Using a Positive Deviance Framework to Learn about Successful Local Health Department Billing Practices

Using a Positive Deviance Framework to Learn about Successful Local Health Department Billing Practices

ASPE Generic Information Collection Request

OMB No. 0990-0421

Supporting Statement – Section A

Submitted: August 24, 2015

Program Official/Project Officer

Amanda Cash

Senior Health Policy Analyst
U.S. Department of Health and Human Services

Office of the Assistant Secretary for Planning and Evaluation

200 Independence Avenue SW, Washington DC 20201

202.260.0362

Amanda.Cash@hhs.gov

Section A – Justification

1. Circumstances Making the Collection of Information Necessary

Background

Historically, many local health departments (LHDs) have been an important provider of clinical services, particularly in rural and underserved communities; however, state and local health departments are under increasing pressure to discontinue clinical services and re-orient services towards population based public health services. Further complicating the ability for LHDs to maintain services are health care reform implementation, Medicaid Managed Care (MMC) expansion, and the turbulent state of Medicaid expansions. Although health care reforms will extend health insurance to more individuals, provider supply may remain constrained, particularly in rural areas where issues around access have not necessarily changed. The extent to which the existing provider infrastructure has the ability to absorb increased demand remains to be seen, particularly in rural, underserved communities with a limited safety net. This exploratory project has two aims: first, we propose to use a mixed-methods, positive deviance approach to identify and learn from those LHDs who have been able to continue providing clinical services in the current policy environment, and second, we propose creating a tool or instrument to monitor any changes over time. Validation of whatever questions or tool we create from this work will occur after this project is finished. We aim to examine where supply of clinical services delivered by LHDs are meeting potential demand for these services as well as any policy or programmatic changes that happened to ensure the delivery of clinical services by LHDs.

2. Purpose and Use of the Information Collection

The aims of this project are to:

• Examine the geographic distribution of clinical service provision for clinical services among LHDs.

• Identify the contextual factors conducive for LHDs to operate as a clinical service provider, particularly in highly vulnerable communities (positive deviants).

• Identify contextual factors that may have lead LHDs to stop or change their clinical service delivery patterns.

• Determine the internal and external mechanisms used by positive deviant LHDs to support clinical service provision.

Aims 1&2: The 2013 State Profile Survey data from the National Association of County and City Health Officials (NACCHO) will be used to examine the geographic distribution of clinical service provision among LHDs. These data will be linked with the Area Resource File (ARF) to identify additional county-level contextual factors reflective of the need for LHDs to act as a clinical service provider. A subset of the linked file specific to states that have implemented Medicaid Managed Care will also be examined to identify LHDs who have maintained clinical service provision within a Managed Care environment. We will use the ARF to identify highly vulnerable counties based on population demographics, provider capacity and access to primary care services and examine the role of LHDs within these counties. Positive deviants will be identified as LHDs in high-need areas that that have maintained clinical services.

Aim 3: We will conduct 1-hour semi-structured telephone key informant interviews with staff from positive deviant LHDs. Of particular interest are the strategies by which LHDs have maintained clinical services. Informants will be in a supervisory role in the LHD with knowledge of the finance and reimbursement systems. All notes will be coded in Atlas.ti within 24 hours of each interview, and thematic analysis will be used to identify consistent strategies used by positive deviant LHDs that can be shared with similar LHDs struggling to maintain clinical services.

After identifying strategies implemented by positive deviant LHDs, we will share these lessons with other LHDs in order to assist them in maintaining clinical services in a changing environment. In addition to scholarly papers, we will write a 1-2 page brief outlining specific activities local health department leaders can take to maintain services in their communities. The brief will have a link to a full report of our work.

This work will not be used to inform policy decisions; it is exploratory in nature. We will use the results to begin to identify and potentially validate questions that may capture some of the information we will collect in this study to monitor any trends over time.

3. Use of Improved Information Technology and Burden Reduction

Data will be collected via telephone interviews. We will use computers to take notes and qualitative data analysis software (Atlas.ti) to conduct data analysis.

4. Efforts to Identify Duplication and Use of Similar Information

To our knowledge, there is no information that has been or is currently being collected similar to these. This is an exploratory study to answer questions that we currently do not have the data to answer.

5. Impact on Small Businesses or Other Small Entities

No small businesses will be involved in this data collection.

6. Consequences of Collecting the Information Less Frequently

This request is for a one time data collection.

7. Special Circumstances Relating to the Guidelines of 5 CFR 1320.5

There are no special circumstances with this information collection package. This request fully complies with the regulation 5 CFR 1320.5 and will be voluntary.

8. Comments in Response to the Federal Register Notice and Efforts to Consult Outside the Agency

This data collection is being conducted using the Generic Information Collection mechanism through ASPE – OMB No. 0990-0421.

9. Explanation of Any Payment or Gift to Respondents

We will not be providing incentives for this study.

10. Assurance of Confidentiality Provided to Respondents

We are not asking any personally identifiable information of respondents, but rather only about their experience in their professional capacity. All data will be de-identified so as not to reveal the respondent.

11. Justification for Sensitive Questions

We will not be asking any questions of a sensitive nature.

12. Estimates of Annualized Burden Hours and Costs

The key informant interviews will take approximately one hour to complete.

Table A-12: Estimated Annualized Burden Hours and Costs to Respondents

Type of Respondent	No. of Respondents	No. of Responses per Respondent	Average Burden per Response (in hours)	Total Burden Hours	Hourly Wage Rate	Total Respondent Costs
Local Health Department Staff	40	1	1	40	$34.21	$1,368.40
TOTALS	40	40		40		$1,368.40

13. Estimates of Other Total Annual Cost Burden to Respondents or Record Keepers

There will be no direct costs to the respondents other than their time to participate in the data collection.

14. Annualized Cost to the Government

Table A-14: Estimated Annualized Cost to the Federal Government

Staff (FTE)	Average Hours per Collection	Average Hourly Rate	Average Cost
Social Science Analyst, GS 11	20	33.00	$660
Social Science Analyst, GS 15	20	76.00	$1,520
Estimated Total Cost of Information Collection			$2,180

15. Explanation for Program Changes or Adjustments

This is a new data collection.

16. Plans for Tabulation and Publication and Project Time Schedule

Aim 1: Examine the geographic distribution of clinical service provision for maternal and child health services among LHDs.

The 2008, 2010, and 2013 State Profile Survey data from the National Association of County and City Health Officials (NACCHO) will be used to examine the geographic distribution of billable clinical service provision among LHDs. These data have been requested directly from NACCHO, along with more specific identifiers not included in the publically available files. These identifiers will be used to crosswalk LHD zip codes and county FIPS codes for each of the LHDs included in the sample.

Within the NACCHO profile data, we will examine responses from the Core profile sent to all LHDs that provide insight into 1) the current landscape of clinical service provision, 2) how this has changed during 2008-2010 and 2010-2013, and 3) LHDs that remain consistent providers of clinical services and currently receive reimbursement for clinical services. Individual LHD responses over the three waves will be linked, and changes in the following variables will be examined.

• Sources of Revenue (Medicaid & other clinical)

• Activities performed by LHDs (most relevant to clinical services and billing capacity)

  • Family planning

  • Prenatal care

  • Obstetrical care

  • EPSDT

  • Well child clinics

  • Comprehensive primary care

  • Home healthcare

Additional time-invariant variables will also be examined and include:

• State

• Governance classification/structure

• Region or county reporting classification

• Organization structure

• Population size served by LHD (7 level categorical variable)

The primary limitation of linking multiple waves of profile data and using crosswalk files to determine zip codes and county FIPS codes is missing data. Consistent responses across LHDs for all three waves remain an unknown at this point. However, should this become an issue we will consider using 2 time points, or possibly limiting the analysis of the above-mentioned variables to the 2013 profile. The ability to measure change longitudinally may be compromised by missing data, but the ability to identify LHDs providing clinical services and recouping some form of reimbursements is not.

Aim 2: Identify LHDs that operate as a clinical service provider for maternal and child health and other reimbursable services, particularly in highly vulnerable communities (positive deviants).

In the context of this study, positive deviants will be defined as those LHDs that have maintained clinical services provision over time—particularly those operating in highly vulnerable communities with limited primary care capacity. A two-tiered stratification approach will be used to identify and contextualize the role of LHDs as a clinical service provider.

The first level of stratifications relates to Medicaid Managed Care (MMC) and Medicaid Expansion under the Affordable Care Act. Given the challenges of maintaining clinical services provision in the context of MMC, the analysis will be stratified by states that have implemented a robust Medicaid Managed Care initiative and those that do not currently have a Managed Care initiative. This will allow for examining changes within a varying policy environment that may influence LHDs ability to provide clinical services. Also relevant to the analysis are the states that expanded Medicaid or implemented an alternative program under the 1115 Demonstration Waiver. State decisions related to Medicaid expansion reflect an additional layer of complexity that could ultimately influence the role of LHDs as a clinical service provider. We are specifically interested in identifying the mechanisms that allow LHDs to bill for clinical services within this challenging environment.

The second tier of stratification provides an additional layer of context for comparing LHDs that are similar to one another by considering area deprivation, community vulnerability, and health system capacity. These factors are conductive for examining the intersection of need or demand for clinical services with the role of LHDs as a clinical service provider in these communities.

An area-deprivation index developed by the co-investigator (Dr. Hale) will be used to characterize the underlying level of vulnerability of the communities in which the LHDs operate—particularly as is relates to clinical service provision. The index collapses multiple social determinants into a single measure that can be interacted with other variables of interest (e.g., rurality) to examine selected outcomes of interest within comparable levels of vulnerability. To derive the index, the NACCHO data files will be linked with the Area Health Resource File (AHRF) and includes the following measures:

• Income (median per capita income, f1322611)

• Poverty (percent of population below 100% poverty, f1332111)

• Education (percent of population with no high school diploma, f1448006)

• Unemployment (percent unemployed, f0679511)

• Single parent homes (percent female head of household, f0874610)

In addition, measures of adequate primary care capacity also derived from the AHRF will also be included to examine the underlying health system capacity in a given county served by the LHD (Health Professional Shortage Area designation, the presence of at least 1 Federally Qualified Health Centers and/or Rural Health Clinic, and primary care physicians per capita).

The data examining the role of LHDs as a clinical service provider will be stratified by the contextual factors noted in the deprivation index and embedded on these larger policy environments (MMC/Medicaid expansion). A series of tables/cross-tabulations of clinical service provision within each level of the contextual factors of interest will be provided and used to identify positive deviants. Examining the role of LHDs as a clinical service provider within these three contexts allows for the identification of positive deviants that are drawn from comparable contextual environments. Given the aggregate nature of the NACCHO profile data and the scope of the study, these analyses are sufficient to adequately identify a sample of positive deviants.

This approach to identifying positive deviants also has limitations. Although the ability to identify positive deviants based on the provision of clinical services is not compromised, using county-level data to characterize the environment in which LHDs operate could exclude some valuable information. LHDs can be county, city, township, or multi-jurisdictional, and aggregating community vulnerability to the county level may not fully reflect the true underlying level of vulnerability or primary care capacity in communities served by LHDs. (For example, approximately 8% of LHDs responding to the 2013 profile survey were part of a multi-county jurisdiction.) As these LHDs are identified, data for counties within the jurisdiction will be combined to better approximate the underlying level of vulnerability. In addition, the analysis will also be limited to only county health departments (60% of all LHDs) and findings compared to the total sample. This approach will provide a general sense of larger circumstances in which LHDs likely operate.

Aim 3: Develop an Interview Protocol to better understand the practices used by positive deviant LHDs to maintain reimbursable services.

We will develop a guide for key informant interviews to be administered to staff from LHDs identified as positive deviants. We do not know how many positive deviants (PDs) will be identified; however, it is likely that there will be more PDs than we are able to interview. As such, we will identify the interview participants by stratifying PDs by contextual factors as outlined above in order to get as varied a sample of interviewees as possible. We anticipate a sample of 30-40 LHDs to be targeted for interviews.

We will develop a telephone interview protocol (not more than 1 hour in length) that will focus on the following topics: types of services provided, changes in service provisions over the past 5 years, LHD funding sources, changes in funding over the past 5 years, interactions with payers (Medicaid, private payers, etc.), and challenges to service provisions and reimbursement, among other topics. Dr. Klaiman will recruit and convene an expert panel of 3-5 people with expertise in LHD services and billing for feedback on the interview protocol prior to implementing it. She will also pilot the interview protocol prior to administering it to participants.

Timeline:

Completion Date	Major Tasks/Milestones
August 2015	Submit request for OMB approval under an existing generic PRA clearance Submit project for IRB Approval Recruit Expert Panel Link Datasets Draft Interview Protocol; send to Expert Panel for review
September 2015	Complete positive deviant identification Receive Expert Panel feedback for Interview protocol Receive OMB approval under an existing generic PRA clearance Receive IRB Approval
September 2015	Draft list of positive deviants sent to MPR Draft interview protocol sent to MPR Receive comments from ASPE on draft list of positive deviants and draft interview protocol Final list of positive deviants sent to MPR Final protocol sent to MPR
October 2015	Identify LHDs for Interviews
November 2015 – April 2016	Conduct Interviews
April – July 2016	Qualitative Data Analysis
August – October 2016	Write-up/disseminate Results

17. Reason(s) Display of OMB Expiration Date is Inappropriate

We are requesting no exemption.

18. Exceptions to Certification for Paperwork Reduction Act Submissions

There are no exceptions to the certification. These activities comply with the requirements in 5 CFR 1320.9.

LIST OF ATTACHMENTS – Section A

Note: Attachments are included as separate files as instructed.

1. Interview protocol (DRAFT)