A practical approach to needs assessment for chemical dependency programs

Joumol

0740-5472/88 $3.00 + .OO Copyright 0 1988 PergamonPress plc

of Substance Abuse Treatment, Vol. 5, pp. 105-I 11, 1988

Printedin the USA. All rightsreserved.

NEEDS ASSESSMENT

TECHNIQUES

A Practical Approach to Needs Assessment for Chemical Dependency Programs JOHN M.

MADDOCK,

BS, DENNIS

DALEY, MSW, HOWARD B. Moss,

MD

Comprehensive Alcohol and Drug Abuse Programs, Western Psychiatric Institute and Clinic, University of Pittsburgh, School of Medicine

Abstract- The authors discuss the concept of Needs Assessment within the context of the planning process for chemical dependency treatment programs. The jive basic Needs Assessment approaches are critically reviewed, and their application in specific forecasting models is addressed. Practical guidelines for the clinical planner using Needs Assessment techniques are dticussed, and recommendations made for the type, timing, and frequency of assessment. The authors note the problems inherent in the muhiplicity of methods used by various states, and call for a uniform approach to Needs Assessment . Keywords-Needs Assessment, alcoholism treatment, chemical dependency treatment, treatment planning, epidemiology.

DURING THE PAST several years chemical dependency treatment has undergone significant changes, evolving into a “major industry” (Holden, 1987). The number of profit based drug and alcohol programs has risen steadily between 1979 and 1982 (Lowman, Bertolucci, Sanchez, Reed, & Patterson, 1984). Alcohol related treatment costs are nearly 15% of the total U.S. health bill (Holden, 1987) and health costs continue to rise dramatically. As more health agencies move toward providing chemical dependency treatment, Needs Assessment becomes a more essential part of planning treatment services.

WHAT

IS NEEDS

ASSESSMENT?

Needs Assessment is an integrative process for planning treatment services and identifying potential problems in program conception. Within the chemical dependency field, that translates into evaluating the drug and alcohol problems of the target community and providing appropriate treatment services based upon this “need” determination. Research activities, such as

Requestsfor reprintsshould

be sent to Howard B. Moss, Comprehensive Alcohol and Drug Abuse Programs, Western Psychiatric Institute and Clinic, University of Pittsburgh School of Medicine, Pittsburgh, PA 15218.

conducting surveys or correlating health/social indicators, are an essential component of evaluating the needs of the target population (Kaelber & Noble, 1982; Warheit, Bell, & Schwab, 1977) and provide the data from which to plan suitable, practical treatment. An important distinction has been drawn between “met needs” and “unmet needs” by some authors. (Kearney & Co., 1979; Ford 1985; Kaelber & Noble, 1982). “Met needs” are defined as those chemical dependency problems for which services are active and available. Utilization rates are most often used to characterize “met needs.” There are segments of the population that are not receiving treatment, either because they do not seek treatment or the services they require are not available and these are the “unmet needs.” It is difficult to measure or estimate “unmet need,” and therefore it is an area often overlooked. The planner must be aware of this “silent” demand group, and use methods to indirectly measure “unmet need.” A simple but important factor in understanding Needs Assessment methods is the difference between “need” and “demand.” This is closely related to met and unmet needs. “Need” encompasses all the chemical dependency problems within a community. “Demand” is the expressed desire for treatment. Within this context “expressed desire” is not limited to the client’s request for treatment, but represents any con-

106

J. M. Maddock et al.

cerned party including family, cIergymen or community groups. WHY PERFORM TREATMENT NEEDS ASSESSMENT? Identifying the need for drug or alcohol treatment services within a community is the first step toward providing adequate, effective treatment for impaired individuals residing in that community. Needs Assessment is the process through which these requirements are examined, thereby allowing one to plan for the appropriate treatment services within a community. As more drug and alcohol treatment programs enter an area the “need profile” of the area changes. In fact, the needs profile of any community is in a constant state of flux. Needs Assessment is, therefore, an ongoing process rather than a static one. When a new chemical dependency treatment program is contemplated, many State regulatory agencies require that needs assessment data be included in the application for a Certificate of Need. Program planners should not, however, regard this process as simply an intellectual exercise necessary to get permission to begin program development. Valuable information as to “who, what, when, and where” can be used to maximize both the therapeutic and economic success of the new program. Any target region can support only a finite number of programs offering the same or similar treatment approaches. Needs assessment models may help identify those treatment services that are lacking, as well as those needs that are filled by existing programs. By becoming familiar with the existing treatment programs within the target community, in terms of capacity, types of services offered and similar aspects, the planners will be able to address those needs that remain unmet and avoid redundant treatment services. Needs Assessment can also help to define the appropriate size of the proposed treatment program. The program must be large enough to meet the requirements of the target area without encountering a high vacancy rate. On the other hand, a program that is too small may have a waiting list and lose prospective clients. Needs Assessment can draw the planners attention to any such potential problems. It is then up to the planner to study the question of appropriate size in more detail. The other dimensions of treatment provision that are correlated with size, include financial and staff resources, facility location and community growth or attrition to name a few. One aspect of the assessment process often overlooked is the selection of a suitable location for the treatment program. In planning the location for services it is useful to study the demographics of the target population. The program may be otherwise optimally designed in every aspect and yet fail to pro-

vide services to the target population due to an unsuitable Iocation. The treatment services must be accessible to the prospective clients. Accessibility is important, but location decisions should be more than a question of client access or available space. Additional consideration should be given to the impact of program Iocation on the community at large. For example, locating a chemical dependency treatment program in close proximity to a parochial school might be frowned upon by the community. Similarly, it may be unwise to locate a new alcohol treatment facility next to a drinking estabIishment. Although staff selection is not a precise science, data generated through Needs Assessment can be utilized to expedite the selection of staff that is compatible with the target popuIation. However, simply matching ethnic background, gender and language is not sufficient. Staff selection factors should include evaluation of interpersonal skills, attitudes toward clients and the special issues of the target population (e.g. military veterans, homosexuals, criminal justice system clients). Needs Assessment is therefore, a complex process. Some of the treatment requirements can be assessed by using mathematical and statistical approaches. Other aspects must be considered on quite different levels, using a myriad of human resource skills and techniques. The planner should, therefore, have a comprehensive understanding of drug and alcohol treatment approaches, the demographics of the prospective treatment population and be well versed in personnel management techniques. The ideal program is designed to meet the treatment needs of its target area in the appropriate treatment modality, in an easily accessible location and of the proper size for the community. WHEN TO ASSESS TREATMENT NEEDS Optimally, the initial assessment should be completed before the program is designed or the treatment site is selected and before hiring staff. Needs Assessment is not restricted to identifying the approximate size of the target area’s chemical dependent population or the ideal treatment approach. It should be an integral part of the planning process at every stage, not only at the inception of the treatment program,but on a regular basis. In reality, Needs Assessment is not always one of the first activities addressed when planning a chemical dependency program. Often the program has already been proposed, the program site has been selected and the program is in the midst of starting up. It is also common for administrative staff to be asked to produce a Needs Assessment analysis to justify the existence of the program at board meetings

107

Needs Assessment Approaches or for regulatory agencies’ Certificates of Needs applications. However, assessment as an afterthought is ineffectual and a poor use of resources. Once the program is underway Needs Assessment should only be initiated with the commitment to modify and reorganize if indicated by the assessment (Warheit et al., 1977). It is far more difficult to modify an existing program than it is to redesign one still in the planning stages. NEEDS ASSESSMENT METHODOLOGIES Needs assessment methodologies vary in cost, complexity, time requirements, comprehensiveness, and relative effectiveness. Choosing an approach can sometimes be confusing and discouraging, especially if the planners are not conversant with the various methodologies by which treatment needs can be assessed. Extensive statistics and data are available for alcohol related problems. And while there has been a nationwide focus on the seriousness of the drug problem, it is a more difficult problem to measure due to the lack of demonstrated health, legal and social indicators. The only cited method we found for predicting treatment service levels for drug abuse was that of Ford (1985). This method uses population demographics, drug arrest records and state treatment capacities to determine need for treatment beds (i.e., detoxification, maintenance, etc.). Drug abuse needs assessment must be approached with this paucity of information in mind. There are five basic types of Needs Assessment approaches according to Warheit et al. (1977) (See Table 1). These basic approaches are the main ingredients to the majority of the needs assessment methodologies in use today. It is germane to this discussion to briefly describe these approaches and their relative strengths and weaknesses.

1. Key Informant This survey approach focuses on individuals in positions that give them access to information regarding the needs of the community at large or its sub-sectors. Examples of key informants include: public officials,

TABLE 1 Needs Assessment Methodologies (Warheit, Bell, & Schwab, 1977) 1. 2. 3. 4. 5.

Key Informant Community Forum Rates-Under-Treatment Social indicators Field Survey

medical and social work professionals, program directors, and clergymen. This is a simple and inexpensive approach. It allows the planner to obtain input from different sources with varying perspectives on the community’s treatment problems and needs. Interviewing key people can be useful in establishing contacts for the proposed treatment program. A notable disadvantage of the key informant approach are the biases such sources will present. Additionally, the impressions of key individuals may not represent the actual needs of the community. The views of the key informants are impressions of the community from a particular perspective, occasionally from a position of self-interest, political partisanship or non-representative socioeconomic status.

2. Community

Forum

This is an approach that utilizes public meetings where residents are invited to come to express their concerns and views of community treatment needs. A community meeting is easily and inexpensively organized. A well advertised, conveniently located meeting can provide input from various community sectors. The needs identified using this method are from the people most affected by the problems. A side benefit is that a meeting can start a “word of mouth” campaign to inform the community of the proposed program. An obvious weakness is the difficulty in drawing representative attendance. Locating meeting places that are accessible to the majority of the community constituents can be difficult. The tendency to draw only concerned, vocal coalitions can lead to collecting unrepresentative data and biases. Another weakness inherent in this type of approach is that it tabulates subjective opinions rather than objective data. 3. Rates-Under-Treatment This method incorporates a detailed inventory of individuals that have previously used treatment services. This can include data from public records, private practice records, or health agencies. The survey data used in this approach is for the most part available and inexpensive. The information gathered can provide a good overview of the services currently rendered within the target area. This material can lead to more efficient planning of services to address unmet needs in the target area. Although this data is readily available, a large portion of the population in need of treatment goes unnoticed by human services. Therefore, it is not recommended to use this approach to study incidence (Kaelber & Noble, 1982). With the growing number of proprietary treatment programs, there may be some difficulty in obtaining data from those sources. Pri-

J. M. Maddock et al.

108

vate programs may be reluctant to share data with potential “competitors.” It might, however, serve all the programs in an area to share information in order to decrease the amount of duplicative treatment services. 4. Social Indicators This is a statistically based approach using such indicators as mortality, health, socioeconomic and criminal records to derive an estimate of alcoholism prevalence. The wide availability of the data used in this approach is its major strength. Equally puissant is the design flexibility and adaptability of using social indicators. It is relatively inexpensive to perform and the results are objective. The weakness most frequently referred to is that this approach is an indirect measure of need. The accuracy and stability of the social and health indicators are the subject of much debate. The weaker the mathematical correlation supporting the indicator, the weaker the assessment. A serious deficiency in this approach is the use of data from sources such as census records which can quickly become outdated and inaccurate. In general, current data sources provide a more accurate analysis (e.g., a survey done every two years is more useful than a survey done every ten years). 5. Field Survey This technique involves collecting information from a sample of the community population through the use of questionnaires. The primary strengths of a survey are flexibility and the capacity to obtain specific, upto-date information from individuals regarding needs and utilization of treatment modalities. The chief drawback to this assessment activity is that it is expensive to conduct. The expense and time requirements of training interviewers and ensuring congruous interviews further complicates this approach. Surveys rely on individuals being willing to admit need and potential utilization of services, and accurately portraying their own use and abuse of substances. Since alcohol and drug use are socially sensitive issues individuals tend to minimize their self-reports of consumption (Edwards, 1973). Survey results may represent “admitted” levels as opposed to actual levels of need and lack total population coverage (Kaelber & Noble, 1982). METHODS, MODELS AND APPLICATIONS IN USE AROUND THE COUNTRY

A model may involve utilizing any or all of the five approaches described above in a simple calculation or a complex, convoluted series of equations and assump-

tions. The purpose of a model is to imitate reality. However, it is highly unlikely that a model will be an exact reproduction of reality. Expecting prophetic results is probably the most common misuse of models. For planning purposes a model can be accurate enough to forecast approximate numbers. Understanding the limitations of a model is important and strengthens the effectiveness of the assessment.

1. Prevalence

Models

Prevalence is the number of problem drinkers or drug abusers in a target area at any one time. Prevalence models utilize social indicators and survey approaches to estimate the number of problem drinkers in an area. Prevalence models do not estimate incidence. Incidence is the number of new cases arising at any one time. There are many prevalence models in use around the country, although there are only a handful of widely used approaches. The most widely used method is the Marden approach (Kaelber & Noble, 1982) which is a survey-based approach. This model utilizes matrices assimilating simple demographic factors such as age, sex, or occupation. Essentially, the target population is arranged by age and sex in a matrix and multiplied by the alcohol problem prevalency rate for each age/sex group using national survey data. It is an easy to use method. The most notable criticism of the Marden technique is the premise that the ratio of problem drinkers is constant throughout the country. This approach ignores demographic and cultural factors. Since this approach is based on survey data, dependability, and timeliness of the survey are of concern. Although this method is the most widely used Needs Assessment tool, it has serious flaws. The second most commonly used prevalency model is probably the most well-known, the Jellinek formula (Kaelber & Noble, 1982; Keller, 1975). This formula is a cirrhosis mortality-based measure. It falls into the category of social indicator approaches. This method relies on the correlation between alcohol problems and deaths resulting from cirrhosis. The formula has been criticized due to several weaknesses. The constant values used in the formula such as the proportion of deaths attributed to alcoholism and the number of those deaths caused by cirrhosis, are regarded as changing more rapidly than assumed by the Jellinek formula. Another serious criticism is that the formula is based on one statistic, cirrhosis mortality. The implication is that the approach is an over simplification of a complex social problem. Other problems with this approach include the contention that alcoholism is a nebulous diagnoses and the questionable accuracy of death certification (Kaelber & Noble, 1982; Schmidt & delint, 1970).

I09

Needs Assessment Approaches 2. Prescriptive

Models

Prescriptive techniques involve the assignment of assumptive values to treatment capacities or need figures. Some commonly used needs assessment models that employ prescriptive elements include the New Jersey treatment needs model (Kaelber & Noble, 1982) and the Nebraska needs assessment procedure (Ford, Luckey, & Wiseman, 1978; Kaelber & Noble, 1982). A commonly used prescriptive figure is that 20% of the alcoholic population will seek treatment. The assumptions used in arriving at the 20% figure are outlined in Procedures for Assessing Alcohol Treatment Needs (Kaelber & Noble, 1982), as follows: 1. There is a 10% increase in the alcoholic population per year (based on literature findings). 2. It is necessary to treat 3OVo of all alcoholics per year to keep up with this increase and with the effects of recidivism. 3. It is also necessary to divide this 30% figure by one-half because alcoholics comprise one-half of the problem drinker population. Thus, 15% of the problem drinker population [need to] be treated. 4. Add 5% as a buffer figure to help keep pace with the growth of the alcoholism problem. 5. Thus, 15% plus 5% equals 20% of all problem drinkers, the number which should at least be seen as an intake interview in alcohol treatment agencies in a given year (Ford & Luckey, 1977; Ford et al., 1978). The drawback of this approach is that the process of prescribing optimum levels of treatment slots or beds is often based more on bureaucratic and factional influences than actual data analyses (Kaelber & Noble, 1982). Although based on actual treatment statistics these formulated values seem more conjectural than empirical in nature. This method’s most critical weakness lies in its speculative nature.

3. Normative

Models

Studying and applying patterns of utilization on a large scale such as national or state trends is the basis for normative models. Derivations of the national or state “norms” can be applied to the target area to estimate treatment requirements. The major weakness in this approach is that it can overlook characteristics that can make two populations of similar size drastically different in their treatment needs. It can provide a rough estimate of need, but one must be aware that there are many qualities that can make communities of comparable sizes very distinct in their specific treatment needs. Data from the Alcohol Epidemiology Data System (AEDS) of the National Institute on Alcohol Abuse and Alcoholism has been utilized to create the AEDS Normative Model (Kaelber & Noble, 1982). This ap-

preach assesses need for alcoholism treatment services using seven indicators of mortality. These indicators are the death rates for liver cirrhosis, alcoholic psychosis, alcoholism, alcohol poisoning, automobile accidents, suicide and homicide for persons aged 15 to 74. Through factor analysis of the nationwide patterns of these indicators two indices have been derived: The Chronic Health Index and the Alcohol Casualty Index. The Chronic Health Index represents long-term chronic health problems. The Alcohol Casualty Index is more typical of acute factors. These two indices are combined to form the Composite Index for overall alcoholic treatment needs. These indices are used to characterize counties by alcohol related problems and size. The model predicts high, low and expected patient and bed numbers for comparison to actual utilization data. The compelling aspect of the AEDS Normative model is its ease of use. Factor analysis of the social indicators lends this method a validity strikingly absent in many of the methods in general use. This model allows for the varying size and make-up of target areas. Although it has its strong points this method also has several limitations. Due to the use of data from the National Drug and Alcohol Treatment Survey (NDATUS) the AEDS normative model results are fixed and limited to NDATUS reporting facilities. Census based data constrains the timeliness of the results and their applicability. The use of composite population data raises the issue of the questionable practice of associating attributes of a target population at large to individual persons within that population (Kaelber & Noble, 1982).

PRACTICAL GUIDELINES FOR THE CLINICAL PLANNER

Prior to performing the Needs Assessment the planners should define their specific goals for this activity. The more specific the definition of requirements, the more appropriate the model can be for the specific purpose. There are many questions that can be asked to help establish or clarify the objectives of the organizers. Some of the broader issues include the basic who, what, why, where, and when. Who is the assessment for? For example is it for a regulatory agency that prefers a specific assessment method? Who is performing the assessment, a proprietary organization or a public health facility? Is there a specific target group and if so, who are they? Identifying special groups such as women alcoholics or dual diagnosis clients is extremely important for designing programs suitable for these groups. What do the planners want to find out? How extensive do the results have to be? How will the

110

J. M. Maddock et al.

results be used? What do the planners intend to do with the outcome? Where is the target area? For instance is the population being assessed rural, inner-city, main-line America or culturally distinct? This will also include considerations about how fast the community changes due to residents moving in and out, urban renewal, business and industry growth or attrition. These are just a few of the factors that can produce social, cultural and demographic changes. When are the results needed? Do the planners have time for an extended study or do they need results more quickly? When is the assessment being performed in contrast to when the data will be used? Why is the Need Assessment being performed? Is a new treatment service being established or is an existing one being evaluated? Is it for private use or is it in response to a state regulatory agency request for a Certificate of Need? If the needs assessment process is undergone to build a case for a state-regulated Certificate of Need for a new program, it is extremely helpful to know the model and method used by the specific state agency. Different states use different needs assessment methods, and this can lead to disagreement in acceptance of a region’s need for a given type of program. For example, if you demonstrate a need for a 20-bed inpatient rehabilitation program in your county using the AEDS Normative Model based on social indicators, but the state regulatory agency uses a county-wide rates-under-treatment method, you and the state may not agree on the need for your program. Since the sensitivity and biases of methods vary, it is most prudent to evaluate program need using a method consistent with the way the regulatory agency will evaluate your application. If you believe that another method is more accurate for planning purposes, then you might utilize both methodologies. The planner must also recognize that there will not always be suitable kinds of data available to perform a specific type of Needs Assessment. Limited financial resources, limited human resources, geographic boundaries, and competitive business practices may hinder data collection to develop a desired model. Therefore, low cost yet valid, public domain approaches such as the AEDS Normative Model should be considered as a viable alternative. COMMENT

The process by which the need for chemical dependency services is assessed is critical in helping program planners identify the potential treatment needs within a given community or geographic region. Identification of needs is a prerequisite to developing specific treatment programs to meet these needs, for example,

determining location of services to ensure optimal accessibility, and identifying the appropriate staffing patterns. The climate today is one in which the need for a treatment program often must be “justified” with data that demonstrates a statistical need for additional or even new services. Thus, the Needs Assessment process helps to facilitate a particular program’s application for approval by regulatory agencies who are looking more closely at proposed programs as they proliferate in today’s society. This process also is an important one because other programs in the geographic area may see the new program as a “competitor” and oppose its implementation. These other programs may argue that additional services are not needed. In those cases in which a formal application for a Certificate of Need is required by law, such opposition often leads to public hearings in which the planners of the proposed program have to respond to the concerns or issues raised. If a formal Needs Assessment has not been completed, the planners are likely to run into serious difficulties in justifying the development of services. The Needs Assessment process is a highly complex one that helps to translate nationally, state or locally derived statistical data into tangible estimates such as how many people in a given area will have a chemical dependency problem, how many are likely to seek services and what type of services will be sought (e.g., detoxification, rehabilitation, outpatient). Data available in a given region may be limited, difficult to find or even unavailable. And data derived from national norms may not be culturally congruent with a particular geographic area. Although it has many limitations, completing a Needs Assessment provides a “rough” estimate of treatment requirements in a community. Needs Assessment is an imprecise science. Successful treatment programs utilize this data with other tools such as marketing techniques, outreach programs, etc. Planners should be intelligent in their use of Needs Assessment techniques. This includes recognizing the strengths and limitations of the approach they have chosen (Ford, 1985). There is no central method shared by all states, and often no centralized organization within the states for Needs Assessment. There is no single “ideal” method for assessing alcohol treatment needs. Therefore, choosing a method is often more a question of money, time constraints or familiarity with only one or two methods. A widely acceptable method is needed for general use to address alcohol treatment needs. One method agreed upon by all the states would render some uniformity to assessing need. It might facilitate developing more accurate, comprehensive needs assessment methodologies and generate better epidemiologic data pertaining to chemical dependencies. A universal

Needs Assessment Approaches

needs assessment method would provide national data that could be used to study the differences between populations and why certain populations exhibit one level of treatment needs while demographically similar populations elsewhere exhibit entirely different needs. It is hoped that an acceptable standardized approach will be developed and used consistently across the country. REFERENCES Edwards, G. (1973). Epidemiology applied to alcoholism: A review and an examination of purposes. Qucrrter/y Journakof Studies on Alcohol, 34, 28-56. Ford, W.E., & Luckey, J.W. (1977). An empirical needs assessment for alcoholism services: Implications for public policy. Lincoln, NE: Nebraska Division on Alcoholism. Ford, W.E., Luckey, J.W., & Wiseman, B. (1978). An empirical needs assessment for alcoholism services: Implication for public policy. Presented at the Second National Conference on Need KY. Assessment-Health and Human Serviceswuisville,

111 Ford, W.E. (1985). Alcoholism and drug and abuse service forecasting models: A comparative discussion. The International Journal of the Addictions, 26, 233-252. Holden, C. (1987). Is alcoholism treatment effective? Science, 236, 20-22. Kaelber, C., & Noble, J. (1982). Procedures for assessing alcohol treatment needs. Rockville, MD: National Institute on Alcohol Abuse and Alcoholism. Kearney & Co. (1979). Population-based mental health needs assessment methodology for use in HSALSP. Submitted to Health Systems Agency of Southeastern Pennsylvania. Keller, M. (1975). Problems of epidemiology in alcohol problems. Journal of Studies on Alcohol, 36, 1442-1451. Lowman, C., Bertolucci, D., Sanchez, D., Reed, P., & Patterson, D. (1984). Epidemiological Bulletin 2: Changes in alcoholism treatment services> 1979-1982. Alcohol Health and Research World\Winter 83/8$ 44-47. Schmidt, W., & de Lint, J. (1970). Estimating the prevalence of alcoholism from alcohol consumption and mortality data. Quarter/y Journal of Studies on Alcoholism, 31, 957-964. Warheit, G.J., Bell, R.A., & Schwab, J.J. (1977). Needsassessment approaches: Concepts and methods. Rockville, MD: National Institute of Mental Health.