Against the Odds: Contextual Predictors of Clinically Competent Dialysis Social Work

Abstract

Purpose: Type II diabetes has reached epidemic proportions in the US, and kidney failure is increasing concurrently. Research shows that clinically competent dialysis social work, as defined by completion of core tasks articulated in the CMS conditions of coverage (psychosocial evaluations, casework and counseling for patients, group work, patient education, and treatment team consultation), results in markedly improved outcomes. However, surveys of dialysis social workers show that a significant majority of respondents have difficulty attending to these key tasks because of other demands on their time. This article aims to explore contextual predictors of core task completion in dialysis social work.

Method: Data was gathered from members of the Council of Nephrology Social Work (N=186). The survey instrument consists of forty-five questions including the type of employer; number of hours scheduled; number of hours spent on core tasks; number of hours spent on non-core tasks; caseload in terms of number of units, number of patients, and type of patients (hemodialysis vs. peritoneal dialysis); the range and most common socio-economic status and education level of the respondent's patients; whether the respondent's primary clinic is rural, urban or suburban; whether and to what extent clerical support may be accessed; staffing ratios; location of clinic by state; management of the social work position; whether the respondent is practicing in isolation from social work colleagues; the level of experience of the respondent as well as his or her tenure at the current primary clinic.

Results: The study uses simple regression and hierarchical linear modeling to test hypotheses around human capital, tertius gaudens, and social network factors as predictors of perceived ability to complete core tasks. Personal human capital factors played no role (p = .824). Patient population (tertius gaudens) plays a role (p=.006). Social network factors were found to be key. A regression model using four of the social network hypothesis variables (employer type p = .004, number of hemodialysis patients p = .023, hours spent on core tasks p = .001, and hours spent on non-core tasks p = .090) and one the tertius gaudens hypothesis variable (presence of lower middle class patients p = .042) was statistically significant to the p <.001 and explained twenty-two to twenty-six percent of the variance observed in the sample. The interclass correlation coefficient showed that nineteen percent of the variance could be accounted for by level two variables identified by employer type, justifying the use of hierarchical linear modeling to gain a more in depth picture of how variables are affecting the outcome.

Implications: Findings emphasize the need to look beyond individual case factors into systemic issues that promote or impair the delivery of dialysis social work services. Interventions must target multiple factors in order to be effective.