Present: Abby
Crocker, Kairn Kelley, Rodger Kessler, Ben Littenberg, Charlie MacLean, Connie
van Eeghen
1.
Start Up: Honeybell
tangelos can be found on Given Courtyard 4th floor South, if you ask
the right person.
2.
Future CROW
group project
a. In
a recent CROW session, the group decided
to learn more about multi-level
modeling through a team-based research study. The ideal project will
result in a publication. The team is considering a research topic centered on opiates,
categorized by long acting/short acting prescription choices (or no pain
management prescription at all) for opiate-naïve adults less than age 65 who
have had a musculo-skeletal injury. Our
question is whether there is a difference in ER utilization based on category
of opiate and dosage (in morphine equivalents).
b. Using
VHCURES (no Medicare yet), we are limited to what was paid for by insurance
over the last five years (started in 2007).
Other issues to keep in mind:
i.
Provider IDs are difficult, as they have more than one
name assigned and more than one MPI. One
solution is to build a master unique ID list that connects all assigned names
and MPIs. Also, claims do not identify
referring providers, so inferences have to be made about what visit caused what
procedure.
ii.
Patient IDs are very difficult: every insurer and prescription
management company uses a different unique ID for its subscribers/members. Sometimes they overlap (encrypted SSN or name
or encrypted subscriber SSN or name), but since encrypted fields have to be
exact matches, this doesn’t help if all we have are names. Otherwise some fuzzy matching will be needed,
including date of birth, zipcode, and clinically related data (diagnosis or
procedure with prescribed medication).
Medicaid is simpler as it pays for its own pharmaceuticals, but harder
because member turnover is thought to be high.
c. Connie
received copies of the data dictionary (the data sets) and the field values
(the codebook). She will share these
with the group. Next step: Steve Kappel will
come to CROW on Feb 7.
3.
Presentation: Abby’s
policy paper on breastfeeding for babies with opiate exposed mothers. Abby asked the group to help complete a
conceptual framework for a paper based on the socio-ecological model to
identify breastfeeding influencers, barriers, and aids. Opiate dependent mothers have a much lower
rate of breastfeeding than the national rate (76%) and current protocols don’t
respond to this specific issue.
Completing the model requires considering both the mother and NAS
(neonatal abstinence syndrome) infant.
a. Basic
framework of nested categories, starting with the largest: (items in red are NAS specific)
i.
Social (policy and cultural norms):
1. Breast
feeding in public
2. Breast
feeding at work
3. Insurance
reimbursement
a. WIC
b. VNA
home visits?
4. Community
attitudes towards breastfeeding
ii.
Community (organization):
1. Hospital
a. Staffing
challenges for lactation support
b. P4P
(lack of incentive – may be an aid, rather than a barrier)
c. Structural:
policies & procedures; standing orders
i.
Referral bias, i.e. the
OB high risk
d. Cultural:
healthy baby initiative hospital, breastfeeding friendly, default bottle
feeding)
e. Education/staff
2. Health
care providers (OB, Pediatrician, SA counselor, SW…)
a.
Attitudes & skills
b.
Uncertainty of mother’s ongoing
use of drugs
3. Post-discharge
a. Home
health visits – avoided by mothers of NAS infants
4.
Transportation –
multiple NAS related appointments; trade-offs with managing work, multiple
appointments and breastfeeding
iii.
Interpersonal (relationships; emotional support):
1. Family
a. Attitude
2. Friends
3. Spouse
iv.
Individual:
1. Baby:
a.
Medical (preemie)
b.
NAS (irritability)
2. Mother
a.
Active user of drugs
b. Attitudes
& knowledge
c. Insufficient
nutrition (500-600 calories/day extra)
d. Fear
of relapse
b. Visual
interpretation of this model:
c. Each
barrier and issue can be addressed in a solution set for policy interventions
in a “bundled way”
i.
Coordination among providers (including lactation
support) with transportation support; preemies in hospital bundle; PCMH based
program
ii.
Social norming: Group meetings for new mothers;
creating expectations r/t breastfeeding
iii.
Consider a hierarchy of interventions: if the provider
doesn’t like it, it won’t happen
iv.
Consider two phases – in which case the model is
different for each phase
1. Breastfeeding
initiation – in the organization; connection with post-discharge support groups
and providers
a. Preemie
initiation
b. Non-preemie
initiation
2. Breastfeeding
maintenance – in the community
a. Early
(1st 4 weeks)
b. First
6 months
c. First
year
d. Structure
of paper
i.
Breastfeeding is good; national rates; NAS rates
ii.
Identify issues for this group (the red issues above); issues are specific to NAS beyond those
already known as barriers to breastfeeding
iii.
Recommended policy interventions for NAS population (as
opposed to policy interventions beyond the NAS population?); a checklist for
all the variables/issues for providers/organizations (a branching logic
flowsheet)
1. Mothers
in a treatment program
2. Mothers
not in a treatment program (who are these: high risk but not anymore? Or
relapsed? Needs regular monitoring?)
a.
Jan 31: Kairn: F31 update
b.
Feb 7: Steve Kappel: Understanding/using VHCURES
c.
Feb 14: Abby:
d. Future
agenda to consider:
i.
Christina Cruz, 3rd year FM resident with
questionnaire for mild serotonin withdrawal syndrome?
ii.
Peter Callas or other faculty on multi-level modeling
Posts
