C6. Level of Integration Measure*

Relevant Functional Domains from the Integration Framework: 

Developer: 

Antioch University

Purpose: 

To rate the degree to which behavioral health providers or behavioral health care is integrated into primary care settings from the perspective of staff and/or providers. 

Date: 

2010

Relevant submeasures: 

None

Format/data source: 

The Level of Integration Measure (LIM) can be answered from various perspectives (e.g., physician, behavioral health specialist, nurse, staff). Respondents will rate the 35 items on a 4-point Likert type scale (1=never; 4=always). Higher scores on the LIM will reflect higher levels of integration. The LIM will include six domains: (1) clinic system integration (“This clinic has a behavioral health specialist on site”); (2) beliefs and commitment (“I believe in integrated care”); (3) clinical practice (“We have a system for referrals to the behavioral health specialist”); (4) interdisciplinary alliance (“Our behavioral health and primary care providers trust each other”); (5) training and consultation (“Our behavioral health and primary care providers learn from each other”); and (6) leadership (“Our clinic has an integrated care champion”).

Development and testing: 

The developer received responses to the LIM, Patient Satisfaction Questionnaire (PSQ), and Measure of Job Satisfaction (MJS) from 36 primary care providers and 82 staff in four clinics. While this is a relatively small sample for psychometric analyses, the developer was able to confirm strong internal consistency for the full scale of the newly developed, 35-item LIM (Cronbach’s alpha of .95), and adequate internal consistency for six of its seven intended subscales (alphas ranging from .71 to .92). Results of this initial exploration indicate that the seventh LIM subscale—three items addressing shared decisionmaking among clinic personnel—should be dropped (alpha = .24). The PSQ measure has four subscales, which yielded Cronbach’s alphas between .74 and .88, well within acceptable range. The MJS also achieved adequate reliabilities, with Cronbach’s alpha ranging from .68 to .89 across its five subscales and .97 for the full scale, comparable to the psychometric data reported by the measure’s developers.

There are no established instruments to assess the level of behavioral and physical health integration in primary care. Therefore, the developer decided to create the LIM based on literature, consultation with experts in the field, and feedback from our clinic partners. First, the developer networked with integrated care experts and our integrated care clinic partners to collect any existing nonpublished measures, as well as to generate an initial list of integrated primary care domains. Second, the developer generated a list of 49 potential LIM items from existing measures, integrated care experts, and brainstorming. Third, the developer collapsed or deleted redundant items and edited the remaining items for clarity, consistency, and readability. Fourth, the developer sought quantitative and qualitative feedback on the importance and quality of each item from integrated care academics, trainers, clinicians, and administrators from around the country. Based on this feedback, the developer edited the LIM again, leaving the current 35-item version. The developer presented preliminary psychometric properties based on the pilot, and has adjusted the LIM accordingly.

Past or validated applications: 

  • Setting: Behavioral health care
  • Population: Behavioral health care providers and/or staff
  • Level of evaluation: Individual or groups of health care professionals.

Sources: 

Personal communication with measure developer

Notes: 

For additional information, please contact:

James Fauth, Ph.D.

Associate Professor, Department of Clinical Psychology

Director, Center for Research on Psychological Practice

Antioch University New England

40 Avon St.

Keene, NH 03431

e-mail: jfauth@antiouch.edu.