Supporting Statement Part B (05.29.14)

SUPPORTING STATEMENT

Part B

Taking Efficiency Interventions in Health Services Delivery to Scale

Version: May 29, 2014

Agency for Healthcare Research and Quality (AHRQ)

Table of contents

B. Collections of Information Employing Statistical Methods 3

1. Respondent Universe and Sampling Methods 3

2. Information Collection Procedures 4

3. Methods to Maximize Response Rates 4

4. Tests of Procedures 4

5. Statistical Consultants 5

B. Collections of Information Employing Statistical Methods

1. Respondent universe and sampling methods

Five practice sites out of 15 that are part of Penobscot Community Health Care (PCHC) will be participating in the study. They are the five largest sites and strategically selected by PCHC to implement the delegate model based on existing staffing models. The universe of respondents varies by data collection methodology as outlined below:

1) Team Survey (TS) – The universe for the TS is all members of all primary care teams in all five participating sites. All primary care providers within the five practice sites work in teams. The composition of the teams may vary within site and across sites. All members of all primary care teams in all five sites comprise the universe for the TS.

2) Key Informant Interviews (KII) – The universe of key informants for the site visits includes the Medical Director, Practice Director, members of primary care teams implementing the delegate model or future implementers, and ancillary staff not part of the team but associated with the work of the team. The Medical Director and the Practice Director are considered “essential” respondents. A convenience sample will be drawn from the members of the primary care teams. The data being gathered is qualitative and patterns and themes of response tend to converge after a relatively small number of interviews (see assumptions in table below). The universe of key informants for the phone calls includes the Medical Director and Practice Director of each participating site.

	Site #1	Site #2	Site #3	Site #4	Site #5	Total
Team Survey
Universe	33	28	49	32	22	164
Anticipated response rate	80%	80%	80%	80%	80%
Anticipated number of responses	26	22	39	26	18	131
Key Informant Interviews (Site visit)
Universe1	29	19	26	19	14	107
Anticipated number of responses2	17	10	12	10	8	57
Key Informant Interviews (Phone calls)
Universe	2	2	2	2	2	10
Anticipated number of responses	2	2	2	2	2	10

¹Assumes approximately half of the providers adopt the delegate model.

²Assumes that Practice Director and Medical Director, half of all primary care providers, one-third of Medical Assistants, and all of ancillary staff will be interviewed.

2. Information Collection Procedures

This exploratory study primarily relies on qualitative data, with limited statistical analyses of quantitative data. The intent of the study is to examine factors that enable or hinder spread of an intervention, and to define and explore the features of that intervention so that others may learn from it and potentially adopt it.

Respondents will be notified of all data collection processes by PCHC staff and encouraged to participate by PCHC staff. The Team Survey will be distributed electronically to the universe of respondents directly by JSI at two points in the study. No sampling is required. The TS will remain in the field for one month. One reminder email will be sent at the two-week mark to non-responders to remind and encourage response.

The site visits for the KIIs will be arranged by JSI directly with the Practice Director of each site. A protocol for the visit and a summary of information to be collected will be forwarded approximately one month before each visit. The Practice Director will arrange interview times with all key informants during the site visit according to the protocol. While individual interviews will be conducted with the Practice Director, Medical Director, at least two providers, and at least two Medical Assistants, group interviews may also be conducted for groups of providers, groups of Medical Assistants, and/or groups of ancillary staff.

The phone calls for the KIIs will be arranged by JSI directly with the Practice Director at a time that is convenient for both the Practice Director and Medical Director to attend. A summary of information to be collected will be forwarded approximately two weeks before the call.

3. Methods to Maximize Response Rates

The Chief Quality Officer at PCHC will encourage all targeted respondents to complete their relevant data activities. For the TS, JSI will send out a reminder email to non-responders approximately half-way through the period that the survey is in the field.

For the KIIs, flexibility in timing is essential. JSI will work with the Practice Directors approximately three months in advance of the intended site visit to make sure that there is ample time to arrange schedules for interviews. Techniques that have worked especially well in the past are to conduct provider interviews early in the morning before appointments, during the lunch hour, and/or after all appointments are completed for the day.

4. Tests of Procedures

The TS will be the only instrument that will need to be tested. It will be tested on one provider and two Medical Assistants. The results of this pilot will be assessed and the instrument revised as needed. If substantial changes are required, another pilot of two providers and two Medical Assistants will be conducted.

5. Statistical Consultants

JSI has a statistician on the project. Contact information is as follows:

Eugenie Coakley, MPH, MA

Accredited Professional Statistician™, 2013–2019

JSI

(617) 482-9485.

5