The toughest job you'll ever love: A Pacific Northwest Treatment Workforce Survey


      The training, recruitment, and retention of the most qualified professionals for the substance abuse treatment workforce is a crucial underlying strategy in the improvement of client care. Conducted in the year 2000, this survey of substance abuse treatment agency directors and clinical staff in the Pacific Northwest of the United States provides the first empirical estimates of issues surrounding these goals and points to the need for more aggressive strategies if a quality workforce is to be maintained and improved. Results of the survey indicate that there is an average of 25% turnover per year among treatment agency staff, and that the vast majority of this turnover is voluntary and stays within the treatment profession. Agency management and direct service staff differ in their perceptions of the recruitment and retention approaches currently in place in their agencies.


      1. Introduction

      Providers of substance abuse treatment services are as varied as the hues on a color wheel. Operating in residential, hospital, outpatient, detoxification, and day treatment settings, they utilize a variety of treatment models and methods. Historically, they have not been united by a single discipline or widely agreed upon standards of care. Practice guidelines are beginning to be defined, but states still differ significantly in their rules for defining the essential elements of the service delivery system, practice procedures, and minimum qualifications for administrative and direct service personnel. These issues of staff diversity, service variation, and inconsistency of practice standards present significant challenges to preparing a workforce that possesses qualifications consistent with other healthcare professions (
      Center for Substance Abuse Treatment
      Preservice preparation of treatment professionals ranges from those who enter the field with little or no training and learn their craft on the job, to clinicians with academic and practice degrees in health or human services, to specialists with clinical degrees that include coursework and field training in addiction treatment. Nationally, over half (55%) of the academic training programs are in community college or two-year institutions (
      • Edmundson E.
      Significant variation in undergraduate training programs.
      ). Many of these offer courses for undergraduate and certification programs for postbaccalaureate students to help them meet the education and field training prerequisites for addiction counselor certification or licensure. The developmental needs of such a diverse workforce are great, yet there is no single body responsible for accrediting preservice training or continuing education programs (
      The Lewin Group
      In 1999 the Northwest Frontier Addiction Technology Transfer Center (NFATTC), in response to a government request to investigate what appeared to be an addiction counselor shortage, began gathering information on the developmental needs of the region's treatment workforce. The NFATTC is supported by the Substance Abuse and Mental Health Services Administration (SAMHSA) Center for Substance Abuse Treatment to help develop the region's substance abuse workforce and initiate enhancements in education, training, and service delivery systems. Despite its rural nature, the Pacific Northwest of the United States is home to a large number of substance abuse treatment providers and direct service personnel. In Washington, Oregon, Alaska, and Idaho the total number of agencies exceeds 900. Those agencies employ an estimated 6,000 direct service personnel who provide patient care in a broad range of treatment settings.
      The developmental needs of those agencies and their employees as a group are largely unknown to policy makers and state administrators. Studies that have been conducted typically focus on the training needs of existing treatment personnel (
      • Hall M.N.
      • Shaffer H.J.
      • Vander Bilt J.
      Measuring substance abuse treatment provider training needs: Developing an index of training need.
      ) and not on the broader issues of staff turnover, recruitment, and retention. A more comprehensive examination of workforce development needs is essential to strategizing and planning system improvements. The first task in designing a data collection effort was to identify the major components of a potential workforce development plan.
      A number of developmental models were found in response to a Web search using workforce development plan as the key words. A synthesis of information from public sector initiatives resulted in the identification of five strategic planning components: staff retention, personnel recruitment, education system capacity building, reciprocity of professional credentialing, and public image enhancement. Subsequent collaboration with representatives of state government, provider associations, addiction counselor educators, and counselor certification boards in the Northwest region identified concerns relevant to the treatment community in each of the five strategic workforce planning domains.

      1.1 Staff retention

      Turnover is thought to be high throughout the region, but no data are available to confirm or disprove the belief. Treatment administrators have long known that they are expending large amounts of money and time in replacing departed staff. The number of vacant positions is reputed to be high, but that contention needs to be empirically verified. In addition, little is known about the nature of the turnover. Is it due to voluntary resignations, forced layoffs, or the result of disciplinary action? And finally, what agencies are doing to try to retain staff and reduce the turnover rate is unknown. What are the sources of job satisfaction and dissatisfaction among the current workforce? Answers to all these questions are necessary before retention recommendations can be made.

      1.2 Staff recruitment

      Nothing is known about how administrators advertise the availability of job openings. What strategies do they use to attract applicants? What advertising media, word-of-mouth strategies, recruitment flyers, and formal communication channels are utilized? How diverse are those efforts and how satisfied are the administrators with the results? Frequently heard informal complaints center on the inadequacy of the applicant pool. Public testimony at a 1999 Portland, OR, hearing sponsored by CSAT consistently cited counselor shortage as a barrier to expanding the capacity of the existing treatment system (
      Center for Substance Abuse Treatment
      ). Those concerns need verification. In addition, do barriers to entering the treatment field exist that limit the group of available applicants? If so, what is the nature of those barriers and how could they be minimized?

      1.3 Education system capacity building

      If the contention that more addiction counselors need to be trained is true, the question then becomes how best to prepare them. Previous studies have revealed a significant gap between the actual skill of entry level counselors and the proficiency needed for them to work effectively with the difficult issues presented by addicted clients and within the necessarily complex policy and procedural guidelines of treatment organizations (
      • Adams R.J.
      • Gallon S.L.
      ). Beginning counselors need quality education and training prior to entering the field. Currently employed personnel, too, need continuing education to stay abreast of new knowledge and the increasing competency requirements of their work. How well prepared are applicants for vacant positions? What is the educational background of the current workforce? Do existing staff receive adequate clinical supervision and continuing education? Are agencies providing sufficient incentives for staff to engage in ongoing learning activities? These are just a few of the issues that need to be understood before workforce planning can accurately target the education and training needs of the treatment community.

      1.4 Reciprocity of professional credentialing systems

      Today's treatment workforce is credentialed by a number of different national and state certification and licensing organizations. There is currently neither a single national credentialing system nor a set of unified credentialing requirements among the States. Differences from state to state can be so significant that reciprocity in the recognition of credentials can be nonexistent, making it difficult for counselors in one state to qualify for treatment positions in another state. In the Northwest those differences are summarized in Table 1. The credentialing matrix demonstrates considerable variation from state to state. Available levels of certification, required education, work experience and examination boards all differ, even more than the summary table indicates. For students or practitioners wishing to move from one state to another, the credentialing process can be confusing and frustrating. Additional classes, more supervised hours, academic degrees, and re-examination are all potential barriers to a smooth transition from state to state. At another level, the existence of such discrepant systems can limit the potential of addiction professionals to influence public policy and practice standards at the regional or national level.
      Table 1A comparison of addiction counselor certification requirements in the Pacific Northwest
      Levels of certification3321
      Range of education requirements150–270 h270 hours–Masters150 hours–AssociatesAssociates or Higher
      Range of required work experience1–3 years0–3 years5–2 years1,500–2,500 hours
      Supervised practice0–300 hours300–800 hours1,000–4,000 hours850 h
      Examination boardNAADACIC&RCNAADACNAADAC or IC&RC
      Note: NAADAC = National Association of Alcoholism and Drug Abuse Counselors; IC & RC = International Certification and Reciprocity Consortium.
      Sometimes well-trained and experienced professionals in one state are required to enroll in multiple education courses and secure a significant amount of supervised experience before they qualify for a comparable position in another state. Qualification standards can also be confusing, with preferred language and treatment practices varying from state to state. The impact of different credentialing requirements within the Northwest region has long been thought to limit mobility and present barriers to professionals interested in securing positions in treatment settings. The percentage of direct service workers who are certified to practice is assumed to be high but, again, empirical data do not exist. Also, the degree to which a lack of certification serves as a barrier to employment is also unknown. Information about the impact of differing credentialing standards could help the states assess the need for a more reciprocal system within the region.

      1.5 Public image

      Community perceptions of substance abuse and chemical dependency are changing, but persons who are abusing substances or in recovery are still perceived to have caused their illness by many in the general population. Such beliefs often contribute to a perception that addiction treatment is not a true health profession, that addiction is not a real illness, and that addiction should not be included as a treatable condition within health insurance plans. Recent studies (
      Drug Strategies
      Americans look at the drug problem.

      Hazelden National Overnight Poll. (1999). Center City, MN: The Hazelden Foundation.

      ) indicate that public opinion is moving away from these stigmatizing beliefs and toward the conceptualization of addiction as a public health problem as opposed to a crime, and as a disease as opposed to a self-inflicted condition.
      Despite that shift, treatment still receives far less funding than the need for treatment would indicate (
      Institute for Health Policy
      National Center on Addiction and Substance Abuse
      ). Treatment providers tend also to be tainted with this stigma and to be seen as unequal to practitioners in other health and behavioral science disciplines (
      Center for Substance Abuse Treatment
      To what extent does stigma represent a barrier to entering or remaining in the addiction treatment profession? How important is public image to current direct service staff? How much attention is given to recognizing staff for their efforts? Answers to these questions will help determine how much effort should be devoted to enhancing the public image of addiction treatment and the professionalism of current providers as workforce development strategies. With the myriad of issues surrounding the profession, the current study was conceived and designed to address the questions raised above and yield information vital to understanding how best to support and build a stable qualified substance abuse treatment workforce.

      2. Materials and methods

      The RMC Research Corporation teamed with NFATTC to conduct a substance abuse treatment workforce survey to assess the developmental needs of both the treatment system and those working in it. Development, administration, and analysis of the workforce survey was a collaborative effort throughout.

      2.1 Instrument

      The Substance Abuse Treatment Workforce Survey was designed by RMC Research and NFATTC, with guidance and feedback from the Northwest Regional Coordinating Council for Substance Abuse Workforce Development. The Council included representatives of treatment provider associations, educational institutions, state government, and counselor certification organizations. The goal of this collaborative effort was to develop a survey instrument that could assess the developmental needs of the treatment system in each state, as well as the professional development needs of treatment administrators, supervisors, and staff. The final survey consisted of 28 items and covered the characteristics and careers of substance abuse treatment specialists, profiled workforce recruitment and retention practices, and assessed levels of proficiency and interest in future training events relating to 21 substance abuse counselor competencies (
      Center of Substance Abuse Treatment
      ). Response formats varied, including forced-choice Likert scale, multiple response (“check all that apply”) and open-ended. A copy of the instrument is included as a Technical Appendix to this article.

      2.2 Sampling

      A complete list of treatment agencies in the four states comprising the Northwest region was compiled from directories of state-certified agencies obtained from each of the four state substance abuse authorities to represent the target population for the survey. States varied in the degree to which they monitored accuracy and updated information in these directories, but state-obtained lists were universally more current than those available from federal sources. Because a representative sample was needed within each state and the size of the populations differed so much across states, sampling proportions were determined individually for each of the four states in the region. The full census of treatment agencies was included in the samples for Alaska (N = 74 treatment agencies) and Idaho (N = 43). With the larger states, sampling fractions were employed. In Oregon, 293 treatment agencies were identified and a sampling fraction of 1/2 was utilized. In Washington, 522 treatment agencies were identified and a sampling fraction of 1/3 was utilized. Consequently, 164 agencies were surveyed in Oregon, and 181 agencies were surveyed in Washington. Surveys were sent to a total of 462 treatment agencies.
      Each treatment agency received three surveys and three self-addressed stamped envelopes. The sample from each agency was to include the agency director (or designated administrator) and at least two clinicians. The cover letter, addressed to the agency director, indicated that blank surveys could be photocopied if more than three staff were interested in completing the survey. Instructions accompanying the surveys stressed the need for a representative sample in the survey effort, and directors were requested to select two or more clinicians who represented different levels of experience in the field and in the agency. Survey respondents were kept anonymous, although an identification number indicated the respondent's treatment agency for survey follow-up purposes.

      2.3 Response rate

      Data were collected during the winter and spring of 2000. Initial response rates were low, and follow-up calls were made to one third of the agencies in all four states that had not responded (n = 102) approximately 30 days after surveys were sent. Due to resource constraints in conducting the survey, not all nonrespondents were contacted . As indicated in Table 2, the final sample included 197 (43%) treatment agencies and 469 (34%) individual staff responses. All staff responses were included in the analysis presented here. Although the sample size is reasonably large, readers are cautioned not to over-generalize the results presented here (see discussion in section 4.4, at the end of this article).
      Table 2Substance abuse treatment workforce survey returns
      StateNumber of surveys sentNumber and percent of surveys returned
      Alaska7474×3=22235 (47%)82 (37%)
      Idaho4343×3=12913 (30%)30 (23%)
      Oregon164164×3=49282 (50%)200 (41%)
      Washington181181×3=54367 (37%)157 (29%)
      Totals4621,386197 (43%)469 (34%)

      2.4 Analyses

      The analyses conducted on these data were guided by an interest in a descriptive exploration of the workforce issues represented on the survey, as opposed to a testing of specific a priori hypotheses. Statistical analyses were conducted using an array of methods available in the Statistical Package for the Social Sciences (SPSS), Version 10.0 (
      SPSS, Inc.
      ). Responses to all 28 survey items were tested for significant differences across the four states, between the two respondent roles in the agency (management/administrative vs. clinical/direct service) and three levels of education (less than a bachelor's degree, bachelor's degree, and more than a bachelor's degree). These differences were tested using chi-square techniques for those items with categorical responses; and t-tests or analyses of variance for those items with ordinal or interval-level response scales. Bivariate correlations and multiple linear regression techniques were employed to explore predictive models relating selected agency characteristics and workforce turnover rates in responding agencies.

      2.5 Descriptive characteristics of respondent sample

      Descriptive characteristics of the respondent sample are reported in Table 3. Respondents' gender, ethnicity, age, and level of education are consistent with other studies conducted in the Northwest region (
      • Adams R.J.
      • Gallon S.L.
      • Arrasmith D.
      • Pantages T.
      • Gallon S.L.
      ). Only the Alaska sample evidenced differences across the states in these descriptive characteristics. Not surprisingly, its respondent sample was more ethnically diverse, slightly younger, and less experienced than those of the other three states.
      Table 3Descriptive characteristics of survey respondents
      CharacteristicPercent of respondents by state
      Chi-square statistically significant (p < .05).
      Age (> 40)
      Chi-square statistically significant (p < .05).
      Education (B.A. or greater)7476726671
      Certified as alcohol and drug abuse counselor6055707769
      Experience in substance abuse field (>10 years)
      Chi-square statistically significant (p < .05).
      Direct service5048454747
      * Chi-square statistically significant (p < .05).
      In general, the sample was predominantly female, primarily Caucasian, and over 40 years old. Most had a bachelor's or more advanced degree, most held current substance abuse counselor certification, and nearly half had been in the field for over 10 years.

      3. Results

      In general, there were no differences among the four states in their samples' responses to the items on the survey. This was an important finding for regional planning purposes; and it guides the presentation of results to a focus on a regional, rather than state, level of interpretation. There were numerous differences in the respondents' roles in their agencies, however, as well as in their level of education. These two comparative strata were, however, highly correlated. Agency administrators had consistently higher levels of education and training than did their clinical staffs. As this inquiry focused on the current status of the regional workforce, rather than its etiology, the authors will emphasize differences based on respondents' current roles within their agencies, not on their education and training backgrounds which may have had much to do with placing them there.

      3.1 Workforce turnover

      The survey inquired as to the frequency of three sources of staff turnover in an agency: staff being laid off due to insufficient funding, termination related to performance issues, or staff resigning for their own reasons. Of these, resignations were by far the most common source of turnover in an agency. On average, treatment agencies lose 1.75 staff members per year due to their own resignations. In contrast, only about one in three agencies lose a staff member due to layoff (.36 staff per agency) and two of every three agencies terminate one staff per year due to performance issues (.67 staff per agency).
      Combining these estimates provides a picture of the overall turnover experienced in treatment agencies in the Northwest. The three sources sum to a loss of nearly three (2.78) staff per year for the average agency in the region. Given that the average agency size in the respondent sample included about 11 staff, this amounts to an average of nearly 25% turnover per year. Although there was a preponderance of small agencies in the region and in the respondent sample—nearly half of the respondents were from agencies with fewer than six staff—this turnover rate poses a serious level of recruitment effort for large and small agencies alike.

      3.2 Recruitment

      Given this high rate of staff turnover, the survey queried agency administrators and staff as to their recruitment approaches and difficulties in filling these positions. Nearly three of four agencies (71%) acknowledged serious difficulties in recruiting qualified staff. They traced these difficulties first to an insufficient number of applicants who meet minimum qualifications (53%) and second to insufficient funding to fill positions that are open (34%). Other reasons frequently cited included the remoteness or rurality of their location.
      When probed further as to the nature of applicants' insufficient skills, the most frequently mentioned deficiencies were in their experience in substance abuse treatment (37%), their lack of applied or practical skills (36%), inappropriate certification (31%), and inadequate theory-based education (24%).
      Respondents' reflections as to why the applicant pool has such inadequacies were sought in many ways. The first simply asked: why are more qualified and talented individuals not entering the treatment field (i.e., what are the barriers to entering this field)? Fig. 1 displays the prevalence of several specified barriers.
      Figure thumbnail gr1
      Fig. 1Prevalence of perceived barriers to entering the field of substance abuse treatment. N = 469.
      A full 84% of respondents felt that low salary was a barrier in attracting more qualified individuals to the substance abuse treatment field, followed by the heavily demanding nature of the work (extra hours, large caseloads, paperwork). Half of the respondents noted the societal stigma associated with the profession and the consequent lack of respect given those who work in it. The cost of education and competition from other fields were also cited by nearly half the respondents.
      Methods of recruitment used by the agencies in the sample were largely the traditional ones: 82% advertise in newspapers, 58% make individual contacts, and 38% make use of human resource departments. Only one in six agencies availed themselves of modern technology via Web sites (17%) or e-mail networking (16%); and one in 10 (11%) advertised in professional journals.

      3.3 Retention

      Given the perceived difficulties in recruiting qualified staff, it may follow that, once staff have been successfully hired, agencies would place a premium on retaining them and further developing their skills. Among the retention methods most frequently cited were providing direct supervision for staff (80%), in-service training (74%), supporting continuing education (74%), and in-house mentoring (43%).
      Perceptions of agency directors and clinical direct service staff differed significantly in these areas, however. Fig. 2 shows that in three of these four areas, significantly fewer clinical staff than agency administrators perceive these retention efforts going on in their agencies. Whether real or imagined, the discrepancies in these perceptions suggest a gap in understanding or practice within treatment agencies in the Northwest.
      Figure thumbnail gr2
      Fig. 2Prevalence of agency management and direct service staff perceptions of various staff retention efforts in their agencies, direct service N = 65, management N = 119, *p< .01, **p = .001.

      3.4 Methods used to develop staff

      Finally, respondents were asked to indicate what could be done to improve the retention of qualified staff in their positions. Their responses, as displayed in Fig. 3, suggest that low salaries and the paperwork demands on treatment providers may be fueling the turnover rate noted in section 3.1. The distribution is somewhat similar to that of the perceived barriers to entering the field: the most prevalent recommendations being increasing salaries (68%) and reducing paperwork (43%). But other methods, relating more to personal growth and advancement, also were mentioned frequently. The opportunity to participate in ongoing trainings was cited by 40% of the respondents and receiving more individual recognition for their efforts was mentioned by 35%. Finally, unlike the perception of existing efforts at retention, agency managers and direct service staff were in strong agreement as to the ways to improve retention in the field. There were no significant differences in their perceptions of the desired methods shown in Fig. 3.
      Figure thumbnail gr3
      Fig. 3Prevalence of perceived ways to improve staff retention in substance abuse treatment agencies. N = 469.

      3.5 Job satisfaction

      To shift the questioning from a needs and problem focus, respondents were also asked what aspects of their work are sources of satisfaction to them. That is, given all the barriers and challenges discussed so far, why do they typically continue to labor in this field for (on average) the 7 (clinicians) or 13 (directors) years previously reported? Consistent with the earlier finding, salary and benefits were not high on this list. Less than one in three respondents indicated their financial compensation (32%) or career growth (24%) were sources of satisfaction to them. Rather, the more personal and human aspects of the work were cited. Nearly half of the respondents indicated that the personal growth (48%), interactions with clients (47%), collegiality with their coworkers (44%), and their commitment to treatment (43%) were satisfying to them. Again, agency administrators and clinical staff viewed their satisfaction in similar ways. Only interactions with clients differed as a source of satisfaction, with significantly more clinicians than administrators mentioning this.

      3.6 Predictors of workforce turnover

      To move beyond a simple description of recruitment and retention across the region, a multiple linear regression (MLR) analysis was conducted to identify significant predictors of staff turnover in the sampled agencies. Based on the interests and hypotheses of key state and local stakeholders around the region, eight variables from the survey were selected as potentially having some explanatory value as to the turnover rate experienced by these agencies across the region: (a) the proportion of its total funding the agency receives from public sources; (b) its rural vs. urban location; (c) the size of the agency in terms of number of staff; (d) the number of treatment models in use at the agency; (e) the years of experience in the substance abuse treatment field of the agency director; (f) the highest academic degree earned by the director; (g) the gender; and (h) the ethnicity of the director. The analysis used only agency-level responses (n = 197).
      Simple, bivariate correlations of each of the eight predictor variables with the criterion, agency-level turnover, are shown in Table 4, along with the results of the MLR analysis. Only two of the candidate predictors evidence statistically significant correlations with staff turnover: the years of experience of the agency director and the proportion of the public funding in the agency. When evaluated in combination with all other predictors via the MLR, each of these also contributes significantly to the prediction of turnover (i.e., they are not redundant or duplicative influences).
      Table 4Simple correlations and multiple linear regression results predicting staff turnover from selected agency characteristics
      VariableSimple correlation (r) with staff turnoverMultiple linear regression (R = .380
      p < .01.
      Standardized regression coefficientt statistic
      Proportion of public funding−.21
      p < .01.
      p < .01.
      Urban/rural location.00−.02−.25
      Agency size (number of staff)−.07−.03−.34
      Number of treatment models in use−.11−.10−1.31
      Years of experience in treatment−.25
      p < .001.
      p < .001.
      Highest degree earned−.01−.02−.19
      Note. Raw regression coefficients displayed only for statistically significant predictors.
      * p < .01.
      ** p < .001.
      The direction of these predictive relationships indicates that agencies that receive the majority of their funding from public sources and have directors with many years of experience in the treatment field are more likely to experience lower staff turnover. A return to descriptive statistics confirms the consistent relationship of years of experience of the director with staff turnover. Those directors with less than 5 years of experience were in agencies reporting over 50% turnover. Those directors having 5–10 years of experience were in agencies reporting far less turnover: about 30% of its workforce. The trend continues with directors with 11–20 years of experience (22% turnover) and 21–30 years of experience (12%).
      Perhaps equally important to an understanding of this issue are those characteristics that were shown to have no influence on staff turnover. The gender and ethnicity of the director, the size or urban/rural location of the agency, or the degree status of the director evidenced no relationship with staff turnover. None of the simple correlations of these predictors with staff turnover even approach conventional levels of statistical significance.

      4. Discussion

      The results of this regional survey confirmed many prior beliefs and predispositions, but added the empirical evidence and specificity needed for concrete guidance to the field. For example, survey results indicated that staff turnover is indeed a problem. In fact, it averaged a distressing 25% per year across the agencies in our sample—a rate that is more than double that of all occupations (11%) across the nation, substantially larger than that of other health care fields (
      U.S. Bureau of Labor Statistics
      ), and similar to that reported in at least one study of substance abuse treatment (

      McLellan, A. T. (2002, April). Strategies for bridging the gap between science and practice. Paper presented at CSAT-sponsored conference Improving Substance Abuse Treatment: Community-Based Approaches to Practice Innovation, Tampa, FL.


      4.1 Staff turnover

      Several more detailed questions about staff turnover are addressed by the data presented here. For example, does this turnover vary among agencies of different characteristics? Two consistent associations emerged from these data and several preconceived notions were not supported. Agencies led by more experienced directors, and those receiving the majority of their funding from public, rather than private, sources were more likely to experience low staff turnover. The authors speculate that experienced directors may have better ideas on how to support and stabilize their staffs than do less experienced administrators. It is possible that less experienced managers are more focused on the fiscal viability of their agency and less on the personal needs of staff that influence their job satisfaction. At an agency level, public agencies often treat more severe clients and typically pay lower salaries than their counterparts in the private sector. Counselors in these agencies, however, may feel a stronger sense of loyalty to their patients and peers.
      Is this turnover due to budget shortfalls or declines in state or federal funding for substance abuse treatment? It does not appear so, because the vast majority of this turnover consists of voluntary resignations by the counselors themselves.
      Is this turnover responsible for the perceived shortage of counselors in the field? Again, not really, because most of this mobility is simply counselors moving from one agency to another, producing virtually no net loss to the profession. There is a shortage, but it is not simply a shortage of counselors in the profession. Rather, it is a shortage of qualified counselors entering and staying in the field. Most treatment agency directors were highly dissatisfied with the applicant pools they faced when trying to fill a vacancy in their agencies.

      4.2 Staff recruitment

      In the face of high staff turnover, agencies are forced to spend time and resources in recruiting new staff. These data indicate that recruitment efforts tend to favor more traditional methods such as newspaper advertising and word-of-mouth solicitation. More technically sophisticated strategies, such as the use of Web sites and trade publications, are used far less frequently. It could be, however, that the most qualified candidates for clinical positions utilize more contemporary or technical job search tools like the Internet and professional publications. Managers who have difficulty recruiting a qualified applicant pool might want to consider increasing their visibility and using advertising mechanisms favored by potentially more qualified candidates.

      4.3 Staff retention

      Another response to an unfavorable and disruptive staff turnover rate might be to concentrate on strategies aimed at retaining staff. This survey reveals an interesting discrepancy between how managers and direct service staff perceive the availability of professional development opportunities within the workplace. Managers consistently report the availability of continuing education, direct supervision and in-service training. Far fewer direct service staff, however, report awareness of such opportunities. Regardless of whose perception is more accurate, it is important to note that direct service personnel are either not aware of the opportunities available or do not have access to as many professional development activities as managers think. In either case this could be a source of dissatisfaction and may contribute to the turnover rate documented here.

      4.4 Barriers to workforce entry

      The factors identified most frequently as barriers to pursuing a career in addiction treatment were low salary (84%), long hours, large caseloads, and considerable paperwork (68%) that tend to characterize direct service positions in treatment agencies. Improving those same conditions were the most frequently reported suggestions for retaining existing staff. At the same time, when asked about job satisfaction respondents did not include financial compensation or career growth as major contributors. Instead, they placed greater value on personal growth, client interaction, and relationships with coworkers. It could be that, because salaries are low and paperwork and caseload demands high, people who remain in the field do so for more personal reasons.
      Similarly, these data have implications for agencies interested in reducing staff turnover and increasing job satisfaction. Managers might do well to balance their effort between enhancing salary and benefits and meeting the more personal needs of staff in areas of recognition for work performed, professional growth opportunities, and organizational development. These strategies could attract more qualified candidates for open positions and serve to more effectively retain valued personnel within the agency.

      4.5 Study limitations

      Although the findings of this survey have a great deal of interest and implication for the substance abuse treatment community, several limitations of this effort limit the generalizability of its findings. First, although the survey sample was designed to be representative of the treatment agency population across the four Northwest states, the response rate is below standards commonly desired in survey research. Second, the absence of reliable and current populationwide data on agency characteristics that parallel data collected on this survey renders it impossible to gauge the statistical representativeness of the obtained sample at either agency or staff levels. For example, at the agency level, SAMHSA's National Survey of Substance Abuse Treatment Services (N-SSATS, formerly UFDS, the Uniform Facility Data Set) contains a great deal of information on services provided and clients served (and achieves an excellent response rate), but includes almost no information common to the NFATTC Workforce survey with which to make such a sample-to-population comparison. Third, in the absence of accurate population lists of clinical staff within agencies, the authors were unable to select a sample of clinicians independently. Instead, they relied on agency directors to distribute copies of the survey to members of their clinical staff with the instruction to select staff that represent the range of clinicians in the agency. Whether this was done effectively is not possible to determine, again due to the absence of population data on the current workforce. However, presentations of these results by the authors to the single state agency directors, their management staff, and a regional advisory board met with much confirmation of the findings. And, the very absence of these data on a more populationwide basis is a reason to invest interest in these findings and use them to promote the initiation of a definitive agency-level reporting system supported by federal substance abuse agencies.
      Finally, the predictive models explored here are indicative of significant associations between the variables under study; but they do not imply causality, nor can they specify the predictive direction of these relationships. For example, while these analyses show that agencies whose directors have more years of experience in substance abuse treatment experience less workforce turnover, it cannot conclude that it is because of this lengthy tenure that clinicians choose to stay in the agency. And, from a statistical perspective, these analyses cannot determine whether the years of experience of the director influences workforce stability in the agency or the inherent stability of the agency influences the director to remain in the field longer.
      As stipulated by the National Treatment Plan (
      Center for Substance Abuse Treatment
      ) the needs of the substance abuse treatment workforce are paramount in the improvement of a system that cares for this highly vulnerable population. Yet there is little data to document the specific nature and prevalence of these needs (
      • McCarty D.
      The alcohol and drug abuse treatment workforce.
      ). Without empirical data such as that presented here for the Pacific Northwest, efforts at improvement will be guided by, at best, anecdotal evidence and philosophical beliefs and, at worst, by the political whims of the day.
      Figure thumbnail fx1
      Figure thumbnail fx2
      Figure thumbnail fx3
      Figure thumbnail fx4


      This research is supported by Cooperative Agreements Nos. UD1 TI 11649 and UD1 TI 13424-01 with the Substance Abuse and Mental Health Services Administration (SAMHSA) Center for Substance Abuse Treatment (CSAT). The authors are indebted to Drs. Frank Mondeaux and Maureen Newby for their early contributions to this survey effort.


        • Adams R.J.
        • Gallon S.L.
        Entry-level addiction counselor competency survey: National results. Northwest Regional Educational Laboratory, Portland, OR1997
        • Arrasmith D.
        • Pantages T.
        • Gallon S.L.
        Salary compensation among chemical dependency and mental health professionals in the Pacific Northwest. Northwest Frontier Addiction Technology Transfer Center, Salem, OR1999
        • Center for Substance Abuse Treatment
        Changing the conversation: The national treatment plan initiative. Author, Rockville, MD2000
        • Center of Substance Abuse Treatment
        Addiction counseling competencies: The knowledge, skills, and attitudes of professional practice. Author, Rockville, MD1998 (Technical Assistance Publication 21, DHHS Publication No. SMA 98–3171)
        • Drug Strategies
        Americans look at the drug problem.
        Peter D. Hart Opinion Surveys. Washington, DC, Author1995
        • Edmundson E.
        Significant variation in undergraduate training programs.
        in: Frontlines: Linking Alcohol Services Research and Practice. National Institute on Alcohol Abuse and Alcoholism, Rockville, MD2002, November: 7-8
        • Hall M.N.
        • Shaffer H.J.
        • Vander Bilt J.
        Measuring substance abuse treatment provider training needs: Developing an index of training need.
        Journal of Substance Abuse Treatment. 1997; 14: 1-13
      1. Hazelden National Overnight Poll. (1999). Center City, MN: The Hazelden Foundation.

        • Institute for Health Policy
        Substance abuse: The nation's number one health problem. The Robert Wood Johnson Foundation, Princeton, NJ1993
        • The Lewin Group
        Substance abuse training needs analysis. Author, Fairfax, VA1994
        • McCarty D.
        The alcohol and drug abuse treatment workforce.
        in: Frontlines: Linking Alcohol Services Research and Practice. National Institute on Alcohol Abuse and Alcoholism, Rockville, MD2002, November: 1-2
      2. McLellan, A. T. (2002, April). Strategies for bridging the gap between science and practice. Paper presented at CSAT-sponsored conference Improving Substance Abuse Treatment: Community-Based Approaches to Practice Innovation, Tampa, FL.

        • National Center on Addiction and Substance Abuse
        Shoveling up: The impact of substance abuse on state budgets. Columbia University, New York2001
        • SPSS, Inc.
        SPSS for Windows (Version 10.0) [Computer software]. Author, Chicago, IL1999
        • U.S. Bureau of Labor Statistics
        Occupational employment, training and earnings data. 2000 (Available: