The World Health Organization (2001)  has reported that depression affects 121 million people worldwide. Older adults are particularly at risk because depression is underrecognized and undertreated in this population (Anderson, 2001; Brown, McAvay, Raue, Moses, & Bruce, 2003; Brown et al., 2004; Bruce et al., 2002). In 1999, a landmark U.S. Surgeon General’s report (U.S. Department of Health and Human Services, 1999) found that 8%–20% of older adults in the community and up to 37% in primary care settings experience depressive symptoms. Research in a large sample of community-dwelling older adults admitted to two community-based long-term care programs in Michigan showed that 40.5% of the sample had recognized mental disorders, 39.6% used psychotropic medications, 24.5% had probable depression, and 1.4% had thoughts of or attempts at self-injury (Li & Conwell, 2007).

The consequences of untreated depression are formidable. Increased rates of disability, hospitalization, institutionalization, and death are well documented (Anderson, 2001; Li & Conwell, 2007; Sherlock, 2005). Our own prior work in Detroit showed that older adults with depression were less likely to fully recover from their physical impairments and more likely to develop cognitive disorders than their nondepressed counterparts (Lichtenberg, 1998). Making treatment more challenging, depression cannot be adequately addressed without also addressing comorbid mental health conditions (Bruce et al., 2002; Dalby et al., 2008).

Yet, evidence exists that carefully designed interventions to improve mental health and reduce depression work. Occupational therapists provide some of those interventions. For example, Matuska Giles-Heinz, Flinn, Neighbor, and Bass-Haugen (2003)  delivered a 24-session “life of wellness” program to a sample of 65 older adults and found improvements in vitality and social functioning. Using a randomized controlled trial, the Well-Elderly study (Clark et al., 1997) showed that occupational therapy improved the mental health of 361 community-dwelling adults ≥ age 60.

An expert panel of mental health and public health experts convened in April 2006 to address the critical issue of optimal interventions for late-life depression (Frederick et al., 2007; Steinman et al., 2007). Although they found that small sample sizes, limitations in research design, and inconsistent use of evidence-based instruments were hampering progress, they also found that interventions with a cognitive–behavioral therapy dimension held promise, particularly when practiced in the everyday environment of patients’ homes and in primary care settings. Occupational therapists working in home health, skilled nursing, and outpatient rehabilitation are therefore in an ideal position to respond to late-life depression.

Occupational therapists have a problem-solving therapeutic orientation that recognizes the critical role of emotions in health and well-being (Legault & Rebeiro, 2001; Sedgwick, Cockburn, & Trentham, 2007). Their work often focuses on reengagement in meaningful occupations and the rediscovery of pleasant events that have fallen away in the face of daily struggles with depression (Goldberg, Brintnell, & Goldberg 2002) and cognitive decline (Wood, Womack, & Hooper, 2009). They also know that to help severely depressed patients, a structured remotivating process (Pepin, Guerette, Lefebvre, & Jacques, 2008) may be required. Therapists know that depressed people are biased toward negative interpretations and negative patterns of thought (Beck, Rush, Shaw, & Emery, 1979) and that an effective path out of this negative cycle is the gradual reincorporation of pleasant activities into one’s daily life. This technique, which is conscious, structured, purposeful, and goal oriented and includes opportunities for personal choice and control, is entirely aligned with occupational therapy therapeutic principles.

Given the evidence that focused interventions can reduce late-life depression, we asked, “Is there a way to strengthen occupational therapy practice skills so that therapists can better identify and help older adults with depression?” Our purpose in this article is to describe our efforts to develop and empirically test an educational intervention designed for this purpose.

The specific goals of this intervention study were twofold: (1) to assess occupational therapists’ (n = 30) mental health knowledge and attitudes before and after intervention using a set of professional education materials delivered in DVD format and (2) to assess change in occupational therapists’ (n = 20) clinical practices using a chart audit (N = 384) at 3 mo before and 3 mo after DVD training.

The educational DVDs were developed by the research team on the basis of a review of the literature and the team’s collective clinical experience. Early on, we agreed that a DVD format would be optimal because it could depict clinical vignettes using real patients and guided demonstrations of assessments. We also agreed that a 6- to 8-hr viewing time was best to avoid viewing the educational materials’ being perceived as too onerous. We opted for a studio-based interviewer–expert conversational style, allowing the therapist to learn from the expert as she or he watched the conversations and demonstrations.

The training package included six 1-hr DVD modules: (1) Introduction, Aging and Mental Health, (2) Understanding and Treating Depression, (3) Medications for Depression, (4) Family Caregiving, (5) Falls Balance and Exercise, and (6) Driving Rehabilitation and Community Mobility. A seventh DVD included beginning-to-end versions of all assessments: the Geriatric Depression Scale (Yesavage et al., 1983), the MacNeill Lichtenberg Decision Tree (MacNeill & Lichtenberg, 2000), Mood Ratings for Behavioral Activation (Lichtenberg, Kimbarow, Wall, Roth, & MacNeill, 1998), Alcohol Use and Disorder Identification Test (Babor, DeLa Fuente, Saunders, & Grant, 1989), the Caregiver Reaction Assessment (Given et al., 1992), and the Activities Balance Confidence Scale (Peretz, Herman, Hausdorff, & Giladi, 2006). A single CD–ROM contained all the PowerPoint slides and assessment instruments in PDF format (for ease of printing and use). Thus, the final educational delivery system was a seven-DVD plus one CD–ROM box set (now available as an official Continuing Education product from the American Occupational Therapy Association [AOTA] at www1.aota.org/shop_aota/prodview.aspx?Type=D&SKU=4859).

The original plan was to evaluate the complete DVD box set. However, financial limitations and staff shortages made it difficult to obtain agency commitments for testing more than the first three DVDs. Fortunately, they contained most of material on mental health knowledge, assessment, and treatment.

Thirty occupational therapists were recruited at three sites: (1) a long-term care facility, (2) an outpatient rehabilitation department, and (3) a home health agency (the home health data were dropped partway through the study; see the Chart Review section). Ten therapists were recruited from each site. The sample was purposeful in the sense that we wanted to examine the mental health practices of therapists who worked with older adults, but it was also one of convenience because therapists were recruited by the managers at each site who had already committed to being study partners.

All procedures were first approved by the institutional review board at Wayne State University and our study partners. The educational intervention, that is, the DVD training, was provided by either Cathy Lysack or Peter Lichtenberg in sessions that took place once or twice per week, all within a 2-mo period. The occupational therapists viewed one DVD per 90-min session as a group. For the most part, the DVD training sessions were held in each facility’s main meeting room, but on a few occasions, they took place in a lounge or corner of a therapy room that was free of patients for the required time period.

We designed questions to test knowledge and attitudes about mental health and late-life depression. The questionnaire was administered immediately before and after participants viewed each DVD. Like the modules themselves, the questionnaires were designed by the university experts on the research team with input from the occupational therapy managers at each site. Questions were true–false except for items that asked the therapists to rate their confidence in using specific assessment tools. At the request of several therapists (and recognizing that doing so would encourage more honest responses), questionnaires were anonymous; no mechanism was in place to link a participant’s questionnaires before and after viewing the DVDs. Figure 1 lists the 14 questionnaire items pertaining to the three DVD modules evaluated.

The chart audit assessed therapists’ practice change after receipt of the DVD training. Performance indicators included assessment tools, communication of assessment results to the treatment team, and behavioral activation techniques when depression was found. We also recognized the need to be sensitive to capturing small gradations of change. For example, therapists might observe depression in a patient and document it, but it would be a further step if they also reported the patient’s depression-related symptoms. Performance indicators were also designed to have behavioral anchors to reduce the inferences chart reviewers had to make. Thus, indicators were either present or absent (e.g., therapists performed depression screening, yes–no). Fifteen indicators were developed. The chart audit was conducted 3 mo before and 3 mo after DVD training.

The chart reviews were supervised by the occupational therapy manager at each site. Managers identified one person on their staff (not an occupational therapist) and trained him or her in the review procedures. This staff person conducted all the chart reviews at that site. All chart entries (e.g., assessment notes, discharge notes, progress notes) were reviewed. Peter Lichtenberg returned to the study site once in the early phase of the review to monitor procedures and provided telephone consultation throughout the study period. With respect to chart eligibility, all charts had to meet two criteria: The patient had to be (1) > age 60 and (2) receiving services from one of the therapists participating in the study at that site.

As a methodological note, the original goal was to review 200 charts before and 200 charts after DVD training, drawing charts from 10 therapists at each of the three research sites (home health, nursing home, and outpatient rehabilitation). Power calculations indicated that this number of charts would be necessary to detect practice changes. However, staff turnover in the home health site was so substantial that we were forced to drop this site from the chart review component of the study. Fortunately, the remaining sites were able to increase the number of charts reviewed to closely approximate our original target.

Data were analyzed using SPSS 16.0 (SPSS Inc., Chicago). Given the nonparametric nature of the questionnaire data, we used a z test of proportions to compare the percentage of correct answers provided by the sample before and after DVD training. (Because the questionnaires contained no names or identification numbers, calculation of a paired t test or a Wilcoxon test was not possible). A z test of proportions was also used in the chart review to compare therapists’ practices before and after DVD training. Additional tests examined differences between the nursing home and outpatient sites. We calculated effect sizes (i.e., the effect of the DVD training on occupational therapy practice) using the formula for Cohen’s δ and the population correlation coefficient r (Cohen, 1988). Interpretation of effect size was based on Cohen’s (1988)  recommendations.

As indicated in Table 1, the occupational therapists in this study gained new knowledge and more positive attitudes after DVD training. The knowledge testing data showed that the percentage of therapists with the correct answer increased 20% to 30% for 5 of the 11 knowledge questions (see Table 1 and Figure 1, Items 2, 3, 4, 9, and 13). Z tests confirmed that these increases were significant.

The patients included in the chart were relatively typical of nursing home and outpatient rehabilitation populations. Most of the patients were women. Rates of comorbid health conditions, especially heart disease, diabetes, and dementia, were higher in the nursing home group. The outpatient group was healthier: No patients had a formal diagnosis of depression, and few had a diagnosis of cognitive impairment. No demographic data for the therapists, or any other data related to therapist characteristics, were collected.

The chart review data are provided in Table 2. The chart review was conducted to assess whether the DVD training resulted in any change in therapists’ practice behaviors—without behavior change, the mental health needs of the therapists’ patients would remain unaddressed. Although slightly fewer charts were reviewed than originally planned (199 before and 184 afterward), the data showed that therapists changed their practices in a positive direction. As seen in Table 2, 14 of the 15 performance indicators assessed increased after DVD training. For example, depression screening by the occupational therapists increased from 3.0% at pretraining to 25.3% at posttraining. (In actual numbers, this equaled 6 older adults before training increasing to 46 older adults after training). The z scores for 9 items were statistically significant. Note, however, that one performance indicator, mention of caregiver, appeared less often after the DVD training in both sites.

The performance indicator data from the chart review for the two study sites separately are presented in Table 3. One can see that the performance indicators were higher in the nursing home group, both before and after DVD training. Before training, it appears that occupational therapists in the nursing home were relatively more focused on residents’ mood and depression and cognitive functioning and communicated this information to the treatment team. When depression was identified, occupational therapists engaged in efforts to facilitate pleasant events in the residents’ lives. Although the same performance indicators increased in the outpatient group, others increased, too. Those indicators included mention of mood or depression and communication of this information to the treatment team. Moreover, noted in the outpatient group only, screening for depression and cognition did not happen at all before training.

The final step with the chart review data was to calculate the effect size of individual performance indicators (evidence of the strength of practice change). These data are presented in Table 4. The calculation found that five items were significant in the nursing home therapist sample, and four items were significant in the outpatient therapist sample. Although the size of the improvement was greater in the outpatient setting, two critical performance items, (1) screening patients for depression and (2) screening patients for cognitive impairment, increased sevenfold and threefold, respectively, in the nursing home group.

The major finding of this study was that the educational intervention delivered by DVD was effective: Our sample of occupational therapists increased its knowledge about mood disorders and depression and demonstrated positive practice change. The therapists talked about depression and cognition with their patients more often, reported patient mood more frequently to the treatment team, and screened patients more frequently for mood and cognition. In the nursing home, therapists also facilitated patients’ recollections of positive events more often.

These findings are encouraging, because the literature has suggested that a major impediment to treatment of late-life depression is awareness of the problem and discomfort with initiating discussions about depression. The data suggest that the occupational therapists in our sample took their new knowledge seriously and transformed their practices accordingly. Although we cannot conclude that the DVD training was responsible for all the changes observed, the magnitude of the change for some items (e.g., the significant increase in depression and cognitive screening), coupled with the temporal aspect of the training, suggests that the DVDs were at least partially responsible for the positive changes observed.

Several unexpected findings require further explanation. First, although on the knowledge questionnaire scores on 5 items increased significantly, scores on 4 items decreased (Item 1, ageism definition; Item 7, time needed for cognitive screening; Item 11, mechanism of antidepressants; Item 12 = side effects of selective serotonin reuptake inhibitors). It is possible that the questions probed DVD content that was more difficult to grasp or recall than other items, or perhaps the questions were less clearly written, making correct responses more difficult. The differences in scores between the two time points on these 4 items were not statistically significant, however, so we believe this issue is relatively minor.

The second unexpected finding came from the chart review. This finding was that “mention of caregiver” (Performance Indicator 13) was less present after DVD training than before. Although the difference was not statistically significant, one should recognize that when therapists devote increased time to their patients’ mental health needs, it may come at a cost to their patients’ caregivers or some other aspect of care. Study data did show, however, that after DVD training, therapists reported more frequently on caregiver coping or stress and referred caregivers to sources of help more often, indicating that when caregiver stress was seen to be a serious problem, therapists did indeed act.

Finally, we found that screening for depression and cognitive impairment was at low rates in the nursing home and nonexistent in the outpatient rehabilitation site before the DVD training. At first glance, especially in the outpatient group, one might assume that screening was not needed because patients were well. But without assessment, can the therapist really know whether depression is present or not? The literature underscores the striking underrecognition of depression in older adults (Frederick et al., 2007; Li & Conwell, 2007; Steinman et al., 2007). In debriefing meetings after the study, therapists related cases in which they had had no sense that depression was present, yet assessment of the patient proved otherwise. These cases were powerful and led both study sites to adopt the depression (GDS) and cognitive screening (MLDT) tools and make them standard practice.

Study results must be interpreted in light of the limitations of the pretest–posttest design. First, although therapists’ knowledge and practice skills increased after training, without a control group it is not possible to conclude that the DVD intervention was responsible for the changes observed. Even if one accepts that the DVDs had some degree of influence, it is not clear what aspect was most influential. Was it the patient vignettes, the visual DVD format, or something else? Literature has suggested that learning is easier when real patient cases are included (Andersen, 2001). Patient cases with which therapists can identify are thought to help learners retain new knowledge and see how that knowledge can be applied in practice. During the study, therapists mentioned how they liked seeing the assessments demonstrated using real patients. Alternative explanations are available, however. The institutional support provided in the form of enthusiastic occupational therapy managers and a motivated sample likely influenced study findings. Therapists also viewed the DVD modules at work, on work time, and at no cost to themselves. We fully acknowledge these factors in shaping study results.

A second limitation pertains to the design of the chart review and the possibility that the blinding process failed in some way. Although occupational therapy managers at each site had the selected charts redacted to remove the name of the treating therapist and from which time period a particular chart was drawn before the chart reviewers commenced their work, it is possible that the review identified more performance indicators at the posttraining measurement point. We acknowledge this potential threat to the validity of findings.

A third issue is the number of patient charts reviewed in relation to the total number of patients seen by therapists. We are more confident in our findings at the nursing home site, given the larger number of charts reviewed (170 before and 150 after training) and the fact that most of charts reviewed at the first time point would have overlapped considerably with those reviewed at the second time point (given the average length of stay of nursing home residents). We estimate that in the nursing home, about 160 charts divided by 10 therapists in this site translated into 16 charts per therapist, or nearly half the number of patients each therapist would typically treat at any given time. We are confident that evaluating this number of charts provided a reasonable glimpse into therapist practice patterns and the influence of the DVD training on practice change. We urge greater caution in interpreting the findings at the outpatient site because fewer charts were sampled and the charts reviewed had nearly no overlap. Offsetting this limitation somewhat is our knowledge that nearly every eligible chart at the outpatient setting was reviewed because so few patients at this site met the ≥ age 60 inclusion criteria.

The final study limitation relates to the measures used. Neither the knowledge questionnaire nor the performance indicators used in the chart review underwent any psychometric testing: The tools were custom designed for the study. Although we believe the measures had face validity and the advantage of the multidisciplinary team’s input, a better measurement tool may have yielded different results.

Our study presents preliminary evidence that strengthening occupational therapists’ mental health knowledge and skills can have a positive impact on older adults’ mental health. Despite evidence that rehabilitation professionals are not particularly adept at recognizing cognitive and affective conditions (Ruchinskas, 2002), our study shows that occupational therapists’ awareness of late-life depression can be increased and that incorporation of simple screening tools is one way to do this. Although our study did not include a mechanism for independent confirmation of undetected mental health problems that were subsequently treated with beneficial outcomes (we documented increased screening and reporting behaviors only), we had sufficient anecdotal evidence to conclude that therapists identified previously undisclosed mental health issues that their patients wanted to talk about and for which they wanted to find answers. This achievement is meaningful.

Time pressures and competing clinical demands are real phenomena, and therapists’ ability to achieve a reasonable level of what Lysaught Altschuld, Grant, and Henderson (2008) have termed competency maintenance behaviors remains an open question. What measures need to be in place for therapists to keep their professional knowledge and skills current? In support of the literature, our study findings suggest that institutional leaders, employers, and workplace supports play a critical role. Without the mental health DVD training coming to them and being designed to fit tight schedules and unique workplace circumstances, we believe we would have had less success.

Future research is needed on how best to strengthen professional competencies related to late-life depression. More rigorous research designs and measurement of salient factors, such as therapists’ experience and barriers and supports in the workplace, are needed. Future studies would also benefit from more comprehensive measurement of occupational therapists’ practice behaviors, including how therapists alter their practices when efforts in one domain are expanded. Future research in home health settings is also urgently needed. As our study showed, home health environments are not always stable or satisfying work environments. Factors that contribute to staff turnover may well be the same factors that challenge the delivery of quality services to patients. Finally, larger studies will be needed to study the sustainability of any new educational interventions over longer time frames and in ways that can measurably demonstrate cost-effectiveness and improved outcomes for older adults themselves.

We believe the educational intervention in DVD format designed to strengthen occupational therapists’ knowledge and skills achieved its intended goals. Occupational therapists in both the nursing home and the outpatient rehabilitation settings learned new information about mood disorders in older adults and they transformed their clinical practices to include screening for depression and cognitive impairment. They also provided, on a more frequent basis than before the DVD intervention, treatment that included behavioral activation and facilitation of more pleasant events in their patients’ lives. Our findings suggest that when clinicians identify brief evidence-based assessment tools and practical therapeutic techniques, they will be used. Establishing the effectiveness of an educational intervention to bolster practice skills is one of many preliminary steps on the path to improving mental health outcomes for older adults.

We acknowledge the funding support of the Retirement Research Foundation (Grant 2006–024). We thank Annmarie Cano, Gerry Conti, Allon Goldberg, Joe Pellerito, Jr., Fredrick Pociask, Mary Beth O’Connell, Stacey Schepens, Cheryl Deep, Brian Golden, Chris Lacavoli, Jan Gregory-Geoffrey, Reyna Columbo, and Pam Poteete. We also thank the occupational therapists who participated in this research.