Goal

Consolidate findings and observations from Year 3 WHPP to create final report.

	Andrea Price Old Account (Deactivated)	Raymonde Uy (Unlicensed)	Julia Skapik (Deactivated)
Last year's 867-02 Final Report	reviewed - explains HCD framework and provides info.
867-02 Last year's Infographic	Use Cases Data Challenges - capture, structure, extraction Best Practices Innovation Strategies
867-03 11 Jun 2021 Harvest Meeting Slides	Reviewed.
867-03 11 Jun 2021 Harvest Meeting Transcript	Reviewed
OCHIN Final Report	Reviewed.
OCHIN Implementation plan	No separate plan - only what was included in the Final Report.
OCHIN Lessons Learned	Reviewed
OCHIN Personas	Final personas in final report. There is a reference to how new situation would change the experience for the persona.
AllianceChicago Final Report	Reviewed.
AllianceChicago Lessons Learned - not on web page	appendix requested
AllianceChicago Personas - Initial
AllianceChicago Personas - Final	appendix requested
AllianceChicago Implementation Plan	Reviewed.
AllianceChicago Training Materials	Implementation could not be fully realized during this 1-year period, so no training materials. In lieu of training materials, AC asked CHCs to look at prototypes and give feedback.

Executive Summary

The goals of this project are to facilitate increased [00:05:00] evidence-based care and shared decision-making around contraception for health center patients in accordance with the US Medical Eligibility Criteria and selected practice recommendations to advance the identification tracking and management of women, identified with gestational diabetes, and enhanced postpartum contraception counseling, and to test and evaluate the National Quality Forum or NQF clinical performance measures for the provision of quality contraceptive care.

In Year 2, NACHC developed a public health informatics structure for approved contraceptive use and diabetes screening and postpartum care for data aggregation and quality improvement, and focused in Year 3 on improving documentation practices and implementing tools to make use of better quality data collection structure with the potential to adapt and scale.

The two partners were three community health Center (CHC) members of AllianceChicago, covering metro Chicago area, and three CHC partners of OCHIN in urban centers in three regions of California: Los Angeles, San Diego and the Bay Area.

Activities

Outcomes

Lessons Learned

Recommendations

Scale and Spread

Goals

Project goals (CDC)

Partners took slightly different approaches to implementing Clinical Decision Support for Contraceptive Practice guidelines, identification of the course and care of gestational diabetes, and identification of patients likely to have a high risk post-partum period.

AllianceChicago Goals

AllianceChicago's goals for this project were to make changes to the maternal health content in the Electronic Health Records (EHR) that would increase tracking and analysis, with special attention to GDM, reproductive planning, and post partum care, to

identify complex/high risk patients during pregnancy with
longitudinal assessment of outcomes from those who have had a pregnancy
support real time identification of pregnant patients for reporting/case management purposes
increase postpartum care and reduce poor maternal health outcomes
Increase outreach that includes contraceptive counseling, use and follow-up

To do this	By doing this
Identify and follow high risk pregnancies Ensure continuation of care in postpartum period for medically complex patients Provide family planning and comprehensive reproductive services	Establish process/develop registry of high risk/medically complex patients during pregnancy Establish process for care coordination for these patients to ensure receipt of postpartum services as per below Postpartum visit(s) attendance with protocol for addressing the following: 2 hr gtt for patients with GDM Linkage to PCP, as needed Ensure patients have comprehensive contraceptive counseling (reproductive plans) during PNC/PPC and provision to their desired method

To do this

By doing this

Identify and follow high risk pregnancies
Ensure continuation of care in postpartum period for medically complex patients
Provide family planning and comprehensive reproductive services

Establish process/develop registry of high risk/medically complex patients during pregnancy
Establish process for care coordination for these patients to ensure receipt of postpartum services as per below

Postpartum visit(s) attendance with protocol for addressing the following:

2 hr gtt for patients with GDM
Linkage to PCP, as needed
Ensure patients have comprehensive contraceptive counseling (reproductive plans) during PNC/PPC and provision to their desired method

Their words:

to develop content to optimize our maternal health content in the Electronic Health REcords (EHR)

help providers document pregnancy episode/outcome
track and identify those patients
- reporting
- case management
- data analysis

OCHIN Goals

OCHIN's goals for this project were to

use data and informatics solutions to optimize care in the post-partum period
increase post-partum visits for those patients at higher risk for being lost to follow up after delivery
ensure comprehensive care by addressing contractive and depression screening during post partum, attention to GDM

Activities

Each partner recruited and engaged three Community Health Centers (CHCs). Alliance Chicago's CHCs covered the metro-Chicago region, while OCHIN's CHCs were located in three different geographical regions of California, Los Angeles, San Diego, and the Bay Area. Partners looked for CHCs with an interest in improving post partum care and had potential champions who could own the project at each site.

AllianceChicago

building a registry of patients identified by each health center to track and follow. This will include a review of data metrics, post-partum follow-up, and additional review of those with GDM. EHR system is Centricity/Athena.
each of the three sites chosen had an interest in improving; b. had begun on a registry and had already id'd gaps in workflows (readiness); c. this funding could improve existing efforts (leverage/synergy)
Engagement strategy: targeted email to CHCs with existing interest/work in this area, meetings.
Implementation Plan - this section is not the actual implementation plan, which is here but rather a comment on the usefulness of the implementation plan and notes
Partners internal: PI is practicing Ob/Gyn in one of the CHCs, Director Informatics, Data analysts, SMEs at CHCs gave real life application of use of EHR, reviewed areas of EHR for improvement.
External: City-wide initiative Chicago Collaborative for Maternal Health (CCMH). Internal: EHR

OCHIN

Created a Postpartum Express Lane, formal launch in late May 2020.
1. The express lane has been automatically made available for all visits with a chief complaint of postpartum. All orders are available within the SmartSet for contraceptive care, patient education available for the AVS, and a link out to the MEC/SPR for provider CDS is incorporated in the express lane. We also created a BPA to look for glucose results and prompt if they are not available.
Training on the postpartum express lane was recorded and distributed to the CHCs, training documentation. Slide deck was created to explain details of the build.
Created three different Smart Texts, one for a two week, one for six week, and one for a telephone visit Any providers using the Express Lane would have access to those notes that were structured to make sure that we collected all of the data that we need. .
Created a best practice advisory (BPA) was also created to prompt lab ordering for all postpartum patients that have gestational diabetes to have their 2-hour glucose screening done. It would look for, had that patient had a two hour postpartum glucose done in the postpartum period? Was there a result back? If the result was not available, it would trigger or prompt the clinician [00:41:30] who was caring for the patient to input an order for the glucose screen, so that it would not get missed.

"...we tried to standardize the project across the service areas, but each service area, what we found really can document their postpartum patients quite different. It's consistent within the service area, but not so consistent across the service areas. Trying to put together one larger project with the pieces of the note, the BPA and the Express Lane, certainly that became challenging just because of the differences in the way the different service area is documented." - Dr. Michele Whitt

Status

From Julia's Harvest Slide

Both partners identified a critical need for improved support for identifying and filling gaps in care
Starts with the pregnancy episode: supported in one partner system but not the other
Changes to the EHR alone can’t implement needed changes to care activities and documentation practices
Must engage health center care teams
Can create decision support to direct users to document follow up
Data quality varies by the type, timing, location of service

difficult to implement quality improvement within a year. There are a lot of moving parts Building in the EHR is a complex process that goes through various iterations and testing before deployment. Mapping of elements and going through our standard data request process could have benefitted from more time. Recruitment of health center participation, even on a quality improvement level, also take time

both partners identified a critical need [00:17:00] for improved support for filling gaps in care and identifying those. And again, I think that really goes back to the lack of support in most of Electronic Health Records for the area of women's health, but specifically the area of obstetrics and pregnancy. The pregnancy episode is something you're going to hear about from the partners. So I won't ruin that spoiler, but changes in the EHR alone, can't implement the needed changes to shift the workflow and teach the care teams about how we are [00:17:30] using a redesigned system, how we are going to maybe get some more systemization to the documentation and the activities of care? So we had to engage health center care teams, and that's what the work has been implemented on the ground with both of our partners. And keeping in mind the data quality varies by the kind of the data when the timing is the location of the service, and we're looking to fill in those gaps as part of quality improvement.

AllianceChicago is in the midst of implementing, finding that, to best meet the needs of the providers, the planning process with CHCs is iterative and therefore takes longer - seek to have the testing done May 2021 and June 2021 for deployment in July 2021.

EHR development
collaboration with CHCs
Alignment of priority areas

Alliance Chicago will be scaling and replicating this pregnancy episode build and change to the Outcomes Documentation Developing a plan to go live in the EHR for pilot sites summer of 2021. We're developing some training to accompany that, and then we're going to get some initial feasibility back.

But the goal would be that this will go network-wide for Alliance Chicago. So all health centers that are caring for [01:37:00] pregnant patients will be able to more consistently document pregnancy episodes and pregnancy outcomes so that we can then do better case management on the front end, and data extraction and analysis on the back end. And so we are very excited. Also, we have some technology coming down the pipe that will supplement the electronic health record to build registries at point of care for our [01:37:30] health center partners, that we hope that the pregnancy episode and outcomes will facilitate that to be even easier than we initially envisioned. So we're looking forward to seeing where this leads for our network.

OCHIN implemented its express lane and is the process of collecting data to see if the metrics are changing/if goals are being met in Fall 2021 and then offer to other CHC members, who could just turn it on. the Best Practice alerts are going into production for everyone in summer of 2021

Contraceptive counseling is often not coded. We know it occurs much more frequently and contraception is provided much more frequently than we see the code itself or the evidence of counseling. And multiple methods of capturing contraception can make it challenging, like to look at multiple data types, and then those transitions of care are unlikely to be [00:15:00] captured as well. And some patients, they'll let the clinic know I'm leaving your practice, I'm moving away. But that information usually doesn't get into the record. So we don't know to count that patient out. Instead, they look like they've been lost to follow-up.

FINDINGS

The biggest finding was that there was not a field to document pregnancy episode, crucial to initiating or tracking post partum care - or even what kind of care the women might need because the outcome of the pregnancy is not known.

data analysis is difficult, must understand multiple confounding factors

We need Input from all levels of users/stakeholders critical, including patients
Comprehensive data about pregnancy takes a LOT of work!

we are still very far away for achieving semantic interoperability in women's health data elements
Redesign of the way we capture data. Input from patients. Pregnancy and PP documentation is not easy and we still have alot of work to do
Need continued focus on PP care to refine IT approach to improve outcomes & health equity.
Good data helps highlight disparities and needs.

consistent documentatino of episodes, consistent documentation of outcomes. pretty much consistency and interoperability consistent documentatino of episodes, consistent documentation of outcomes. pretty much consistency and interoperability
the opportunity on a national scale, in terms of consistent documentation for maternal and repro health that will faciliate better clinical care and research/analysis

we need an HIE like data collection at the outpatient level for community registry

AC These tasks take time: recruitment, discovery, informatics design, feedback from SMEs, working on processes with health centers is an iterative process, and that takes time

Discovery: patients could have been followed up if there had been a field for pregnancy outcome in the EHR

Discovery: Ability to track in EHR would help AC/CHCs align with clinical best practices
HCD/Results/Discovery c. For the Persona, this improvement would have given more context to the symptoms the patient was experiencing (and thus, better care)
Having the pregnancy episode would allow AC/CHCs to build cohorts of live births for a certain period and study factors in cases of miscarriage or loss, and mortality.
Environment - COVID, Documentation is uneven - some use EHR, some have a manual process - result: both lead to individuals lost to followup (why would use of EHR result in loss?)
Project works well because both AC and CHCs are have a high interest in improving this area

Results wins

Has data flow changed? No - made us more aware for the need of support to have a more standardized way of data capture
Results/Finding: Review of the value set authority codes (VSAC) uncovered the need for our proprietary codes to be mapped to provide the data requested.

Process wins:

Greater understanding for AC of limitations of EHR - what can and cannot be done w/in system
The project lent weight/validity to previous, lingering inquiries that had not been resolved: OB updates to the EHR system were needed to effectively support the providers and case managers meeting with patients who experienced pregnancy, regardless of the outcome.
Process win, halo: project's existence brought AC team closer to CHC staff to improve workflows and outcomes
Hosting focus groups on workflow challenges at the CHCs, get their wishes to improve documentation and tracking. With the feedback from these focus groups, work with our internal teams at AllianceChicago to create content specific to the sites.
Given the initial persona, the changes this year would have meant knowing the end of the pregnancy story for a patient with GDM who decided on control via diet control rather than insulin. The provider was able to document in the chart the end of the patient’s pregnancy and outcomes related to the delivery.

Outcome wins

Having the pregnancy episode in EHR will help providers track individual pregnancies and create registry of all patients - to create plans to meet their healthcare needs.
Outcome win: The episode will make back-end data extraction easier. Currently, many variables are used in determining a pregnancy event, which can lead to error or not capturing all pregnant patients.
Outcome Win: A template that standardizes our codes to VSACs to use for data mapping for other projects as well as this one. (Halo effect)
Win: Can better create a registry of postpartum and gestational diabetes patients based on the implementation of the pregnancy episode content
WIN: building and implementing a postpartum express lane across all three CHC partners.

Unexpected benefits

the topic became a focus of discussion; 2. provided dedicated time to develop a tool and design feedback to informatics; 3. More contact with CHC SMEs

OCHIN

Process win: The workflow has markedly changed. Prior to express lane creation there was no centralized place that brought together postpartum notes, procedures, or medication orders. A new workflow was created for all service areas. The notes were individualized for the group of providers expected to use the express lane.
Outcome win in progress: exploring ways to correlate express lane use with increased glucose, contraceptive, and postpartum depression screening.
Outcome win: We identified that not everyone is receiving their glucose screen. Developed a best practice alert to alleviate this issue.
The biggest win is that because everyone had [00:44:30] bought into the project, that we were able to get this Express Lane built, we were able to standardize our notes for the clinicians to use.
- We were able to standardize on orders, even though they were different in each service area, in order set that provides us to be able to collect data on the backside about how well the Express Lane is used, and making sure that we have a standardized data set so that we can report back to the service areas that yes, you're doing a good job at making sure that your postpartum patients get contraceptives, [00:45:00] or yes, you're doing a better job now of making sure that everyone gets their gestational glucose screen done.
- It's a one-stop shop that has both your document, your orders, your referrals, everything is just on one screen and that one panel and when it is but well, you don't have to add additional orders or anything, and they're customizable to your preference list.
being able to do some data analysis on the backside to prove that patient care, certainly from the standpoint of meeting [Aycock standards 00:46:39] and what you want from the standpoint of gestation diabetes, the EPDs and contraceptive care is being met and being able to be able to provide feedback to the service areas is a big win for patients that get more consistent guideline based care.

"We knew that Express Lane existed, but had ignored it when it popped up for us because it seemed like a distraction, but being able to create it for something that we knew there was such a great need and really work on it from the ground up and partner. Also, knowing that there's this larger force that's not just me, the provider or my clinic, but really knowing this is coming from such a national level is important really helped us go along and really get into the nitty gritty, which I ensure probably drove Michele and AnnMarie crazy at times, but I think it's important." - Kim Cardoso, Nurse Midwife

Express Lane

Lessons Learned and Best Practices by AllianceChicago and OCHIN

Issue	Best Practices	Source (not necessary in final report)
Project overran its hours for AC Informatics	Use this year's hours for Informatics Staff to estimate future labor hours to get adequate coverage.	AllianceChicago Maybe don't include this?
Learned Collaboration does not come without engagement and in retrospect - takes time	Engage health centers early in the project, and expect it to take a longer than one year.	AllianceChicago
Staff transition - champion moved to another role, meant meeting with new-to-the-project clinicians to introduce and make changes to CHC workflow as needed.	Be prepared to orient new people, recruit two champions at each CHC at the outset.	OCHIN
(what was the issue difficulty in getting started?? reluctance to try something new?)	Planning teams need to build in ways to increase ownership by service area staff.	OCHIN
An Excel spreadsheet was not enough to help clinic staff understand the tool in the planning phase.	A detailed sample build helps CHC staff envision the future project. Make sure notes are are exactly how staffers want them. We would come up with a planning guide rather than the spreadsheet we used.	OCHIN
Scoping this project was complicated. The HC’s did not have a focus at the start and did not understand the system enough to know what was possible or relevant.	Through discussion, identify key members of the clinical team that know and use the EHR daily and centered them in build and workflow discussions moving forward.	OCHIN
Patients are always moving in and out of post partum, and since it was a relative date, identifying the postpartum period was difficult. There is always a patient moving in and out of postpartum. It is always based on a relative date, identifying their postpartum period was difficult to interpret. When we looked at the data, we had to be careful about how we interpreted the window of when the data was valid. For example, glucose screening timeframe is not fully reflected in any 30-day given window.	We found it was better to assess the data for the postpartum period in a 90-day window rather than 30 days.	OCHIN
Partners needed training, some more than others. The lack of feedback on specific training needs until training time was an obstacle for creating implementation documentation tailored to the HC needs.	Service area specific training documentation and demonstration.	OCHIN
Not enough people had access to REL to test the build and give feedback. There was only one tester at each CHC.	Arrange to have more than one tester at each CHC implementation site.	OCHIN
Do not assume all clinics have the same workflow. Each service area has their own workflow and as a result, how they wanted the express lane needed more individualization.	We learned their workflow and as we became more familiar, we were able to provide better insights on individualizations to help them realize their build. Creating workflow documentation that exemplifies best practices for each HC.	OCHIN
Too many choices for members made builds complicated and too customized for each health center.	We would provide a more detailed sample build next time.	OCHIN
Trying to standardize workflows in order to create the Express Lane and Best Practice Alerts was made more difficult by the fact that the service areas documented their postpartum patients differently - and wanted to customize the processes to match their workflows.	Discuss proposed changes, have Champion explain the processes. It is not helpful to present a model as a given, but to propose it and work iteratively to negotiate which elements can be standardized - less custom. The result was a set of standardized orders.	OCHIN

Wish List

EHR integration
Involve the patient more in the process, for example, a say in the example of the woman with the gestational diabetes, if she's tracking her sugar, is there a way that we can upload that in the EHR so that the provider has that information prior to the visit to have that dialogue? Is there a way they could end the episode? Again, there's a whole host of things, I think, that at the patient level, we can have them more involved in the process.
more resources

Data from Year 3

We found that about 6% of pregnancies... Well, first of all, we found that it was 177,000 pregnancies in our sample of about 1, 300,000 people. 6% of those pregnancies also had a diagnosis of diabetes mellitus with it. We also found that 9% of the pregnancies had a postpartum [00:55:30] visit, and we had found that about 6,700 patients also had a diagnosis of gestational diabetes. When we looked at the people that had the gestational diabetes diagnosis, we found that 71% had some diabetic screening, and we found that 38% of these people had gestational diabetes and [00:56:00] diabetes diagnose. You can see here the distribution of contraception in people that were pregnant.
The 6% of patients that have diabetes and were pregnant, at least 46% of them had a diabetes diagnosis prior to their delivery date and about 20% had a diabetes diagnosed after that delivery date, [00:58:30] which suggests that they became diabetes after their pregnancy.
We did find that the 3.8% of pregnancies have gestational diabetes, so about 1200 people, but we also found that 5,500 [00:59:00] patients had a gestational diabetes diagnosis, but in our assessment of who was pregnant, who was not, they were deemed non-pregnant. You can see some of the noise in the data in here and some of the questions, and that's why our analysis was thrown into a wrench because we couldn't match our data, and then we figure out, "Oh, when you ignore the fact [00:59:30] that they're pregnant, then you get a lot more people with gestational diabetes." We were not able to do the analysis of the screening if was temporary order. For example, if the screening was done to diagnose the patient, or if the screening was done post pregnancy, so this is something we still want to look at it.
Patients have gestational diabetes and diabetes. We couldn't figure out yet the temporal order of those. We're continuing to take a look, but you can see the number of people that had diabetes after their delivery date,those number of similar; one thing that we're missing here is the counseling, but that is something that we need to discuss because we need to understand also the temporal order of that and how it applies to this
- lack of INteroperability may be the cause of the lack of infomratoin on prescritpions that might have been given in the hospital such as birth control pills or LARC
Patients that were lost to follow up could be a for any one of a number of reason -
- Came in for a pregnancy test anbd were not pregnant
- the prgnancy did not continue
- transferred care
- If you're undocumented, you can retain your Medicaid coverage for a certain number of days after the pregnancy, so those people are unlikely to have a regular source of care besides the health center and they might be trying to squeeze in their postpartum visit because that's the last time they'll maybe have access to their insurance.

-Discussion Kate asks: Pedro, you talked a lot about temporal sequence and it seemed like a lot of that was not very clear both for the gestational diabetes data and the contraception data, and I was wondering if you could just talk about that a little bit more. So for example, in the contraception data, do we know that that contraception data is after the pregnancy episode, or is that any contraception [01:10:30] during the year, where there was also a pregnancy or could you just talk through sort of what temporal things we're sure of and what we're not?
Pedro Carneiro: Yeah, definitely. So I would say half of the pregnancy sample has an estimated delivery date that is valid within the timeframe, right? So those people we can actually pinpoint [01:11:00] and use it as things have a temporal order. Some of the patients were identified through diagnose codes. So then we have to assume that the first time that they got the diagnose is when they got pregnant. So that's a little bit of a stretch right there, right. For the diagnosis, sometimes they have dates that [01:11:30] are continuous. Sometimes they only have one date or they have a date that we are not sure if that's the first time that the person got diagnosed or if that is just the date where the diagnosis being taken care of. So some of those questions still remains, then sometimes the data will be categorical and will happen within an encounter.
So [01:12:00] the question becomes, is that encounter when that data happened, or does that just indicate that the data is checked. Ongoing. So we still have some of those questions to answer, but in terms of, I think for the LARCs, what we did was, this person is pregnant, they've had multiple visits, [01:12:30] what was the contraception of choice in each of the visits, right? Most of the times the same, and we just reported the last one. The most recent one, which in theory would be after they're... in the visit after their delivery. But there's still some questions there as well, right. So [01:13:00] we still have to do some [QI 01:13:03] with the partners to just understand a little bit more of the temporal order of the diagnosis and the medications and things like that. [crosstalk 01:13:16].
- Julia - At some point they seem to be pregnant and pregnancy was definitely completed and follow-up occurred.
- John we did get specific delivery dates for a good chunk of the data. And others were just based on other estimates.
- Julia - moving towards the next steps of the project

Comparison to Baseline (Year 2?)

Scale and Spread/Recommendations

Bring additional partners passionate about maternal and infant outcomes from health centers
Design workflow to support coordinated perinatal and post-partum care
Implement via Agile continuous improvement in health centers and iterate
Build a model and toolkit to support best practices and integrated maternal care approach
Importance of clinical champions within health centers who create value and demand for EHR vendors to adhere to standards that support Women's Health data elements
consistent documentatino of episodes, consistent documentation of outcomes. pretty much consistency and interoperability
EHR Express Lane for Post-Partum visit
Building a pregnancy registry
Everyone should have access to pregnancy episodes to allow management of follow up and prenatal care
Collaboration across health centers for pregnancy management
Learning Community on obstetrics informatics

Conducted a panel at the American Medical Informatics Association (AMIA) CIC May 19, 2021 S06: Inequity in Health Information Technology: Women's Health Data Elements, Interoperability, USCDI v2 and Terminology Challenges

Pull quote

We are really excited to be working with our partners who do bring great enthusiasm. I echo everything that Roxanne said that when there is a project like this, that really touches upon something that [00:32:00] addresses a gap that's not academic per se, but very practical and important, and will make a big difference. The response that we get is very, very different and so this one had been returned with great enthusiasm. We're just thrilled to be able to offer this and be able to bring it to the forefront and look forward to seeing the mark that it makes on our network, moving forward, both in terms of a workflow for case management purposes and then for data analysis, and learning more about the populations we serve, which is really why we come to work every day. - Dr. Lisa Masinter, AllianceChicago Harvest Meeting

Potential future directions

Assess whether the tools developed by the two partners will result in better reimbursement due to more comprehensive documentation leading to more billing.

John that anyone implementing any software could create in their software, basically a smart on fire application, in which the patient would be able to provide credentials and say to that system, I give you authorization to go pull my data from - I consent for you to pull such and such data, or only this data, s this patient rule was designed to enable a system to pull data from either a third-party claim system like a healthcare provider
Julia - the promoting interoperability rule requiring electronic health records to support a fire release, I think it's release for, API actually go into effect on July 1st. It's obvious that many of the HRs aren't necessarily going to meet that benchmark. I know that [OCHIN's 01:16:51] EHR does and will be able to do that, but the beauty of that API is that basically you can [01:17:00] request and receive data using fire. And as John said, some of the patient access changes to the rule have made it clear that you no longer need the direct consent of the patient to ask for information about the patient directly related to their care.
Look at the impact of care on longer term outcomes
Look at the impact on the care team and utilization/ effectiveness of care
Work with ACOG on pregnancy registry concept, content and approach
Involve the patient more directly
Create and assess the impact of a hybrid (virtual and in person) care approach
Integrate virtual monitoring for high risk patients?

Which proportion of these patients are truly lost and really potentially at risk for bad outcomes, because they're not getting any care for some reason, versus which one of them have just gone elsewhere and they won't need us. And to your point, Michelle, if someone else has taken care of them, it's going to be fine. I think it's hard for us to know that, right? And I think one of the big opportunities here is if we can figure [01:22:00] out how to really coordinate care around the pregnancy episode and how to appropriately attribute patients and transfer information about them, transfer their care. That opens that value-based Pandora's box that everyone searching for, how do we do this appropriately?
So it potentially could create a method for us to work with payers, to get better bundle payments or incentives to demonstrate. We know that all of our patients, what's happening to them. And we can [01:22:30] demonstrate follow-up at X rate. And I think we see here a follow-up rate which is very consistent with what we see for Medicaid patients nationally, but it's not even 50%. So what we want to get to is the place where if we haven't done the follow-up, we know who did it, and or why not. Maybe the patient left the area and missed the window, but [01:23:00] at least to Lisa's point, we can rest our heads and know that everyone's being managed to the best of our ability.

it's been shown that adverse pregnancy outcomes, including gestational diabetes are associated with a lot of longterm risks, like cardiovascular disease.

So I'm currently working with now my interest and what you guys are finding is to try to see whether there could be lessons learned for cardiovascular disease, risk in women. If there's some way that all of these outcomes can be appropriately documented in an electronic [01:24:30] health record that providers even five, ten years down the road, or even if there's just a single interaction of a woman back into the healthcare system, that person could be flagged, for potentially needing early screening or intervention for cardiovascular disease, which, as we know, is the number one cause of mortality of women in the US. So that's just my plug for making sure that I think it matters that [01:25:00] pregnancy outcomes are appropriately documented.

CMS also created in their inpatient, their IPPs rule this year, a requirement for inpatient facilities to provide a report within, I think, 72 hours of an inpatient stay admission discharge. [01:31:30] So one would hope that the possibility would, between the FHIR API requirement and that requirement to send a message, which probably at this point will still be an ADT message, right? Just the standard brief HL7 message. But hopefully we could see that start to trigger that information to come to the system and to encourage us at least to use [01:32:00] our analog approach of calling and asking for the record. Even if the electronic message doesn't contain very comprehensive data.

And that's something I think very much would be an interesting item on my list for a future project.

(Michele 1:39) that we linked the closure of an episode with a pregnancy outcome. So that if you've opened an episode and then, quite frankly this is just workflows [01:39:00] within your office. I mean, it should be that if a patient hasn't, they've missed three appointments or whatever it is, whatever you want to do as the criteria, that pregnancy episode needs to be closed, or you're going to call or try and get that patient in. But once all of that process is done and when the pregnancy episode is closed, that you link that to... They either delivered and put a date, or they had a molar pregnancy or an SAB or whatever, put a date. Or that they were lost at [01:39:30] follow-up, you don't know what happened to them.

Then there's another one... Oh, or they transfer care and you could put where they went. So if you have transfer care and quite frankly, they do that in Allscripts. Allscripts has a way to do that. But if you put that they transfer care, they were lost at follow up, you've done the best you can, you don't know what happened to them, or that it's got one of these outcomes with the date. I think that that would go a long way to, at least, finishing out or closing, so [01:40:00] that there's not something hanging about that patient. And you could then run reports. How many patients are you losing to follow up? Are just dropping off the map and you really don't know what happened to them. And I think that's a different question because you could very easily figure out what's going on with that, or at least make some effort to figure out why you're losing so many patients and the rest of it, they had an outcome and you know.

Kim Cardoso - different health centers, we are all different businesses, right?

Essentially competing for the same patient population. However, what we've decided to do over the past couple of years is we formed a group where we are standardizing care throughout our area, our region, since our patients are so mobile, they might start their pregnancy in one health center, be in the middle in another one, have a gap in care, follow-up post-partum with someone else. [01:42:00] So we've been meeting regularly to talk about how we want to take care of our patients as a whole, and also how we want to share data. So we've been trying to, so if one of us builds a registry in Tablo, a dashboard, then we share the know how with the other one and... Yeah, with this concept of really much larger sharing to care for the patient who moves through all of our systems, keeping the patient at the center of the care.

So whatever support you can get for that [01:42:30] from a data standpoint or a health information standpoint, and forgive me for not having the right language as a clinician, would be really amazing, because the follow up and the continuity is really important to us, whether they're our patient or not, as long as they were our patient at some point.

Support for building pregnancy registries at the CHC level, but owned by the health care provider. One type of Medi-Cal has a pregnancy registry that covers a diverse area covered by three public health departments - but some patients done have Medi-Cal insurance so they are not in that registry. an HIE is hopsital-based infromatoin.

what should happen is as soon as we know someone's pregnant, we should start populating a longitudinal set of data [01:48:00] that tell us about the pregnancy so that we can say, "Is the person progressing the right way? Are they hearing off in a high risk direction? Do they need additional touches and additional treatment? Are they showing up for their appointments? And if not, how can we bring them back or identify the fact that they've left the practice? And are they not only getting information transitioned to and from their [01:48:30] inpatient facility, but are they getting the postpartum visit? And then are we transferring their care back to the non obstetric providers who used to see them?" See if morbidity, mortality and other bad outcomes are reduced

IGNORE THESE NOTES

Andrea Price Old Account (Deactivated) Notes on AllianceChicago final report

CHCs

3 CHCs in the metro Chicago area

I. A1. Goal: identification of patients for follow-up

A2. These tasks take time: recruitment, discovery, informatics design, feedback from SMEs, working on processes with health centers is an iterative process, and that takes time

B1. Environment - COVID, Documentation is uneven - some use EHR, some have a manual process - result: both lead to individuals lost to followup (why would use of EHR result in loss?)

C. Activities

3 sites

a. each had an interest in improving; b. had begun on a registry and had already id'd gaps in workflows (readiness); c. this funding could improve existing efforts (leverage/synergy)

building a registry of patients identified by each health center to track and follow. This will include a review of data metrics, post-partum follow-up, and additional review of those with GDM

C. Engagement strategy: targeted email to CHCs with existing interest/work in this area, meetings.
II. Implementation Plan - this section is not the actual implementation plan, which is here but rather a comment on the usefulness of the implementation plan and notes
C. Partners internal: PI is practicing Ob/Gyn in one of the CHCs, Director Informatics, Data analysts, SMEs at CHCs gave real life application of use of EHR, reviewed areas of EHR for improvement.
D. External: City-wide initiative Chicago Collaborative for Maternal Health (CCMH). Internal: EHR
F. Timing - We seek to have the testing done May 2021 and June 2021 for deployment in July 2021.

d. RESULTS - with the improvement:

a. patients could have been followed up if there were a field for pregnancy outcome in the EHR
b. Ability to track in EHR would help AC/CHCs align with clinical best practices
c. For the Persona, this improvement would have given more context to the symptoms the patient was experiencing (and thus, better care)
d. Having the pregnancy episode would allow AC/CHCs to build cohorts of live births for a certain period and study factors in cases of miscarriage or loss, and mortality.
C. Has data flow changed? No - made us more aware for the need of support to have a more standardized way of data capture
Results/Finding: Review of the value set authority codes (VSAC) uncovered the need for our proprietary codes to be mapped to provide the data requested.
E. a. The project lent weight/validity to previous, lingering inquiries that had not been resolved: OB updates to the EHR system were needed to effectively support the providers and case managers meeting with patients who experienced pregnancy, regardless of the outcome.
b. Outcome win: Having the pregnancy episode in EHR will help providers track individual pregnancies and create registry of all patients - to create plans to meet their healthcare needs.
c. Outcome win: The episode will make back-end data extraction easier. Currently, many variables are used in determining a pregnancy event, which can lead to error or not capturing all pregnant patients.
d. Process win, halo: project's existence brought AC team closer to CHC staff to improve workflows and outcomes

IV Lessons Learned

a.. Challenge - Project overran its hours for AC Informatics - overcome by everyone pitching in.
Learned Collaboration does not come without engagement and in retrospect - takes time
c. Learned Project works well because both AC and CHCs are have a high interest in improving this area
d. Process win: Greater understanding for AC of limitations of EHR - what can and cannot be done w/in system
e. Learned Engage health centers early in the project
f. Learned: Hosting focus groups on workflow challenges at the CHCs, get their wishes to improve documentation and tracking. With the feedback from these focus groups, work with our internal teams at AllianceChicago to create content specific to the sites.
V. A.
Outcome Win: A template that standardizes our codes to VSACs to use for data mapping for other projects as well as this one. (Halo effect)
Win: Can better create a registry of postpartum and gestational diabetes patients based on the implementation of the pregnancy episode content.
difficult to implement quality improvement within a year. There are a lot of moving parts Building in the EHR is a complex process that goes through various iterations and testing before deployment. Mapping of elements and going through our standard data request process could have benefitted from more time. Recruitment of health center participation, even on a quality improvement level, also take time

B2. Unexpected benefits - 1. the topic became a focus of discussion; 2. provided dedicated time to develop a tool and design feedback to informatics; 3. More contact with CHC SMEs

Andrea Price Old Account (Deactivated) Notes on OCHIN Final Report

Goal
1. Use data and informatics solutions to optimize care in the post-partum period.
2. increase post-partum visits for those patients at higher risk for being lost to follow up after delivery
3. ensure comprehensive care by addressing contractive and depression screening during post partum, attention to GDM
Status/Outcome: We are currently in the implementation phase of the project and have not collected data regarding outcome.
CHCs

3 CHCs in located in different geographical regions of California
Announcemets made during OCHIN workgroup meetings, then email follow up and then meetings.
These CHC’s were selected because they demonstrated a strong interest in the work and had provider champions advocating for the project within their health centers

Major activity:

Created a best practice advisory (BPA) was also created to prompt lab ordering for all postpartum patients that have gestational diabetes to have their 2-hour glucose screening done.
created Postpartum Express Lane, formal launch in late May 2020.
1. The express lane has been automatically made available for all visits with a chief complaint of postpartum. All orders are available within the SmartSet for contraceptive care, patient education available for the AVS, and a link out to the MEC/SPR for provider CDS is incorporated in the express lane. We also created a BPA to look for glucose results and prompt if they are not available.
Training on the postpartum express lane was recorded and distributed to the CHCs, training documentation. Slide deck was created to explain details of the build.

Results

Process win: The workflow has markedly changed. Prior to express lane creation there was no centralized place that brought together postpartum notes, procedures, or medication orders. A new workflow was created for all service areas. The notes were individualized for the group of providers expected to use the express lane.
Outcome win in progress: exploring ways to correlate express lane use with increased glucose, contraceptive, and postpartum depression screening.

Lessons Learned

Staff transition - champion moved to another role, meant meeting with new-to-the-project clinicians to introduce and make changes to CHC workflow as needed.
Importance of consistent champion - mitigation plan: be prepared to orient new people, recruit 2 champions at the outset.
Planning teams needs to build in ways to increase ownership by service area staff.
A detailed sample build would help CHC staff envision the future project - an Excel spreadsheet was not enough to help them understand in the planning phase. Make sure notes are are exactly how staffers want them. We would come up with a planning guide rather than the spreadsheet we used.
Challenge Clinics all have different workflows - Lesson Learned - Look for a balance between customizing for each CHC and creating a standard so it can be generalized to other health centers. Customizing was labor intensive: Each service area has their own workflow and as a result, how they wanted the express lane needed more individualization. Best to try to get all to agree on creating workflow documentation that exemplifies best practices for each HC.
Difficulty in describing something that is complicated but keeping things simple enough to be understood, implemented, and generalizable to other CHCs.
Patients are always moving in and out of post partum, and since it was a relative date, identifying the postpartum period was difficult - advice from this project is best to keep it wider. Example: glucose screening timeframe is not fully reflected in any 30-day given window. We found it was better to assess the data in a 90-day window.
Pre-assessment of training needs and iterative training options could be explored: The lack of feedback on specific training needs until training time was an obstacle for creating implementation documentation tailored to the HC needs.
Give more people ability to test - there was only one tester at each CHC

Project will be made available to other CHCs after the original CHCs evaluate and recommend changes.

Consolidated

Lessons Learned

Date	Challenge or Best Practice	Category	Description (Use your own words)	What went well and why?	What is blocking you or did not go well and why?	How did you address or would you correct this for next time?	Is there any follow up needed?	Solution Team
MM/DD/YY	Challenge or Best Practice	Data, Implementation, Roll out, Workflow, Experience	Issue	+	-	?	Next Steps	Who can speak about this issue

OCHIN

Date	Challenge or Best Practice	Category	Description (Use your own words)	What went well and why?	What is blocking you or did not go well and why?	How did you address or would you correct this for next time?	Is there any follow up needed?	Solution Team
2021-02-15	Challenge	Implementation	Scoping this project was complicated.	Through discussion we were able to identify key members of the clinical team that would provide input on the project.	The HC’s did not have a focus at the start and did not understand the system enough to know what was possible or relevant.	We addressed this by identifying key team members that know and use the EHR daily and centered them in build and workflow discussions moving forward.	No follow-up necessary.	OB Care Coordinator and/or Clinicians
2021-03-15	Challenge	Experience	Same HC champion throughout project vs. changes halfway through	Having a consistent HC champion made planning and implementation move smoothly.	Two of the HC’s had champions change halfway through then right at the end of the planning period.	We addressed this by being flexible and accommodating new clinicians into the project. For the future, we will include more than one champion from the start to avoid any team member drop off.	No follow-up necessary.	OCHIN and HC partners
2021-04-15	Challenge	Implementation	Do not provide too many choices for members.	Robust discussions about possible interventions.	Made builds complicated and too customized for each health center.	We would provide a more detailed sample build next time.	No follow-up necessary.	Clinical Informaticists
2021-04-20	Challenge	Data	We identified that not everyone is receiving their glucose screen.	We were able to address this with the project build.	Nothing blocked us once this issue was identified.	Developed a best practice alert to alleviate this issue.	No follow-up necessary.	Clinical Informaticists
2021-04-25	Challenge/Best Practice	Data	Floating timeframe of data.	We learned a wider window gives you a better understanding of your data.	There is always a patient moving in and out of postpartum. It is always based on a relative date, identifying their postpartum period was difficult to interpret.	When we looked at the data, we had to be careful about how we interpreted the window of when the data was valid. For example, glucose screening timeframe is not fully reflected in any 30-day given window. We found it was better to assess the data in a 90-day window.	No follow-up necessary.	Clinical Informaticists
2021-04-27	Challenge	Roll out	Partners needed training, some more than others.	During training sessions, clinicians openly asked questions and addressed any concerns that came up related to the postpartum express lane.	The lack of feedback on specific training needs until training time was an obstacle for creating implementation documentation tailored to the HC needs.	Service area specific training documentation and demonstration.	No follow-up necessary.	Clinical Informaticist
2021-04-30	Best Practice	Workflow	Do not assume all clinics have the same workflow.	We learned their workflow and as we became more familiar, we were able to provide better insights on individualizations to help them realize their build.	Each service area has their own workflow and as a result, how they wanted the express lane needed more individualization.	Creating workflow documentation that exemplifies best practices for each HC.	No follow up necessary.	Workflow Engineer
2021-05-25	Best Practice	Implementation	Create Best Practice Alert (BPA) based upon discussions about HC’s perceived weakness for providing adequate patient follow-up.	We were able to build from the existing development of the OB Dashboard.	We had to enlist the help of the lab interface analyst to standard naming conventions for labs in question.	This was just part of the process and not something we would or could correct for next time.	No follow-up necessary	Lab Interface Analyst
2021-05-25	Challenge	Roll out	More people need to have access to the testing environment.	There was at least one tester at each HC.	Not enough people had access to REL to test the build and give feedback.	In the future we would identify more testers.	No follow-up necessary.	Site Specialist

Browser not supported