035.[Medline]
  32. Fink A, Elliott MN, Tsai M, Beck JC. An evaluation of an intervention to assist primary care physicians in screening and educating older patients who use alcohol. J Am Geriatr Soc. 2005;53:1937-1943.[Medline]
  33. Satre DD, Knight BG, Dickson-Fuhrmann E, Jarvik LF. Substance abuse treatment initiation among older adults in the GET SMART program: effects of depression and cognitive status. Aging Ment Health. 2004;8:346-354.[Medline]
  34. Kofoed LL, Tolson RL, Atkinson RM, Toth RL, Turner JA. Treatment compliance of older alcoholics: an elder-specific approach is superior to "mainstreaming.". J Stud Alcohol. 1987;48:47-51.[Medline]
  35. Atkinson R, Tolson R, Turner J. Factors affecting outpatient treatment compliance of older male problem drinkers. J Stud Alcohol. 1993;54:102-106.[Medline]
  36. Wiens AN, Menustik CE, Miller SI, Schmitz RE. Medical-behavioral treatment for the older alcoholic patient. Am J Drug Alcohol Abuse. 1982;9:461-475.[Medline]
  37. Schonfeld L, Dupree LW, Dickson-Euhrmann E, et al. Cognitive-behavioral treatment of older veterans with substance abuse problems. J Geriatr Psychiatry Neurol. 2000;13:124-129.[Medline]
  38. Kashner TM, Rodell DI, Ogden SR, Guggenheim FM, Karson CN. Outcomes and costs of two VA inpatient treatment programs for older alcoholic patients. Hosp Community Psychiatry. 1992;43:985-989.[Abstract/Free Full Text]
  39. Oslin D, Liberto JG, O'Brien J, Krois S, Norbeck J. Naltrexone as an adjunctive treatment for older patients with alcohol dependence. Am J Geriatr Psychiatry. 1997;5:324-332.[Medline]
  40. O'Malley SS, Jaffe AJ, Chang G, Schottenfeld RE, Rounsaville B. Naltrexone and coping skills therapy for alcohol dependence: a controlled study. Arch Gen Psychiatry. 1992;49:881-887.[Abstract/Free Full Text]
  41. Volpicelli JR, Alterman AI, Hayashida M, O'Brien CP. Naltrexone in the treatment of alcohol dependence. Arch Gen Psychiatry. 1992;49:876-880.[Abstract/Free Full Text]
  42. Johnson BA, Ait-Daoud N, Akhtar FZ, Ma JZ. Oral topiramate reduces the consequences of drinking and improves the quality of life of alcohol-dependent individuals: a randomized controlled trial. Arch Gen Psychiatry. 2004;61:905-912.[Abstract/Free Full Text]
  43. Kiritze-Topor P, Huas D, Rosenzweig C, Comte S, Paille F, Lehert P. A pragmatic trial of acamprosate in the treatment of alcohol dependence in primary care. Alcohol. 2004;39:520-527.
  44. Conrad KJ, Lutz G, Matters MD, Donner L, Clark E, Lynch P. Randomized trial of psychiatric care with representative payeeship for persons with serious mental illness. Psychiatr Serv. 2006;57:197-204.[Abstract/Free Full Text]
  45. Drake RE, Yovetich NA, Bebout RR, Harris M, McHugo GJ. Integrated treatment for dually diagnosed homeless adults. J Nerv Ment Dis. 1997;185:298-305.[Medline]
  46. Feeney GF, Connor JP, Young RM, Tucker J, McPherson A. Alcohol dependence: the impact of cognitive behaviour therapy with or without naltrexone on subjective health status. Aust N Z J Psychiatry. 2004;38:842-848.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS