Importance of Societal Factors + Impact on Health Decisions

Hospitals and Health Systems are working on capturing social determinants of health data in efforts to improve patient and community care. Identifying systemic causes that lead to health inequalities include societal factors such as the following: 

Z CODES. The International Classification of Diseases version 10 (ICD-10), provides a group of codes for reporting factors that influence a patient’s health status and their interaction with health services. The Z codes (Z00 to Z99) may be used when:

Care is provided, but there is no specific disease or injury to report. 

E.g. Z71.1 Person with feared complaint where no diagnosis is made. 

When something other than an illness or injury affects a person’s health status.

E.g. Z59 Problems related to housing and economic circumstances. 

We all recognize that societal factors can influence health. JBH Solutions considers it best practice for hospitals to report Z codes. Having a standardized approach to documenting, screening, and coding social needs, allows us to deliver on the promise of patient-centered care.  Including these issues within the health record, will increase awareness among case management, physicians, and other clinical providers. It can improve care coordination in the acute and post-acute care settings. There are downstream benefits including improvement in key performance indicators, especially for those facilities that are operating under performance-based initiatives, or those receiving population-based payments. 

Who can report Z codes? With very few exceptions, all diagnoses must be documented by the treating physician or allied professional provider, to qualify as a reportable principle or secondary diagnosis. The Z codes are a major exception as noted in the FY2022 ICD-10-CM Coding Guidance.

CODING GUIDANCE:  FY2022 ICD-10-CM Official Guidelines (page 100-101) - SDOH Codes in ICD-10-CM For social determinants of health, such as information found in categories Z55-Z65, Persons with potential health hazards related to socioeconomic and psychosocial circumstances, code assignment may be based on medical record documentation from clinicians involved in the care of the patient who are not the patient’s provider since this information represents social information, rather than medical diagnoses. For example, coding professionals may utilize documentation of social information from social workers, community health workers, case managers, or nurses, if their documentation is included in the official medical record. Patient self-reported documentation may be used to assign codes for social determinants of health, as long as the patient self-reported information is signed-off by and incorporated into the medical record by either a clinician or provider.

At the time of a hospital admission, the physician must weigh the risks of outpatient management. The ever-present question is ‘could the patient be managed in an alternative setting or lower level of care (e.g. home, skilled nursing facility)?’ In its simplest terms, this translates to ‘is there an increased risk that bad things will happen if the patient is not hospitalized?’ For the physician advisor and utilization management team, the inclusion of social determinants of health can be the difference in supporting medical necessity for an inpatient hospitalization, as opposed to observation services or an outpatient treatment plan.

Here are two cases to illustrate how social determinants affect the inpatient versus outpatient equation: 

Patient A has a simple pneumonia. The Pneumonia Severity Index calculation  of 72 points, shows he falls into the Class III mortality risk category (0.9 to 2.8% risk of death) and is appropriate for either outpatient management, or a brief hospitalization with observation services. The attending places an order for hospitalization with observation services. The point range for this group is 71-90. The insurance company may argue that with 72 points, the patient is at the low end of illness severity, and outpatient management is more appropriate.

Patient B also has a simple pneumonia. The Pneumonia Severity Index calculation shows a score of 62, consistent with Class II category (0.6 to 0.9% mortality risk) and is appropriate for outpatient management. The physician places an order for observation services. The case manager screening for medical necessity, documents that criteria support outpatient management.

However, what if in both cases, the doctor documents the presence of several social determinants of health including;

• Homelessness
• No insurance or money to pay for the antibiotic prescription
• No transportation to get to a pharmacy, or physician office
• No primary care physician
• Health care illiteracy
• Lack of a social support network

Now the patients have a much higher risk of adverse events and treatment failure if they are managed in the outpatient setting. When the social determinants are included, there is a much stronger argument that hospitalization is medically necessary.

Here is a real-life example of how capturing Social Determinants of Health (SDOH) can impact the Medical Necessity of hospitalization. My classic case example is a woman who presented to the ED after falling at home, and found to have an uncomplicated fracture of the humerus. This is appropriate for outpatient management. However, when I met her in the Upper Peninsula of Michigan during a February snowstorm, my documentation stated this was "A 92 year-old woman, living alone, with no close family or friends. She was treated one month earlier for an acute upper GI hemorrhage due to an NSAID-induced gastric ulcer, and has previously developed acute encephalopathy when taking oral opioids. She uses a front-wheeled walker for mobility due to a previous stroke with residual left hemiparesis. She will be hospitalized for skilled, restorative physical and occupational therapy. She will require close monitoring of potential side effects, and her clinical response to the initiation of medications to treat her acute pain."

The fracture, taken in isolation, does not require hospital management. However, she relies on that arm for weight-bearing on her walker. She has a high risk of complications if started on NSAID agents for pain control, and may develop side-effects when taking opioids. She will not be able to perform basic activities of daily living. Without family or friends nearby, she is at increased risk of complications including additional falls with injury, pressure injuries, and many others. I would argue that discharging the patient from the emergency department would be ill advised.  The inclusion of several Z-code issues adds clarity around the medical necessity of the admission.  Side note: Just as important, the documentation clarifies the admission is for active management, not custodial care.

Taking this concept to the next level, we can leverage the Z codes through an AI-based automated alert system to impact the hospital’s key performance metrics. There is a wide range of opportunities, and for illustration, we’ll use findings from a recent JBH on-site physician advisor assessment:

CASE STUDY - SYSTEM ISSUE. Length of Stay (LOS) for inpatient hospitalizations was significantly above budget. This was in part due to unnecessary emergency department (ED) visits and hospital readmissions. These events contributed to ED overcrowding, slow throughput, and decreased patient satisfaction due to the system delays. In addition, overall hospital LOS was increased. It is typical for patients readmitted within thirty days of the index (initial) hospitalization to have a longer LOS during the second admission.

STEP 1. Gather baseline data. We screened by unique patient identifier (medical record number) all ED visits over the last 12 months to a community teaching hospital of > 200 beds.  

STEP 2. Identify opportunities for improvement. A sample of individuals with 15 or more visits in the past year were reviewed by the physician advisor. The primary goal was the identification of modifiable risk factors that contributed to this high volume of clinical encounters.

 PROMINANT FINDINGS: 

TRANSPORTATION ISSUES. There have been calls for specific Z codes related to transportation issues. However, they are currently reported under Z59 - problems related to housing and economic circumstances. Many people in this target community were found to have difficulty getting to their outpatient physician appointments. This is especially true for sub-specialists whose offices may be 1-2 hours away. Missing an appointment has a greater impact than just annoying the clinic staff. The failure to receive ongoing management of their medical issues, leads to clinical worsening,  return visits to the ED, and often an avoidable  inpatient hospitalization. 

DIALYSIS ISSUES. Z91.15 - noncompliance with renal dialysis. We identified patients frequently presenting to the ED, having missed one or more dialysis sessions, but not needing emergency dialysis. The reasons for missing treatment varied from transportation issues, mild illness, or general non-compliance due to other psychosocial issues. Despite the absence of other active issues,  these patients were hospitalized for 2+ days for what amounts to outpatient therapy.

STEP 3. Develop interventions based on the observations.

SOLUTION INTERVENTIONS INCLUDE:

TRANSPORTATION ISSUES. Any barriers to the post-hospital care plan, including transportation challenges are ideally identified during the index hospitalization and addressed in the discharge plan. However, we know that even in the best of times, things slip through the cracks. One solution is an FHIR-based alert system, identifying all patients with an early return to the emergency department. This population can be targeted for specialized interventions. Following our recommendations, this health system is exploring the use of satellite telehealth centers to minimize the transportation issues, increase compliance with outpatient followup, decrease unnecessary ED visits, and decrease hospital readmissions.  

DIALYSIS ISSUES. For patients that present to the ED, having a dialysis session, and not requiring emergency treatment, the answer seems simple. I.e. just have them go to their outpatient dialysis center for a session. However, as is true for many clinical issues we encounter, the correct action is overly complex and difficult. It is much easier to admit the patient to the hospitalist, consult nephrology for dialysis, and involve case management to coordinate the post-discharge plan. This invariably leads to 2-3 days in the hospital. A FHIR-based solution is available to identify patients on dialysis that present to the ED. Early notification allows coordination between nephrology services and case management, to determine if diversion from the ED to the dialysis center for same-day, or next-day treatment, is appropriate.  Similarly, for hospitalized patients, automated alerts are available for early identification of dialysis patients with a discharge order. Payers do not typically pay for dialysis services provided in the inpatient setting on the day of discharge. Therefore dialysis performed as the patient is leaving the hospital, can have a significant negative impact on hospital finances. Early notification allows the care team to coordinate resumption of outpatient dialysis services, including the first session on the day of discharge if needed.

To learn more about how triggers and AI can help support your care team, contact us today.