Empirically supported psychological interventions controversies and evidence

Assignment: Accessing Information About Evidence-Based Practices

Introductory Principles of Social Work Research

Bruce A. Thyer

The scientific approach to unsolved problems is the only one which contains any hope of learning to deal with the unknown.

A -Bertha Capen Reyno lds (1942, p . 20)

n emphasis on the value of scientific research has always characterized professional social work education and practice. Indeed, this emphasis is one of the hallmarks that distinguishes genuinely "professional" services from other forms of private/public philanthropy and charity and the provision of social care motivated by religious, familial, altruistic, or

philosophical reasons. In the history of social work in )Jorth America and Great Britain, as well as in other European nations, the system of poor laws and other rel- atively unsystematic attempts to care for the destitute gave rise during the latter part of the 19th century to an orientation labeled scientific philanthropy. Coincident with the emergence of "friendly visiting;' settlement houses, formalized academic train ing, and ot her prec ursors to the professionalization of social work, the development of charitable se rvices gui ded h y a sc ienti fic orienta ti o n has evolved to the present day.

Social work historian John Graham provides a good case study o n a To ronto charity hom e for women called The Haven, established in 1878 by re li gio us elites, that gra dually made Lhe tra nsition Lo a rn o re secularl y o riented and p rofess ional service. Gr aham (l.992) describes the completion of this tra nsition in 1927 ::is follows:

Profess ional social work, therefore, had been firm ly installed at The Haven, and the last vestiges of the benevolent philanthropy of the nineteenth century were aban- doned. A growing sense of professional identity moreover demanded a strict delin- eation bet.ween the social worker and the social agency volunteer. Differentiating the former from the latter was a scientific knowledge base and specialized skills which were the social worker's alone. (p. 304, italics added)

Such a transition can be said to characterize the. majority of social work programs across orth America by the early part of the 20th century. Currently, one widely used definition

of social work can be found in The Social Work Dictionary published by the N'ational Association of Social Workers- "the applied science of helping people achieve an effective

2 THE HANDBOOK OF S OCIAL WORK R ESEARCH M ETHODS

le\rel of psychosocial function and effecting societal changes to enhance the well-being of all people" (Ril rker, 2003, p. 408, italics added). Many states further defme the practice of clinical social work, and Florida's definition provides a representative examp le of the inter- connec tedness of social work and science: "The ' practice of clinical social work' is defined as the use of scientific and applied knowledge, theories and methods for the purp oses of describing, preventing, evaluating, and treating, indiv idual, couple, fa mi ly o r gro up behav- ior " (Florida Departmen L of Hea.lth, 2008, ita lics added) . These definitions illustrate the close lin kage between the practice of social work and the world of scientific inquiry.

\'\'here do we social workers come from organizationally? V\lc have many roots, but a central one was the establishment in 1865 of the American SocjaJ Science Association (ASSA), a generalist organization influenced by French sociologist Auguste Com te's then novel philosophy of science labeled positivism, which called for the objective study of human society and behav io r using the same tools of scientific inquiry that were proving so successful in the biological and physical scie nces. rrom the ASSA sprouted numerous o ffs hoots, some of which thrive to this day, although the parent g roup crumbled in 1909. from the ASSA, in 1879, eme rged the Co nfe rence of Charities, which in 1.881 evolved into the Nat ional Conference of Charities and Correction (NCCC), described as "a forum for the communication of the ideas and values co nnccLcd with scientific char ity" (Germain, 1970, p. 9). In turn, the NCCC was renamed the Na tional Conference on Social Work in 19 17. This label lasted until 1957, when it was altered to the National Conference on Social Welfare, which gradually expired during the 1980s.

More recently, in 1994, a small group of social workers led by Janet B. W. Williams estab- lished a new scientifically oriented social work membership organization known as the Society for Social Work and Research (SSWR). AIJ social workers with a n interest in scien- tific research in social wo rk are eligible to join. T he SSWR quickly grew from 27 1 members in 1995 to more than 1,300 in 2009, and the organization has an active news letter and program of annual in tern ational conferences. The first professional SSWR co nference was held in 1995 in Washingto n, D.C., and has been followed annually since that time with very successful and high-quality conferences (see www.sswr.org) . The SSWR conferences offer a hos t of competitively reviewed symposia, papers, and posters; P.lcnary addresses by promi- nent social work researchers; and an awards program that recognizes outstanding e..xamples of recen tl y published social work research. Because of its superb organization and the top quality of its presentations, the SSWR conference has rapidly become Lhe preferred venu e for social work researchers to present their research findings. Moreover, it has become the conference ol choice for schools of social work to seek interviews w it h potential new faculty and fo r potential new faculty to seek academ ic positions. In 1999, lhc SSWR began provid- ing its members a subscription to Lhc bimon thly peer-reviewed journal Research on Social Work Practice, an in dependen t periodical established in 1991. This grow th of the SSWR augurs well for the continuing voice of science within mainstream social work.

A related bu t independent development was the establishment of the Institute for the Advancemc11l of Social Work Research (IASWR) in 1993. The mission of the IASWR is to create infrasrructure for social work research, to lead advocacy efforts to fund social work research, to help stakeholders view social work research as valuable, to provide training and professional development programs for social work researchers, to persuade social workers to undertake careers in research, to provide a free Web-based research-focused newsletter, and to promote disc ip l.in ary and inte rdisciplinary resea rch collaboration . Pive nalional pro fess io nal social work organizations contributed to the developrncn l o f the IASWR and are represented on its governing board. Its origi nal p urpose of advocating for the establishment of a federally funded Na 1ional Center for Social Work Research failed in the face of fiscal austerity, bu t the IASWR has expanded its remit as described above (see http://ww\>v.iaswresearch.org/) .

(MAPTER l • INTRODUCTORY PRI N CI PLES OF SOCIAL W ORK RESEARCH 3

Anolhcr organizalional reso urce for social work research is the Social Work Topical Interest Group (TIG) found within the American Evaluation Association (AEA) . The AEA has about 5,000 members, and several hundred of these comprise the social work TIC. The AEA holds an annual conference as well as regional ones, has an active journals program, and provides training and consultation services, and its Web site has a wealth of useful resources (e.g., locating measurement instruments, how to locate an evaluator; see hup://www.cval.org/aboutus/organization/aboutus.asp).

The National Association of Social Workers is the largest professional social work group in the world, with about 150,000 members. Almost aJJ are M.S.W. and B.S.W.-lcvcl trained professionals, and the organization primarily serves Lhc needs of ils practitioner member base, not those of social work research ers. The NASW does not host an annual conference but does have one research journal, Social Work Research J\ new initia tive is a social wo rk resea rch Well page (see www.socialworkers.org/research/), cosponsored with the IASWR, which is itself ostensibly independent but is actual1y h oused within the NJ\SW offices in Washinglon, D.C.

Social work resea rchers also find welcoming organizational suppo rt from various dis- ciplinary (e.g., American Psychological Association, American Sociological Associatio n, Associa li on for Behavior Anal ysis) and in terdisciplinary (e.g., Am erican P ublic Health Association, Associatio n fo r Advancement of Behavioral and Cognitive Therapies, American Orthopsychiatric Association, the Gerontological Society of America) groups. These groups typically have thriving annual conferences, a wcll-cslablished journals program, and training opportunities social workers can take advantage of. Thus, both budding and experienced social workers have ample opporlunities to network with research -oriented colleagues both within and oulsidc of lhe discipline.

Scientific Perspectives on Practice

The role of scientific research in social welfare can be seen through many early writings, including a11 article titled "Scientific Charity," presented at the 1889 meeting of the NCCC (cited in Germain, 1970, p. 8), and one titled "A Scientific Basis for Charity" (Wayl and, 1894), which appeared in the influential journal The Cha.rities Review. Such perspectives cu lmi n ated in the publication of Richmond's (1917) Social Diagnosis, an influenLial text that wholeheartedly extolled the virtues of positivist science. lndeed, in 1921, Richmond received an honorary M.A. degree from Smith College for "esLablishing th e scientific basis of a new profession" (cited in Germain, l 970, p. J 2).

The possible examples of conference talks, journaJ articles, chapters, and books illus- trating the central reliance on scientific research as a guiding force within early social work arc roo numerous to mention further here. Germain (1970) remains one of the very best reviews of this "ancient" history of our profession. More recent is the history of the Social Work Research Group (SWRG), a short-lived professional membership organ ization established in 1949 that became one of the original seven constituents of the l\'ational Association of Social Workers (NASW) in 1955, transmogrifying itself into the NASW's Research Section. In 1963, this became the NASW's Council on Social Work Research, where it gradually faded from v iew by the mid-1960s as the NASW allowed the research mission established in its bylaws to Largely lapse. Graham, Al-Krenawi, and J3radshaw (2000) have prepared an excellent historical study of the rise and demise of the SWRG.

Coinciden t with these organizational and policy developments related to the integra- tion of science and social work during the past quarter century have been t hree related perspectives on practice. The first is known as empirical clinica.l practice (ECP), the second

4 THE HANDBOOK OF SOCIA i WORK RFSFARCH MFTHOr>S

is called empirically supported treatments (ESTs), and the third is labeled evidence-based practice (F.BP ). Th ese are reviewed briefly in turn.

Empirical Clinical Practice

Empirical clinical practice was the name of a book authored by social workers Siri Jayaratne and Rona Levy (1979), who describe the characteristics of the ECP model they espouse: "Empirical practice is conducted by clinicians who strive Lo measure and demonstrate the effect of their clin ical practice by adapting traditional experimental research techniques to clinical practice" (p. xiii). The authors focus on teaching social workers the use of relatively simple research methods ca lled single-system research designs to empirically evaluate the outcomes of their work. l'hey be lieve t hat "clinical practice that can empirically demonstrate its effect prov ides the basis for the best service to the client" (p. xiv). They contended that ECP ca n be ::idopted by p ractitioners using vir- tually any theoretical model of practice so long as it is possib le to measure changes in the client, re late t·hese changes (provis ionally) Lo soc ial work inle rvc nlion , and Lhen base future services on these observations. The auth ors advoca te that social workers should rely on previo us research to help guide their cho ices of interve11tions that they offer clients. In their words, "The clinician would first be inte rested in us in g a n in terventi on strategy that has been successful in the past . . .. When established techniques are avail - able, they should be used, but they should be based on objective evaluation rather than subjective feelin g" (p. 7) . ECP involves the careful and repeated measure of client func- tioning, using reliable and valid measures repeated over time, combined with selected treatments based on the best available scientific evidence. Their entire book is devoted to describing how to do these activities. A similar social work text by Wodarski ( 1981 ), titled The Role of Research in Clinical Practice, advocated for much the sa m e thing- a preference to make use of psychosocial treatments that scientific research had really demonstrated to be of benefit to clients, measuring client functioning in reliable an

The banner of ECP was picked up by a number of subsequent social workers, and a rather large (a nd not uncontroversial ) literalure has grown around Lhese nolions (e.g., Corcoran, 1985; Ivanoff, Blythe, & .8riar, 1987; Ivanoff, H.obinson, & Blythe, 1987; G. MacDonald, 1994; Thyer, 1996). The influence of ECP has not been inconsiderable. For example, in 1982, just 3 years fol lowin g the publicalion of F.mpirical Clinical Practice (Jayaratne & Levy, 1979) , the curriculum policy statement of the Co un ci l on Social Work Ed ucation (CSWE, 1982) included a new mandate Lh a l research co urses musl now Leach "designs for the systematic evaluation of the student's ow n practice . . . [and should] pre- pare them systematically to evaluate their own practice and co ntr ibute to the generation of knowledge for practice" (pp. 10- 11). Similar sta ndards still ca n be found in the curren t CSWE guidel ines. Insistin g that individual practi tio ners co ndu cl systema tic outcome evaluations of their own services was a remarkable professional standard, one that has not ycl bee11 cm ul a lcd by educational and practice guidelines wiLhin clinical psychology o r psychiatry in the present day. Reid ( 1994) provides a ni ce overview of the rise, influence, an d dissemjnation of the ECP movement.

Empirically Supported Treatments Subscquenl lo Lhc ECP movement within social work, a rclaled iniLiaLive developed within clinical psychology called empirically validated treatments. During the mid - l 990s, the president of Section lll (Society for a Science of Clinical Psychology) of Division 12

CHAPTCR 1 • IN l ROOUtTORY P RI NCIPLES OF S OCIAL W ORK R ESEARCH 5

(Clinical Psychology) of the American Psychological Association convened a Task Force on Promotion and Dissemination of Psychological Procedures, a gro up charged with two functions: (a) develop a scientifically defen sible set of criteria that can be used to deter- mine whether a given psychological technique ca n be called empirically va lida ted and ( b) cond uct co mprehensive reviews of the research literature, apply these cr iter ia, a nd come up with, in effec t, lists of psychological procedures that fulfill these criteria and, therefore, can be co nsidered, in a scientific sense, empirically validated.

Th e evid enti ary sta ndards ultimately decided o n by the task force were actually rather modest, consisting of the following criteria:

I. At least two good between-group design experiments demonstrating efficacy in one or mo re of the following ways:

A. Superior to pill or psychological placebo or to another treatment B. Equivalent LO an already established treatment in experiments with adequate

statistical power

II. A large ser ies of single-case design expe rim en ts ( N > 9) demonstratin g efficacy that must have done the following:

A Used good experimental designs B. Compared the intervention to another treatment (as in I.A. )

Among the further criteria are that the psychological techniques must be based on well-proceduralized treatmenL manuals, that the characteristics of the client sam ples are clearly defined, and th at the positive effects must have been demon strated by at least two different inves tigators or investigatory teams. A psychological treatment m eeting the preceding criteria co uld be said to be well established. A som ewhat less stringent set of cr i- teria could be followed to potentially label a treat ment as probably efficacious (Chambless et al. , 1996).

With the criteria in place, the task force busily got to work in seeing which psycholog- ical treatments co uld be labeled empirically validated and probably efficacious, and reports soon began appearing indicating empirically validated inLerventions for a wide array of psychosocial disorders such as depression, panic disorder, pain, and schizophre nia. As with the ECP movement within social work, the task force within psychology did not escape controversy. For one thing, the task force recognized that labeling a treatm ent as empirically validaled see med to close the discussio n off, implying perhaps a stronger level of research evidence than was justified. Subseq uent reports of the task force used lhe more t empered language of empirically supporled lreatments (ESTs) . Entire issues oflead- ing professional journals (i.e., a 1996 issue of Clinical Psychology: Science and Practice, a 1998 issue of the Journal of Consulting and Clinical Psychology, a 1998 issue of Psychotherapy Research) were devoted to the topic, as were considerable independent lit- eratures (e.g., Sanderson & Woody, 1995). The influence of the EST movem ent also has been strong, and the work of the Division 12 task fo rce was commented on extrem ely favorably in Mental Health: A Report of the Surgeon General (Hatcher, 2000). The volume titled A Guide lo Treatments That Work (Natha n & Go rman, 2007), now in its third edi- tion, i.s an exemp lary resource for social workers seeking relatively current information about empirically sup ported treatments for a wide va ri ety of m ental hea lth prob lems. Division 12, Sec.:Lion HT (The Society for a Science of Clinical Psychology) co ntinues its work in defining the cr iteria and language used to describe empirically suppo rted treat- ments and maintains a Web site providing curre nt information on this influential initia- tive (see http://www.psychology.sunysb.edu/eklonsky-/divisionl2/index.html).

6 Tll E HAIWBOOK OF S OCIAL W ORK R ES EARCH METHODS

Evidence-Based Practice Coinc ident with the EST initiatives in clinical psychology have b een related activities in med ic ine labeled evidence-based practice, defined as "the conscientious, explicit, and judi- ciou s use of the current best evidence in making decisions about the care of individual patients" (Sackett, Richardson, Rosenberg, & Haynes, 1997, p. 2). On its face, EBP would not seem to be a radical notion, and indeed, most readers would ass ume that such a sta n- dard already was in place in most of the hea Ith professions. St'td ly, to a great extent, this is not the case, altho ugh a small but in fluen tit'tl grou p of health care providers is attempting to make it so. EBP and EST actually are much more sophisticated var iants of th e earlier ECP model of social work, but the spirit and intent of all three movements ECP (devel- oped within social work), EST (developed within psychology), and EBP (developed within medicine)-are the same. EBP is gradually supplanting the ECP and EST initia- tives within social work and psychology. The current president of the Society for the Science of Clinical Psychology (a section of Division 12 of the Amer ican Psychological Association) pub lished an edito rial titled "Evidence-Based Psychoth erapy: A Graduate Course Proposal" (Persons, 1999 ), and some social workers have begun using the EBP language, most notably Gambrill (1999) w ith her thoughtful arLicle titled "Evidence- Based Practice: An Alternative to Authority-Based Practice," which introduced El3P to the socia l work literature. The past decade has seen the publication of enough social work books on the EBP topic to fill a bookshelf. The melding of these disciplinary perspectives i11lo an interdisciplinary human services movemen t generically ca lled evidence-based prac- tice seems likely. Consider Perso ns's ( J 999) description of EBP:

T he evidence- based practitioner:

• Provides informed consent for treatment • Relies on the efficacy data (especially from RCTs [randomized cl inical trials]) when

recomm ending and selecting and carrying out treatm ents • Uses the empirical literature Lo gu ide decision-m aking • Uses a systematic, hypothesis -testing approach to the treatment of each case:

o Begins with careful assessment o Sets dear and measurable goals o Develops and individualized fo rmulation and a treatment plan based on the

formulation o Monitors progress toward the goals frequently and mod ifies or ends treatment

as needed (p. 2)

WeU, perhaps Jayaralne and Levy ( 1979) were simply two decades al1cad of their time. An issue of the NASW News contained an article on the Surgeon General's Report on Mental Health and noted, "A challenge in the near term is to speed transfer of new evidence-based treatments and prevenlion interventions into diverse service delivery settings and systems" (O' Neill, 2000, p. 6, italics added ). The Surgeon General's repo rt itself states clearly,

Responding to th e calls of managed menta l health and behavioral heaJth ca re sys- Lcms for evidence-based interventions will have a much needed and discernable impact on practice .... It is essential to expand the supply of effective, evidence- based services throughout the nation. (Hatcher, 2000, chap. 8, p. 453 )

EBP requires knowin g what helps socia l work clients and what does not help them. It requires being able Lo distinguish b etween unverified opin-io ns a bout p sychosocial

CHAPTER l • I NTRODUCTORY P RINCIPLES OF SO CIAL WORK RESEARCH 7

interventions and facts about their effectiveness. And separa ting facts from fictions is what science is prelly good at doing. Jot perfectly, and not without false starts, but the publicly verifiable and potentially testable conclusion s of scientific research render this form of knowledge buil ding an inherently self-correctin g one (in the long nm), a con- siderable advan Lagc over o ther "ways of knowing."

EBP differs from its precursor initiatives in that it does not tell socia l workers what interven tions should be provided to clients. TL d oes not list so -called best practices, create practice guidel ines, or develop lists of supposedly emp irically based treatments. Nor docs it unduly privilege certain forms of evidence above all others. Each of the above three sen - tences represents commo n misconceptions of EB P. EBP is actually a process of inquiry offered to practitioners, described for physicians in Straus, Richardson, Galsziou, and Haynes (2005), but readily adaptable to providers in all of the h uman service professions. These steps are as follows (from Straus ct al., 2005, pp. 3-4) :

Step l: converting th e need for information (abou t prevention, diagnosis, prognosis, therapy, causa tion, etc.) in to an answerable qucs Lion.

Step 2: tracking down Lhe besl evidence with which to answer that question.

Step 3: critically appraising that evidence for its validity (closeness to the truth), impact (size of the effect), and applicability (usefulness in our clinical practice) .

Step 4: integrating the critical appraisal with our clinical expertise and wil11 our patient's unique biology, values, and circumstances.

Step 5: Evaluating our effectiveness and efficiency in executing steps 1-4 and seeking ways to im prove them b oth for next time.

Each chapter in Straus et al. (2005) addresses on.e of these steps, and they have been adapted for use by soc ial workers in an exce ll ent series of entries appearing in 171e Social Worker's Desk Reference (see Roberts, 2009, pp. 1115-1182). EBP states that social workers need to be familiar with the best available evidence addressing the questions related to client services and to their particular practice situation and to integrate their appraisal of this information into an assessment of their own skills, the client's preferences, relevant professional and personal val ues and ethical standards, cost, feasibility, and resources . All of these factors a re rcleva n L, no t just what the research evidence indicates. And by best evidence, what is meant is not so-called gold-standard st udies such as randomi zed con- trolled trials or meta -analyses (see later chapters on Lh esc LOpics in thi s book) but simply t he best available relevan t ev idence. If there are no stud ies of superlative quality, then you locate and assess those of lesser quality. Lots of evidence can go in to th e mix, including quasi-experimental studies, single-subject studies, corrclational studies, descriptive work, epidemiological evidence, qualitative investigations, case h istories, theory, and infom1ed clinical opinion. There is always evidence for a social worker to consult, even if it is nol evidence of the highest quality. As with ECP, EBP also encourages practitioners to evalu- ate the outcomes of their work with individual cl ients using a research methodology called si ngle-subject designs .

Another option is for social workers to co nsu lt systematic reviews (SRs) of the research evidence related to various answerable que~tions invo lving assessment and interven tive meth ods. The two gro ups most responsible for preparing high-quality and independent SRs are called the Cochrane Collaboration (sec www.cochrane.o rg), focusing on issues related to health care, and the Campbell Co ll aboration (see www.campbellcollaboration .org), focusing on social welfare, education, and criminal justice. SRs are prepared by

8 THt HAN DBOOK OF SOC IAL W ORK RtSEARCH M El HODS

qualified research teams who obtain articles and reports from all over the world dealing with a specific issue. These reports are minutely analyzed and critiqued and the collected information surrunar ized in a readable format, with a take-away .message something like Treatment Xis well-supported as an effective treatment for clients with Problem Y; The avail- able evidence indicates that Treatment X is ineffective in helping clien ts with Problem Y; Clients with Problem Y who receive Treatment X demonstrated impaired outcomes, com - pared to clients who receive no treatment. You can see hmV" this information would be of immense value to social workers. Here is a sampling of SRs currently available on the Cochrane database that is of relevance to social workers:

• Behavioral and cognitive-behavioral therapy for obsessive-compulsive disorder in children and adolescents

• Family intervention for bipolar disorder • Family therapy for depression • Psychological debriefing for preventing posttraurnatic stress disorder • Psychotherapy for bulimia nervosa and binging • Short-term psychodynamic psychotherapy for common mental disorders

And here are some fonnd on the Campbell Collaboration Web site:

• Cognitive-behavioral therapy for men who physically a buse their partner • Cognitive-behavioral intervention for children who have been sexually abused • Interventions intended to reduce pregnancy-related outcomes among adolescents • School-based educational programs for the prevention of childhood sexual abuse • Work programs for welfare recipients

These systematic reviews represent the highest quality and up -to-date critical appraisals of the existing research literature addressing particular psychosocial and health problems e:>..'})erienced by social work clients. They are a wonderful resource for practitioners seeking such information and are integral to the conduct of evidence-based practice.

To sum marize, ECP suggested that social work treatment should be chosen based on support via randomized controlled studies and that social workers need to evaluate the outcomes of their practice with clients using single-system research designs. The EST ini- tiative came up with a list of evidentiary criteria needed to label a given treatmen t as "empirically supported." Once these criteria were in hand, lists of psychosocial interven- t ions meeting these standards were published. RBP provides more of a process to guide clinical and practice decision making, which explicitly embraces evidence from many sources (albeit urging one to pay particular attention to evidence of the highest quality) and explicitly includes nonscientific considerations such as client preferences and values into this decision -making process. In many ways, EBP is a more sophisticated and mature conceptualization of the conduct of practice than ECP and EST, and these latter two in i- tiatives largely have been subsurn.ed by EBP.

On Terms

The preceding brief overview helps to bring us to the present, wherein social work is attempting to really implement our original aspirations pcrtainiillg to being based on a foundation of scientific research. As in most intellectual undertakings, it always is helpful