Practices of Depression Care in Home Health Care: Home Health Clinician Perspectives

We sought to recruit administrators and nurses from five HHC agencies in five states in the Northeast, South, and Midwest between February and November 2012. These five agencies had recently participated in a nursing team–randomized trial of the Depression CAREPATH (Care for Patients At Home) intervention for older HHC patients (9,10). We determined three types of key informants to provide complementary perspectives: HHC nurses on a CAREPATH intervention team and HHC nurses on a usual care team, nurse supervisors of CAREPATH intervention teams, and medical or clinical directors who played a leadership role in facilitating their agency’s participation in CAREPATH.

A liaison at each agency identified potential informants, whom the research team then contacted by e-mail or phone to recruit and schedule phone interviews. No potential informants we approached refused to participate. Each provided signed written consent and received a $75 gift card for participating. The study received approval from the Weill Cornell Medical College Institutional Review Board.

The research team developed semistructured interview guides for different types of informants that covered several domains, including typical practices of depression care prior to CAREPATH. [The guides are included in an online supplement to this report.] The guides were pilot-tested with a nurse, a nursing supervisor, and a director from agencies that did not participate in CAREPATH and iteratively revised throughout the interviews.

The first author conducted one-on-one telephone interviews with all informants. Audio-recorded interviews were transcribed by a third-party service and analyzed by using grounded theory method (11), an established qualitative analytic approach by which researchers generate overall themes by iteratively reading and reviewing interview text, attributing labels (codes) to salient text, and documenting recurring themes that “emerge.”

Two investigators (YB and AAE) independently coded each transcript and met to discuss the definitions of codes and negotiate consensus. To enhance analytical rigor, a third investigator (JER) independently coded selected transcripts and participated in meetings to provide an outside voice and help negotiate consensus. The process was documented throughout and led to the development of a codebook and themes.

Data were organized and managed by using the qualitative analysis software Nvivo 9 to facilitate text and code consolidation and, in turn, theme development.

We conducted interviews with six directors, five supervisors, and nine nurses. All but one were women. Average years of HHC experience were 16 for directors, nine for supervisors, and ten for nurses. Average length of interviews was 45 minutes (range 33–56 minutes).

We grouped currently known best practices of depression care into five clinical functions: screening, assessment, case coordination, antidepressant management, and patient education and goal setting (9). Participants reported that all five functions overlapped with routine HHC nursing. However, based on their description of typical practices prior to CAREPATH, gaps of various extents existed between typical and best practices in each of the functional areas, which we detail below. [A table in the online supplement provides a summary.] The best-practice protocol recommends initial screening with the two-item Patient Health Questionnaire (PHQ-2), a depression screening tool included in the mandated, start-of-care (SOC) patient assessment for all Medicare HHC patients. Several agency directors and supervisors believed that the addition of the PHQ-2 greatly enhanced clinician awareness of depression. However, some informants suggested that the typical approach by nurses may have been inadequate or ineffective in detecting depression. Participants noted that nurses would often “go right in and just read off the answers” rather than conduct a clinically informed interview. Although there was a perception among nurses that it might take several weeks or visits before a patient would feel comfortable talking about depression and that the SOC visit was usually too involved to do a good screening, none of the five agencies had a clinical management system that prompted nurses to reassess PHQ-2 at follow-up visits.

For patients who screen positive on the PHQ-2 (a score of ≥3), the best-practice protocol recommends further assessment of depression severity with, for example, the PHQ-9. However, such practice was adopted before CAREPATH by only one of the five agencies. For patients with clinically significant depression (a PHQ-9 score of ≥10), the protocol recommends that HHC clinicians follow the course of patients’ symptoms each week or each visit by using the PHQ-9 to support case coordination, antidepressant management, and patient education. However, prior to CAREPATH, none of the five agencies practiced weekly follow-up assessment with the PHQ-9.

Consistent with the best-practice protocol, HHC nurses in our study routinely reported to physicians about patients’ depression as they would for other conditions. However, they also reported issues related to nurse-physician communication. A handful of nurse informants expressed frustration about not being able to communicate with patients’ physicians in a timely manner and indicated that they mostly spoke with nurses and receptionists at a physician’s office and rarely with the physician in person. This seems to be a general issue that does not pertain specifically to depression. Primary care physicians (PCPs), with whom nurses most often communicated, were reportedly receptive to HHC nurses’ reports but often wanted to evaluate the patient themselves before making a decision (for example, ordering a new antidepressant). Accordingly, nurses perceived their roles as raising PCPs’ awareness of an issue and facilitating independent evaluation by PCPs—for example, scheduling an office visit for the patient, which is consistent with the best-practice protocol. However, our results suggest that visits to PCPs alone did not guarantee adequate attention to depression, because, according to our informants, older patients might be “afraid or not sure how to ask the doctor or tell the doctor [about their depression].” Several informants believed that HHC nurses had a role in educating and coaching patients to make the most out of their visits.

Nurse informants from all five agencies reported that prior to CAREPATH, depressed patients were generally referred to social workers at their agencies. Nurses believed that social workers were better trained and more effective at addressing patients’ psychosocial issues. Nurses also perceived that referral to social workers lessened burdens for clinicians “going in and dealing with the medical piece.”

Meanwhile, care coordination between nurses and social workers appeared to be limited once patients were referred. Several nurses mentioned reading social workers’ clinical notes on shared patients, but routine integration of social workers into the weekly nursing team case conference—an important mechanism for case management and coordination in HHC—was rare. In general, nurse informants expressed little knowledge of the types of interventions social workers conducted with depressed patients. In addition, a general perception shared by informants from four of the five agencies was that their agencies had too few social workers to meet the plethora of psychosocial needs of patients.

HHC nurses reported that they conducted patient education and medication reconciliation (by comparing the patient’s list of current medications with the list of physician orders) for antidepressants as they did for other medications. Although nurses reported that many patients had significant depressive symptoms even though they were taking antidepressants, prior to CAREPATH nurses largely did not consider antidepressant management—the process of finding an effective regimen—within their purview. Several nurses believed that “it’s up to the doctor to change it [antidepressant] if it is required.”

Educating patients about depression and its treatment and assisting patients with setting personal goals is highly consistent with routine HHC and was strongly endorsed by our informants. The informants also perceived a role in connecting patients with community resources (for example, senior centers) to increase social interactions and pleasurable activities for depressed patients.

In regard to barriers, our results indicated that prior to CAREPATH, HHC nurses typically did not consider depression care within their scope of practice and preferred to have another provider, such as a social worker or the PCP, take full responsibility. Almost all supervisors and agency directors believed that HHC as a profession attached greater importance and urgency to general medical conditions than to depression or other mental health conditions. Nurses saw themselves as task oriented and their roles as primarily delivering skilled nursing care.

Lack of knowledge and training among nurses to adequately recognize and treat depression may have led to low self-efficacy in addressing depression. Several agency directors and supervisors believed that prior to CAREPATH, their nurses commonly held misconceptions about depression and depression treatment—for example, that depression was a normal part of aging and that if a patient was taking antidepressants, no further management by HHC clinicians was needed. When reflecting on their practices prior to CAREPATH, several nurses expressed a feeling of insufficiency—because they would not know what to do about it, they naturally avoided discussing depression with the patient.

Several informants also acknowledged a sense of uneasiness when assessing depression and engaging patients in depression care, which suggested stigma. For example, as a nurse described her experience, “I just felt uncomfortable asking too many questions. I tried to stay [as] positive as I could and so I didn’t go there.” An agency director believed that some nurses felt that “if they put it [depression] on the patient’s record, they’ve kind of given them [a] scarlet letter.”

Our informants observed an overall low level of antidepressant management undertaken by patients’ PCPs, who accounted for an overwhelming proportion of prescribing physicians in this population. Reportedly, PCPs tended to “never increase the dose” or not change medications for patients who were not responding to their current antidepressant. [The online supplement summarizes information about how the barriers map with quality gaps in each functional area of depression care.]

Using qualitative reports of HHC clinicians and administrators, this study found that although the clinical functions of depression care fit within routine HHC practice, gaps of varying extents existed between typical and best practices in all functional areas. These gaps were associated with HHC nurses’ beliefs about their scope of practice, lack of knowledge, low self-efficacy, and stigma in regard to depression and depression care; ineffective interclinician communication and collaboration (especially between the nurse and the PCP); and poor antidepressant management in primary care.

Our findings elucidate important factors pertaining to gaps in depression care and shed light on potential strategies to close them. Prior to the CAREPATH trial, HHC nurses’ limited knowledge and self-efficacy contributed to the low priority they assigned to depression and partly explained their inaction and tendency to delegate depression care to others. Interventions are needed that boost nurse self-efficacy and provide practical tools that enable them to play an active role in depression care in accord with best practices.

Our findings highlight PCPs as an important link in depression care quality improvement for HHC patients. Almost a third of elderly Medicare fee-for-service patients receiving HHC take antidepressants (12), and PCPs are responsible for most of their antidepressant prescriptions (13). Given that medication-related decisions ultimately rest with the physician and not the HHC nurse, PCPs’ lack of knowledge and self-efficacy in regard to antidepressant management (14) constitutes a major barrier. Adding to the concern are challenges in timely and effective HHC nurse–PCP communication: depression care management may be especially susceptible to communication breakdowns because of gaps in training and practice and lower priorities assigned to depression care. Future interventions should seek to engage clinicians at the PCP’s office in addition to addressing communication skills and effectiveness of HHC nurses.

Although this study focused on barriers pertaining to clinicians, addressing these barriers alone may not be sufficient to close the quality gaps. Our interviews revealed factors at the system and HHC agency levels that provide misaligned incentives for depression quality improvement in HHC (15). Although recent reforms such as accountable care organizations may help align incentives, explicit policy tools are needed to achieve accountability for mental health outcomes.

Our study had some limitations. By design, the qualitative data we collected reflected the (highly pertinent) perspectives of HHC nurses and administrators and not the perspectives of other stakeholders, such as HHC patients. The five agencies may have perceived a greater need than the average agency to improve the quality of depression care, indicated by their voluntary participation in a depression study. Finally, this research, like qualitative research in general, is not intended to be generalizable. However, the findings may be “transferrable” to other circumstantial or organizational contexts.

On the basis of qualitative reports by HHC nurses and administrators, gaps between typical and best practices existed in each area of depression care in HHC. Strategies to close the gaps should enhance HHC nurses’ knowledge and self-efficacy with regard to depression care, address the low level of antidepressant management by PCPs, and improve the quality of HHC nurse–physician communication.

The authors thank Catherine Reilly, M.P.H., for assistance with recruitment of study informants.