Abstract
Background
Nonattendance in alcohol and other drug (AOD) treatment has been a persistent issue for service provision. The study reports on the outcomes of implementing a group intervention, titled Getting Ready for Change (GRFC), as the default entry pathway into an AOD counseling service that aimed to improve initial attendance and retention through reduced wait days and improved clinical capacity.
Methods
Clients of the service (N = 274) were offered either an individual appointment (baseline) between September 2015 and February 2016 or a group-based appointment (intervention) between September 2016 and February 2017. The samples were compared in terms of demographics, principal drug of concern, wait days to initial and follow-up appointments, and attendance.
Results
The implementation of GRFC reduced wait days to initial appointment from 15 to 5 days and improved initial attendance rates by 24%. Wait days to follow-up were reduced from 10 to 8, retention rates improved by 24%. Further, there was an increased service capacity to meet community demand.
Conclusions
A group entry model into AOD treatment is a novel intervention, is easy to implement, improves attendance and retention in treatment, reduces wait days, and enhances clinical capacity.
Introduction
It is well recognized that retention in treatment is a significant predictor of positive treatment outcomes for clients of alcohol and drug other - alcohol and other drug (AOD) services. 1 – 4 Yet, many people who seek treatment for substance use disorders do not attend their first scheduled appointment, and substantial numbers that commence treatment disengage before their treatment is complete. 2 , 5 Nonattendance rates in AOD outpatient settings range between 35% and 50% 2 , 3 , 6 – 8 and impact service efficiency. 2 Attendance rates are mediated by client factors, including type and degree of substance use, demographics, employment, family circumstance, and service factors, including, environment and staff attitudes. 4 , 9 , 10 Time to treatment is critical, 4 , 9 , 10 with longer waiting times to initial appointment interpreted by clients as devaluing their problems and reducing motivation and engagement 4 , 9 in treatment. Conversely, minimal delay (i.e., treatment within 24 hours to 3 days) improves attendance and engagement. 4 , 10 , 11 Aiming to promote access, services have trialed various flexible appointment models, including client-driven scheduling, (“walk-ins”) and “overbooking” (2 clients booked per appointment). 3 None of these fully address issues, however, and their impact on nonattendance is variable. 3 , 8 , 10 Another strategy that may promote attendance is provision of a “menu” of interventions from which clients can select 12 to promote fit with client motivation and expectations. 4 Additionally, orientation to the service and provision of opportunities to explore treatment expectations and problem solve for barriers can improve retention.8
With retention in treatment critical to achieving best outcomes as well as limited resources, services are obliged to act proactively. This study examines the impact on attendance, retention, and clinical capacity with the introduction of a novel group intervention, titled Getting Ready for Change (GRFC), as a default entry pathway for clients into AOD counseling.
The research question to be explored was the following: Does the group intervention improve initial attendance rates, improve retention in treatment, and reduce clinical time loss associated with nonattendance?
Methods
Participants
Participants of the study were adult clients who contacted a government, outpatient, tertiary AOD treatment service in Australia, seeking counseling treatment.
Procedure
Clients were triaged by trained phone counselors. To be deemed eligible for an initial appointment, clients must have met Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5), 13 substance use disorder criteria (moderate to severe), not had an acute medical illness, and been ready and willing to engage in treatment. Data were obtained on clients calling a phone triage service at 2 time-matched periods 1 year apart. The baseline sample was composed of clients triaged by phone and offered individual counseling appointments with Allied Health practitioners, between September 2015 and February 2016. The intervention sample was composed of clients who were triaged by phone and offered the GRFC group appointment between September 2016 and February 2017. Exclusion criteria included people under 18 years, those who stated being unable to attend the scheduled group appointment times, or those who reported being unable to cope with a group setting. Those excluded were provided individual appointments (i.e., standard procedure). At baseline and intervention, clients were not offered a follow-up appointment when the client and Allied Health practitioner agreed that brief interventions provided at the initial appointment had met current needs and treatment was considered complete.
GRFC comprised a semistructured 30-minute psychoeducation and motivational enhancement session overviewing local AOD services, treatment options, self-care strategies, and support systems as well as completing a change plan (modified from Self Management and Recovery Training (SMART) recovery 14 ). The session was followed by a brief 10-minute individual session to assess risk and review the change plan. GRFC was offered twice per week, facilitated by 2 staff and capped at 6 participants per session.
There was no change in service resourcing between sample periods.
Measures
Client demographics, principal drug of concern (PDOC), appointment dates, attendance/nonattendance, and clinical notes were entered, as standard procedure, in the statewide record-keeping system (the database) used by government AOD treatment services. Appointments, the number of group sessions available, attendance, and nonattendance were recorded in the Allied Health Service client appointment calendar.
Clinical time for the baseline and intervention sample interventions was calculated, by the research team, based on allocated diary time. Each individual intervention was allocated at 1.5 clinical hours, based on 1 staff member to provide direct client service and related administration time. Each GRFC appointment (group plus individual components) was allocated 3 clinical hours per session, based on 2 staff providing 1.5 hours each for direct client service and related administration time. For baseline, “lost clinical time” was recorded for any canceled, postponed, or nonattendances. For intervention, when 2 or more clients attended, 2 staff were involved and there was no “lost clinical time” recorded. For group sessions where only one client attended, one of the staff was not required and so half the allocated session time was deemed as “lost clinical time” (i.e., 1.5 hours). For sessions with nil client attendance, the full allocated time (i.e., 3 hours) was deemed as “lost clinical time.”
Retention in treatment was considered achieved when clients deemed eligible for an initial assessment returned for a follow-up appointment. Wait days were measured in 3 ways: “wait days to initial” referred to the days between first phone contact and initial appointment, “wait days to follow-up” referred to the days between the initial appointment and follow-up appointment, and “total wait days” was defined as the wait days between the first phone contact and the follow-up appointment.
Design and analyses
Data were extracted (and subsequently de-identified) from the database on sample characteristics (age, gender, and PDOC) as well as wait days to initial appointment, attendance, whether a follow-up appointment was provided, wait days until the subsequent appointment, and whether this was attended. Data were analyzed using IBM SPSS Statistics 23. 15 Data were coded as per the following: age (continuous), gender (dichotomous; male/female), PDOC (categorical), wait days (continuous; initial, follow-up, and total wait days), and attendance (dichotomous; yes/no).
Analysis involved coding and checking data frequencies, followed by observing the bivariate relationships and calculation of standardized residuals. Following this process and to address the research question, relationships between the variables were investigated. An alpha level of .05 was used for all statistical tests. Also, Cramer's V (øc) and Cohen's d were calculated to provide an estimate of effect size. As suggested by Aron and Aron, 16 Cramer's V of around 0.10 was considered to be a small, around 0.30 a moderate, and around 0.50 or more a large effect size. A Cohen's d of 0.20 was considered a small, 0.50 a medium, and over 0.80 a large effect size. 17 Levene's tests were conducted for homogeneity in t tests, and where this was breached, adjusted values were reported.
Ethics
This project was reviewed by the Royal Brisbane and Women's Hospital, Human Research Ethics Committee (reference: HREC/17/QRBW/361).
Results
Sample description
The baseline sample comprised 117 clients triaged and offered individual counseling appointments. The intervention sample comprised a total of 157 clients who were deemed eligible for service. Of these, 129 met inclusion criteria for the group appointment, with the remaining 28 (18%) not meeting criteria and provided individual appointments (standard procedure). There were no differences relating to age groups between those who attended (M = 38, SD = 9.7) and those who didn't attend (M = 37, SD = 8.9) (t(272) = −0.77, P = .44, d = 0.09). The relationship between gender and initial attendance was not significant (χ2(1) = 0.49, P = .29, øc = 0.04). There were also no significant differences between those who attended and those who did not attend the initial appointment in relation to their PDOC (χ2(4) = 5.88, P = .21, øc = 0.14). In relation to follow-up attendance, there were no differences relating to age groups between those who attended follow-up (M = 37, SD = 9.3) and those who didn't (M = 38, SD = 10) (t(272) = 0.47, P = .64, d = 0.09). The relationship between gender and follow-up attendance was not significant (χ2(1) = 0.32, P = .36, øc = 0.04). There were also no differences between those who attended follow-up and those who didn't attend in relation to their PDOC (χ2(4) = 6.37, P = .17, øc = 0.17) (see Table 1).
The analyses indicate significant differences between the samples on wait days to initial attendance, attendance at the initial appointment, wait days to follow-up, follow-up attendance, and absolute retention. Total wait days from initial contact to follow-up reduced from an average of 23 at baseline to 12 in the intervention period.
The only significant demographic difference found between the samples was gender, with significantly more females in the intervention sample (43.9%) than the baseline sample (20.5%). To explore this further, additional analyses were conducted on the key variables in relation to gender. Wait days until appointment for males (M = 9.9, SD = 7.9) was slightly higher than for females (M = 8.42, SD = 9.2), although this was not significant (t(272) = 1.37, P = .17, d = 0.17). Wait days until follow-up were also not significantly different between the male (M = 8.3, SD = 3.3) and female (M = 8.65, SD = 5.2) groups (t(136) = −0.48, P = .63, d = 0.08). There were no differences in age between males (M = 37, SD = 9.2) and females (M = 38, SD = 9.6) across the sample (t(272) = −0.62, P = .53, d = 0.08). There were also no significant differences between gender and PDOC (χ2(4) = 1.68, P = .79, øc = 0.08) or whether follow-up was offered (χ2(1) = 0.54, P = .33, øc = 0.05).
Clinical time
A total of 43 group sessions were scheduled. Of these, 30 were facilitated as group-based sessions, with 13 groups either having only one or nil clients attend. The number of available appointments, number of allocated appointments, and lost clinical time are shown in Table 2.
Characteristic and comparative analysis between baseline (n = 117) and intervention (n = 157) samples.
Standardized residual: ±1.98.
Clinical time and service capacity.
Discussion
Nonattendance in AOD treatment has been a persistent issue for service provision. 10 Contrary to other literature, this study found age, gender, and/or drug of concern to have little impact on attendance or retention, but it did support wait days being the most influential variable. 4 , 9 , 10 The subject service prior to implementation of the group entry model experienced attendance rates consistent with the literature, with nonattendance rates for initial appointment at 56% and nonattendance for follow-up appointments at 33%, with 26% of clients seeking treatment being retained.
Implementation of GRFC reduced wait days to initial appointment from 15 to 5 on average and nonattendance rates reduced to 31%. The reduction in lost clinical time associated with nonattendance and individual appointment scheduling allowed for service flexibility, with follow-up appointments provided on average within 8 days and nonattendance for follow-up appointments reduced to 18%. Retention rates improved overall, with 50% of clients seeking treatment being retained. There were no significant differences between samples in relation to percentage of clients offered a follow-up appointment, demonstrating that usual business practices for offer of follow-up appointment remained consistent between samples. GRFC allowed both the initial appointment and scheduled follow-up appointment to be provided on average within 12 days of first contact, compared with 23 days at baseline. As reflected by the literature, this may have demonstrated to the client the priority placed on their health concerns and so could have impacted retention. 4 ,9
Service capacity improved due to reductions in lost clinical time and number of appointments that can be offered. Time loss associated with nonattendance at initial appointment reduced, with 97.5 hours lost at baseline, compared with 45 hours at intervention, showing an overall improvement of 52.5 hours. Further, nonattendances impacted less on allocated clinical time in the group model. At intervention, the number of available appointments increased by 169.
The study had some limitations. Although the study did not evaluate the content of group sessions to determine whether this impacted attendance, the intervention was designed on evidence based principles, where clients are informed of their options, actively involved in the treatment planning and matching, and provided with a ‘menu’ of treatment options. 12 , 18 It would be useful to investigate further the influence of group content on retention. Further research could investigate whether the results would be found in specific populations such as youth, older adults, or in other AOD settings. There was a significant gender difference between the groups, although the differences were not related to the outcome variables. Although sampling bias is possible, it cannot be explained, as both samples included all callers to the service. Gender difference in clients attending AOD treatment has been well established. 19 , 20 As GRFC was initiated as a service improvement activity, this study did not collect data on other client or service factors that may have impacted, mediated, or moderated attendance and retention described in the literature. 4 , 9 ,10
This study demonstrates that GRFC, a group entry model into AOD treatment, is a novel intervention that is easy to implement, improves attendance and retention in treatment, reduces wait days, and enhances clinical capacity in a tertiary AOD treatment service. As attendance is one of the greatest challenges to engagement and loss of clinical time, efficiencies must continue to be sought to improve outcomes for clients.
Footnotes
Acknowledgments
The authors would like to acknowledge all clients of the service. The authors would like to acknowledge the assistance of the staff of the Allied Health Service, Metro North Mental Health Alcohol and Drug Service, who assisted in the design and the facilitation of the group program and in data collection. The authors also acknowledge the assistance of De Bonda Luck, psychologist, for assistance with data entry and literature review during her student placement in 2015. Library services and access to databases were provided by Queensland University of Technology, Brisbane, Australia.
Funding
This was a service quality improvement initiative undertaken within existing resources of the Metro North Mental Health Alcohol and Drug Service. In the absence of dedicated funding, this study was conducted with in-kind support of the authors’ employing organization. The organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Author contributions
Kim Sander, [B. Soc. Work], led the research team and provided strategic guidance to implement the change to clinical service delivery. Hollie Wilson, PhD, managed the data analysis and interpretation as well as supported staff of the triage team to transition to and implement the change to service entry. John Kelly, [BA (Psych) Post Grad Dip Psych], co-authored in design, methods and results as well as provided guidance in development of group content, group facilitation processes and trained colleagues to ensure quality and consistency in service delivery. Anthony Bligh, [M. Psych (Clinical)], analysed pre-data, co-authored the literature review and reviewed the final manuscript.