Reducing Readmissions using Health I.T.

Introduction

Healthcare costs are rapidly increasing in the United States. In the year 2000 we spent about $1.4 trillion on healthcare, and in 2010 we spent $2.6 trillion. We almost doubled the cost of healthcare in 10 years, which now accounts for nearly 18% our GDP [1]. A majority of the increase in these costs is driven by the rise in care rendered in the hospital (from $415.5 billion in 2000 to $815.9 billion in 2010).

The government has recognized the growth in healthcare spending is unsustainable. It has therefore introduced a number of metrics, as part of the Affordable Care Act of 2009 (ACA) to try to curb this spending.

The Readmission Reduction Program is one such metric. [2]

Medicare Readmission Measures

Section 3025 of the Affordable Care Act, established the Hospital Readmissions Reduction Program [3]. This program requires CMS to cut payments to hospitals for excess readmissions within 30 days of discharge, beginning Oct 1, 2012.

Under this program the financial penalties will be levied against hospitals that have readmissions in excess of the calculated rate for certain diseases, including acute myocardial infarction (AMI), congestive heart failure (CHF), and pneumonia. The program takes into account different patient characteristics (patient demographics, co-morbidities, and patient frailty), by calculating risk adjusted readmission rate [2] in an attempt to capture the variability of patient populations that are served by hospitals.

Per CMS calculations, approximately two-third of US hospitals will receive penalties up to 1% of their reimbursement for Medicare patients starting Oct 2012; these penalties will increase to 3% in 2015.

Penalizing hospitals for readmissions, however is controversial. The controversy is mainly centered around two arguments [4]:

  1. Holding hospital accountable for factors outside their control and
  2. The inadequacy of attributes used to calculate predicted readmission rate

Controversies aside, the penalties are here to stay.

Predicting Readmission Risk in Patients

The loss of revenue from readmissions is a big threat to the bottom line of hospitals. To counteract this threat, hospitals have started looking at readmission risk prediction tools, hoping to predict which patients are at higher risk of readmissions, and taking proactive measures to prevent them from coming back to the hospital.

There are many readmission risk prediction tools published in the literature. Some of the major tools currently in use are described below:

1. LACE Index

The LACE index was published in Canadian Medical Association Journal in 2010 [5]. The mnemonic LACE includes:

  1. Length of stay
  2. Acuity of admission
  3. Co-morbidity of patient measured using Charlson comorbidity index
  4. Emergency department use (measured as the number of visits in the six months before admission)

The LACE index is calculated by summing the points of the attributes that apply to the patient (Table 1). The expected probability of death or readmission within 30 days ranges from 2% for a LACE score of 0 to 43.7% for a LACE score of 19.

The LACE index is simple, clinically relevant, and can be reliably measured. However, reliability of the index in other populations, other than the study population is limited.

The LACE Index
Attribute Value Points
Length of Stay (“L”) < 1 0
1 1
2 2
3 3
4–6 4
7–13 5
>=14 7
Acute (emergent admission) (“A”) Yes 3
Charlson Comorbidity Index Score (“C”) 0 0
1 1
2 2
3 3
>=4 5
Visits to Emergency department during previous 6 months (“E”) 0 0
1 1
2 2
3 3
>=4 4

2. Charlson Comorbidity Index

Charlson Comorbidity index was developed in 1987 for classifying co-morbid conditions, which might alter the risk of mortality. This index was developed predominantly for use in longitudinal studies [6]. In recent years, this index has been used for multiple other purposes including readmission risk prediction.

This system assigns a co-morbidity score, and an age score. The comorbidity score assigns a numerical value to many common diseases (e.g. Myocardial infarction, CHF, COPD). The age score is also a numerical value assigned based on the patients age. The Charlson score is a total of the comorbidity score and the age score. This score is then used to calculate the Charlson probability viz., 10 year predicted mortality. [7]

Charlson Score = Co-morbidity score + Age Score

Charlson probability = Calculated from Charlson Score

3. Project Boost

The Boost Team at the Society of Hospital Medicine has identified patient specific risk factors and created a tool, 8P scale. [8] This risk assessment tool is completed at admission, and identifies patients who are at increased risk of adverse event after discharge. The advantage of this tool is that, it allows clinicians to influence and change many of the risks during the period of hospitalization, which in turn could mitigate post discharge adverse events.

The 8 P’s in this tool are:

  1. Problem medications
  2. Psychological
  3. Principal diagnosis
  4. Poly-pharmacy
  5. Poor health literacy
  6. Patient support
  7. Prior hospitalizations
  8. Palliative care

Health Information Technology and Risk Prediction Tools

Health Information Technology is in a unique position to capture data that is required by readmission risk prediction tools – at the point of care.

Current generation Electronic Health Records have the capability to build these tools in the system, to capture and present this information to clinicians. Furthermore, data modeling techniques using electronic data collected via multiple EHR’s have the potential to develop more sophisticated algorithms. Such algorithms, in the future may have the capability to take these multiple data points into account, and risk stratify readmission risk based at the individual level, and not at the population level.

Besides development of algorithms & models based on the readmission risk assessment measures discussed above, Health IT can influence readmission risk in multiple other ways. Some of these methods are discussed below.

Transition of Care

Currently, due to paper processes, critical information is lost in transitioning care of the patient from one healthcare entity to another. Patients and caregivers often report that their needs are unmet during and after discharge; [9] this lack of information and unmet needs leads to readmissions. Health IT is uniquely in the position to enable information sharing between these disparate healthcare entities, and by proxy enabling providers to cater to the needs of their patients.

1. Health Information Exchange (HIE):

Establishing HIE’s to connect different Electronic Health Records will allow exchanging crucial information (viz. Allergies, problem list, medications, summary of care) that is often lost during the transfer of patient. When physicians in the community following up patients recently discharged from the hospitals lack this information, the risk of re-hospitalization increases. [10]

2. Mergers & Acquisitions:

Mergers and/or acquisitions of different health care entities may encourage them to adopt common Electronic Health Records, or push for greater information sharing across different healthcare entities.

3. Other modes of Electronic Communication:

Utilization of other methods to exchange information electronically (e.g. The Direct Project) may also aid in improved patient care and reduced hospital readmissions.

Post Discharge Follow up

After a patient is discharged from the hospital, follow up care is extremely important, and has shown to reduce readmissions by catching problems early on. Post discharge follow up can occur in various ways:

1. Phone call:

A nurse, pharmacist or other healthcare provider typically calls the patient a few days after discharge to confirm the patients’ health, medications and follow up appointments. If any of these elements (or others as determined by individual organizations) are missing, the healthcare provider takes necessary corrective action. [11] These actions can include scheduling a home nurse visit, ensuring patient fills his/her prescriptions, and/or confirming physician appointments.

2. Case Management:

Case management generally encompasses phone calls, home visits and home care. Chronically ill patients, who have the highest risk of readmissions, typically should have case managers and/or Patient Care Coordinators assigned to their care. These case managers routinely follow up with their patients to follow their symptoms, ensure that they are taking medications, following physician recommendations and appointments. They are trained to catch early warning symptoms, and intervene to prevent deterioration of patients’ health thereby preventing readmission. These case managers use disease management care pathways in existing EMR’s and other tele-health technologies to reduce readmissions. [12]

3. Patient monitoring devices:

Patient monitoring devices have been in use for quite some time, although in only a handful of situations e.g. Home Holter monitor for arrhythmia, Lifeline medical alert systems, blood glucose monitors. Technology is becoming cheaper every year. This is opening up new avenues for home monitoring, which were previously thought impossible. Furthermore, the advent of the Internet has made the transmittal of this information near instantaneous. Examples include a BP monitor, and weight scale from Withings that records information and can wirelessly upload this information to patients Personal Health Record, from where it can be accessed by a physician (assuming the appropriate security and privacy measures are in place), who can then take action if needed.

Education and Engagement of patients

Education and sharing information with patients is extremely important to engage patients in their care. Engaged patients have better health outcomes, make better use of resources, [13] and by extrapolation may have a reduced risk of admission and readmission.

There are various ways in which patients can be engaged to take part in their care:

1. Access to their data:

Today, patients can access medical information on the Internet, but they do not have access to their own data. This patient data is locked away in paper charts (and even electronic health records) in physician offices and hospitals. Allowing patients access to their own information will help them understand their health conditions, and therefore motivate them to be more engaged. The OpenNotes Project clearly demonstrated that if physician notes are shared with patients, health outcomes improve. [14]

2. Educational tools during hospitalization:

Patients are generally sick, fatigued and frightened while they are in the hospital. Traditional methods of teaching patients in these situations, which is mainly a one way information delivery by physician or nurse, are generally ineffective. In fact, patients forget most of the verbal information that is delivered to them in the hospital. [15] Providing written material to patients at discharge allows patient to refer back to their medical information at home .

The real opportunity here is in delivering rich multimedia based education either by using in room televisions or bedside kiosks. Several companies such as GetWell Networks have products that can deliver audio, video and other multimedia content via these avenues, and at a time when patients’ are ready to assimilate this information. Such improved information delivery methods that are under the control of the patient, may allow patients to understand their health conditions much better. These educated patients are, in turn more likely to follow healthcare advice, recognize alarm symptoms earlier, and possibly decrease their utilization of critical hospital services including readmissions.

3. Patient portals:

Patient portals are the patient facing side of the electronic health record. HealthIT.gov defines a patient portal as, “A patient portal is a secure online website that gives patients convenient 24-hour access to personal health information from anywhere with an Internet connection.”

Providers can choose to push information to the patient portal, such as educational material, results of lab and radiology tests, reminders for health maintenance exams and other patient pertinent information. This allows patients to take control of their information, be better informed and more engaged. [16]

4. Personal Health Records (PHR):

Patients control their Personal health records (PHR’s), as compared to Patient Portals, which are controlled by the hospital. Medicare.gov defines a PHR, as “A Personal Health Record (PHR) is a record with information about your health that you, or someone helping you, keep for easy reference using a computer. You control the health information in your PHR and can get to it anywhere at any time with Internet access.”

Patients’ have the ability to pull in data from multiple patient portals, besides adding their own information. They can track their health statistics such as blood pressure, blood glucose, weight etc, by linking monitoring devices directly to their PHR. Access and use of a well designed PHR allows patients to track their own health data, and take action as needed. [17]

Implementation and Incentives Alignment

As discussed above, there are multiple tools that can be implemented to reduce readmissions, increase patient safety, satisfaction and outcomes. However, there are also a number of business challenges.

Implementation of these tools can be very costly and require huge resources from organizations. Therefore, strong leadership support, and support from other stakeholders, including patients is critical for the success of these projects. Furthermore, incentives need to be aligned so that all parties benefit from the implementation, patient outcomes improve and health care costs are reduced.

Conclusion

Medicare readmission measures and payment penalties are here to stay. Health IT is in a unique position to transform the entire healthcare landscape by providing the necessary tools in the hands of leaders and frontline healthcare professionals to reduce the risk of readmissions.

These tools can take the data entered into electronic health records, risk stratify patients based on ever-improving algorithms, and impart this information to all the parties (including patients) to improve the quality and value of care that is delivered.


  1. CMS. “National Health Expenditures; Aggregate and Per Capita Amounts, Annual Percent Change and Percent Distribution: Selected Calendar Years 1960–2011.” Centers for Medicare & Medicaid Services, Office of the Actuary, National Health Statistics Group. April 19 2013..  ↩
  2. “Readmissions Reduction Program.” Center for Medicare & Medicaid Services. April 19 2013.  ↩
  3. “The Patient Protection and Affordable Care Act: Section‐By‐Section Analysis.” United States Senate Democrats. April 19 2013.  ↩
  4. Joynt, KE, and AK Jha. “A Path Forward on Medicare Readmissions.” N Engl J Med 368.13 (2013): 1175–77.  ↩
  5. van Walraven, C, IA Dhalla et al. “Derivation and Validation of an Index to Predict Early Death or Unplanned Readmission After Discharge From Hospital to the Community.” CMAJ 182.6 (2010): 551–57.  ↩
  6. Charlson, ME, P Pompei et al. “A New Method of Classifying Prognostic Comorbidity in Longitudinal Studies: Development and Validation.” J Chronic Dis 40.5 (1987): 373–83.  ↩
  7. FPNotebook. “Charlson Comorbidity Index.” Family Practice Notebook. April 19 2013.  ↩
  8. SHM. “Risk Assessment Tool: The 8ps.” Society of Hospital Medicine.  ↩
  9. Grimmer, KA, JR Moss, and TK Gill. “Discharge Planning Quality From the Carer Perspective.” Qual Life Res 9.9 (2000): 1005–13.  ↩
  10. van Walraven, C, R Seth et al. “Effect of Discharge Summary Availability During Post-Discharge Visits on Hospital Readmission.” J Gen Intern Med 17.3 (2002): 186–92.  ↩
  11. Dudas, V, T Bookwalter et al. “The Impact of Follow-Up Telephone Calls to Patients After Hospitalization.” Am J Med 111.9B (2001): 26S–30S.  ↩
  12. Darkins, A, P Ryan et al. “Care Coordination/Home Telehealth: The Systematic Implementation of Health Informatics, Home Telehealth, and Disease Management to Support the Care of Veteran Patients With Chronic Conditions.” Telemed J E Health 14.10 (2008): 1118–26.  ↩
  13. Coulter, A, and J Ellins. “Effectiveness of Strategies for Informing, Educating, and Involving Patients.” BMJ 335.7609 (2007): 24–27.  ↩
  14. Delbanco, T, J Walker et al. “Inviting Patients to Read Their Doctors’ Notes: A Quasi-Experimental Study and a Look Ahead.” Ann Intern Med 157.7 (2012): 461–70.  ↩
  15. Flacker, J, W Park, and A Sims. “Hospital Discharge Information and Older Patients: Do They Get What They Need?” J Hosp Med 2.5 (2007): 291–96.  ↩
  16. Urowitz, S, D Wiljer et al. “Improving Diabetes Management With a Patient Portal: A Qualitative Study of Diabetes Self-Management Portal.” J Med Internet Res 14.6 (2012): e158.  ↩
  17. Tang, PC, and D Lansky. “The Missing Link: Bridging the Patient-Provider Health Information Gap.” Health Aff (Millwood) 24.5 (2005): 1290–95.  ↩