Evaluating the Rollout of Estonia's Enhanced Care Management System

Last registered on December 20, 2021

Pre-Trial

Trial Information

General Information

Title
Evaluating the Rollout of Estonia's Enhanced Care Management System
RCT ID
AEARCTR-0003661
Initial registration date
May 01, 2019

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
May 06, 2019, 12:05 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
December 20, 2021, 11:05 AM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Region

Primary Investigator

Affiliation

Other Primary Investigator(s)

PI Affiliation
Harvard University
PI Affiliation
World Bank
PI Affiliation
World Bank
PI Affiliation
World Bank

Additional Trial Information

Status
On going
Start date
2018-12-03
End date
2023-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
The Enhanced Care Management intervention consists of training and coaching doctors to develop care plans and undertake targeted outreach to the subset of patients who are at high risk of experiencing deteriorating health so to increase their utilization of health services. The goal is to extend the use of preventive care, which improves patient health and reduces the need for curative medical services. ECM includes heightened tracking of tests and referrals, follow up after hospital discharges, tracking medication adherence, monitoring between clinic visits, and a focus on achieving clinical quality. It includes four elements: identification of high risk patients through risk stratification, development of care management plans by the primary care physician, proactively linking care providers together, and developing a team approach with patients and their caregivers. ECM is reflective of global primary care reforms that aim to focus the attention of doctors on high-risk groups and improve continuity of care (see Peikes et al, Health Affairs, 2018).

In a pilot of ECM conducted in 2017 with 10 doctors, take up of the intervention was high – pilot providers made 40% more calls to patients with cardiovascular disease (CVD) or one of the “triad” of hypertension, diabetes, and elevated blood lipids; were 11% more likely to have patients on appropriate statin prescriptions; had patients 25% less likely to be hospitalized for CVD-related conditions; and were 11% more likely to follow up within 30 days in the event of an acute CVD incident. However, this pilot was conducted with a purposively selected group of 10 doctors who were expected to be highly motivated early adopters. Even this motivated group expressed that ECM, while generally seen as clinically worthwhile, brought with it additional work in the form of additional time spent on electronic medical records.

Our Estonian counterparts in EHIF and Ministry of Social Affairs (MoSA) expect that take up will be significantly lower when the program is expanded to the general population of primary care physicians. One hypothesis to explain low take-up is that many Estonian doctors are confident in the efficacy of their pre-existing practices. Doctors may be reluctant to adopt ECM because they do not realize that ECM patients, although comprising roughly 6.5% of the patient register, represent a substantially larger share of the burden of disease among their patients – for example, they make up 11.3% of acute admissions. By shifting efforts towards these patients (via the ECM system), they are more closely aligning their efforts to the burden of disease that they face. We will implement 3 treatment arms designed to maximize uptake of ECM by primary care doctors through communicating distinct features of the above ECM rationale. Each aims to shift the appreciation of ECM by doctors along a specific dimension of their work practice.

In the first arm, providers will receive a “big push”-style mentorship introduction to the ECM program over the course of one week from one of the champion providers who participated in the pilot. In the second, providers will have an ongoing, weekly call with the same type of ECM coach which will have a similar overall budget for the implementer per provider. The third arm will receive both treatments. All groups, as well a comparison group, will receive weekly reports on their own performance relative to ECM expectations for patient management.

Registration Citation

Citation
Croke, Kevin et al. 2021. "Evaluating the Rollout of Estonia's Enhanced Care Management System." AEA RCT Registry. December 20. https://doi.org/10.1257/rct.3661-1.1
Former Citation
Croke, Kevin et al. 2021. "Evaluating the Rollout of Estonia's Enhanced Care Management System." AEA RCT Registry. December 20. https://www.socialscienceregistry.org/trials/3661/history/107030
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
The Enhanced Care Management intervention consists of training and coaching doctors to develop care plans and undertake targeted outreach to the subset of patients who are at high risk of experiencing deteriorating health so to increase their utilization of health services. The goal is to extend the use of preventive care, which improves patient health and reduces the need for curative medical services. ECM includes heightened tracking of tests and referrals, follow up after hospital discharges, tracking medication adherence, monitoring between clinic visits, and a focus on achieving clinical quality. It includes four elements: identification of high risk patients through risk stratification, development of care management plans by the primary care physician, proactively linking care providers together, and developing a team approach with patients and their caregivers. ECM is reflective of global primary care reforms that aim to focus the attention of doctors on high-risk groups and improve continuity of care (see Peikes et al, Health Affairs, 2018).

In a pilot of ECM conducted in 2017 with 10 doctors, take up of the intervention was high – pilot providers made 40% more calls to patients with cardiovascular disease (CVD) or one of the “triad” of hypertension, diabetes, and elevated blood lipids; were 11% more likely to have patients on appropriate statin prescriptions; had patients 25% less likely to be hospitalized for CVD-related conditions; and were 11% more likely to follow up within 30 days in the event of an acute CVD incident. However, this pilot was conducted with a purposively selected group of 10 doctors who were expected to be highly motivated early adopters. Even this motivated group expressed that ECM, while generally seen as clinically worthwhile, brought with it additional work in the form of additional time spent on electronic medical records.

Our Estonian counterparts in EHIF and Ministry of Social Affairs (MoSA) expect that take up will be significantly lower when the program is expanded to the general population of primary care physicians. One hypothesis to explain low take-up is that many Estonian doctors are confident in the efficacy of their pre-existing practices. Doctors may be reluctant to adopt ECM because they do not realize that ECM patients, although comprising roughly 6.5% of the patient register, represent a substantially larger share of the burden of disease among their patients – for example, they make up 11.3% of acute admissions. By shifting efforts towards these patients (via the ECM system), they are more closely aligning their efforts to the burden of disease that they face. We will implement 3 treatment arms designed to maximize uptake of ECM by primary care doctors through communicating distinct features of the above ECM rationale. Each aims to shift the appreciation of ECM by doctors along a specific dimension of their work practice.

In the first arm, providers will receive a “big push”-style mentorship introduction to the ECM program over the course of one week from one of the champion providers who participated in the pilot. In the second, providers will have an ongoing, weekly call with the same type of ECM coach which will have a similar overall budget for the implementer per provider. The third arm will receive both treatments. All groups, as well a comparison group, will receive weekly reports on their own performance relative to ECM expectations for patient management.


Intervention Start Date
2021-06-01
Intervention End Date
2023-06-01

Primary Outcomes

Primary Outcomes (end points)
Outcome #1 – Provider participation (constructed)

Outcome #2 Take-up: Number of CVD or metabolic triad patient follow-up calls
Unit of observation Patient
Data source EHIF medical records
Sampling method Observed universe of records
Number of rounds of measurement N/A, ongoing
Dates of measurement N/A, ongoing
Data access confirmed? Expected date of confirmation Confirmed

Outcome #3 Patient likelihood of statin prescriptions, conditional on CVD or lipid condition
Unit of observation Patient
Data source EHIF medical records
Sampling method Observed universe of records
Number of rounds of measurement N/A, ongoing
Dates of measurement N/A, ongoing
Data access confirmed? Expected date of confirmation Confirmed

Outcome #4 Patient likelihood of CVD-related hospital admission
Unit of observation Patient
Data source EHIF medical records
Sampling method Observed universe of records
Number of rounds of measurement N/A, ongoing
Dates of measurement N/A, ongoing
Data access confirmed? Expected date of confirmation Confirmed

Outcome #5 Proportion of acute CVD care episodes receiving 30 day follow up
Unit of observation Patient-episode
Data source EHIF medical records
Sampling method Observed universe of records
Number of rounds of measurement N/A, ongoing
Dates of measurement N/A, ongoing
Data access confirmed? Expected date of confirmation Confirmed
Primary Outcomes (explanation)
Provider participation measures will depend on the data structure of the program rollout that is decided by the implementing partner and what measures are available; i.e. the number/proportion of eligible patients enrolled into the ECM program or a similar indicator.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Treatment group 1 (Big-push mentoring) Providers in the first treatment group will be given an intense week of mentoring on ECM practices by a peer-doctor who has already successfully adopted the reform during the pilot phase. The peer will spend a week at the treatment provider's practice, working collaboratively to explain and work through each stage of adoption of ECM practices. Throughout that week, a structured curriculum will guide the mentoring sessions and cover all aspects of the reform for the new provider to adopt. Mentors have already been identified and screened as both knowledgeable regarding the ECM reform, personable, and engaging. They were chosen as early adopters with the aim that they would provide technical assistance to the rollout documented here.

Doctors in this treatment group (and all others) will then receive, on a weekly basis, an automatically-generated ECM Performance Report with data visualizations (graphs and charts) showing their adherence to ECM practice standards, and how their adherence has evolved over time. This report will be drawn from comprehensive EHIF medical records, meaning that providers will receive up-to-date information on their management of all ECM patients, as well as specific updates on individual ECM patients as appropriate. This report will be delivered to all three treatment arms as well as to the comparison arm.

Treatment group 2 (Persistent mentoring) Providers in this treatment group will receive ongoing support in ECM implementation, delivered in weekly one-hour sessions by an "ECM Coach" whose total budgetary cost to the Government is equivalent to that of the training for Treatment Group 1, and time investment approximately the same. This person will, on a weekly basis, respond to any questions about the ECM program, assist with specific case management, and discuss the treatment provider's performance and its evolution over time. They will support ongoing discussion about approaches to improve performance based on the information that is fed back from the ECM Performance Report.

Treatment group 3 (Big-push and persistent mentoring) Providers in this treatment group will receive the same ECM Performance Report, the intensive mentoring week, and the ongoing phone-based coaching (from a distinct mentor doctor).

The first comparison group will consist of the remaining 40 physicians in the rollout sample, who will receive, on a weekly basis, the automatically-generated ECM Performance Report with data visualizations (graphs and charts) showing their adherence to ECM practice standards, and how their adherence has evolved over time. This ensures that access to information on performance is equivalent across the three study groups and the comparison group.

We will also be able to construct a distinct control group at the patient level. We expect to have sufficiently rich data at the patient-interaction level over time to compare ECM providers' management of ECM practices before, during, and after their transition into the program; as well as against ECM-eligible patients served by providers who are not selected into the rollout sample.
Experimental Design Details
Randomization Method
randomization done in office by a computer
Randomization Unit
Health clinic
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
160 Health Clinics
Sample size: planned number of observations
267,200 patients
Sample size (or number of clusters) by treatment arms
40 clinics – Big-push mentoring
40 clinics - Persistent mentoring
40 clinics - Big-push and persistent mentoring
40 clinics - Information only
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
OUTCOME 1: Take-up – Number of follow-up calls per CVD or metabolic triad patient Proposed sample size Treatment 1: 40 providers Treatment 2: 40 providers Treatment 3: 40 providers Comparison: 40 providers Assumed attrition: N/A Assumed take-up of intervention: N/A Assumed baseline mean of variable: 1.8 calls Assumed standard deviation of variable: 2.6 calls POWER CALCULATIONS Data source for power calculations: EHIF Medical Records Power: 0.80 Power 0.90 Number of clusters 160 160 Size of cluster 8 – 618 (mean = 170) 8 – 618 (mean = 170) Significance level (alpha) 0.05 0.05 Sample size 27,200 27,200 Minimum detectable effect 0.76 calls 0.88 calls OUTCOME 2: Statins – Proportion of CVD or lipid-indicated patients with statin prescriptions Proposed sample size Treatment 1: 40 providers Treatment 2: 40 providers Treatment 3: 40 providers Comparison: 40 providers Assumed attrition: N/A Assumed take-up of intervention: N/A Assumed baseline mean of variable: 0.379 (38%) Assumed standard deviation of variable: 0.485 POWER CALCULATIONS Data source for power calculations: EHIF Medical Records Power: 0.80 Power 0.90 Number of clusters 160 160 Size of cluster 1 1 Significance level (alpha) 0.05 0.05 Sample size 20,800 20,800 Minimum detectable effect 0.052 (5.2p.p.) 0.061 (6.1p.p.) OUTCOME 3: Hospital admissions – Patient likelihood of CVD-related hospital admission Proposed sample size Treatment 1: 40 providers Treatment 2: 40 providers Treatment 3: 40 providers Comparison: 40 providers Assumed attrition: N/A Assumed take-up of intervention: N/A Assumed baseline mean of variable: 0.0120 (1.20%) Assumed standard deviation of variable: 0.09 POWER CALCULATIONS Data source for power calculations: EHIF Medical Records Power: 0.80 Power 0.90 Number of clusters 160 160 Size of cluster 126 – 2974 (mean = 1670) 126 – 2974 (mean = 1670) Significance level (alpha) 0.05 0.05 Sample size 267,200 267,200 Minimum detectable effect 0.004 (0.04p.p.) 0.004 (0.04p.p.) OUTCOME 4: Acute follow-ups – Proportion of acute CVD care episodes receiving 30-day follow up Proposed sample size Treatment 1: 40 providers Treatment 2: 40 providers Treatment 3: 40 providers Comparison: 40 providers Assumed attrition: N/A Assumed take-up of intervention: N/A Assumed baseline mean of variable: 0.59 (59%) Assumed standard deviation of variable: 0.49 POWER CALCULATIONS Data source for power calculations: EHIF Medical Records Power: 0.80 Power 0.90 Number of clusters 160 160 Size of cluster 1 – 70 (mean = 20) 1 – 70 (mean = 20) Significance level (alpha) 0.05 0.05 Sample size 3,200 3,200 Minimum detectable effect 0.09 (9p.p.) 0.11 (11p.p.)
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials