Find Patients in Healthcare Data: 6 Steps

Life sciences and Biotechnology companies with specialty and rare disease therapies are uniquely challenged to find patients. The number of applicable patients may be small, their diagnosis may be challenging to come by, and their transit across the healthcare system is often incomplete. Further complicating this challenge is using dated, incomplete data and inefficient methods to identify these patients and their providers.  

So, how can those charged with commercialization, engagement, and market access find patients and providers? Simply using raw claims data falls short of pinpointing patient-level results. Whether for assessing new therapy market potential or targeting providers for engagement, data users require a more holistic, timely, complete, and granular view to find the right patients and providers at the right time. Patient-level data helps Life Sciences and Biotech users target only the healthcare providers and organizations currently seeing rare disease patients and those most likely to care for these patients in the future. Kythera Labs delivers more accurate results by linking de-identified datasets and remastering data to enable more accurate analysis, which helps patients get the treatment they need.

Let’s look at wet macular degeneration as an example of the six steps for finding rare patients and providers in remastered real world data (RWD). This guide is a suggested approach and is not intended to be a comprehensive list of treatments, codes, etc.

Wet Age Related Macular Degeneration
Age-related macular degeneration (ARMD) is an eye disorder and the leading cause of incurable blindness in the elderly worldwide. In the United States, 11-15 million people have ARMD, and of these, 10% of patients with dry ARMD will develop wet ARMD. Approximately 200,000 new cases of wet ARMD are diagnosed each year, leading to 90% of legal blindness. Untreated, wet ARMD leads to loss of vision in the majority of patients.

Wet Age-related Macular Degeneration (Wet ARMD)
Samuel D. Hobbs; Kristine Pierce.

How to Accurately Find Patients Using Granular Data

Using data, Life Sciences and Biotech companies can build a path to finding these patients and their providers with higher accuracy. Here are the 6 steps:

1. Diagnosis Codes

Identify patients who received a diagnosis of wet ARMD within remastered claims data during a specific lookback period ranging from last week to January 2016 using applicable ICD-10 codes.

Diagnostic Codes: Wet-Age Related Macular Degeneration (wet AMD)

2. Symptoms

Examine EHR data and use symptoms typically reported with wet ARMD.

  • Visual distortion or blurring of central vision (ICD-10 H53.15)
  • Metamorphosia  (ICD-10  H53.15)
  • Micropsia  (ICD-10 H53.15)
  • Scotoma  (ICD-10  H53.429)  
  • Other vision complaints ( (ICD-10  H54.7)
  • Decreased best-corrected visual acuity (BCVA) (ICD-10  H54.7)
    Fine AM, Elman MJ, Ebert JE, Prestia PA, Starr JS, Fine SL. Earliest symptoms caused by neovascular membranes in the macula. Arch Ophthalmol. 1986 Apr;104(4):513-4. [PubMed]

3. Risk Factors

Look for known risk factors associated with ARMD within claims, EHR, and other data sources.

Medical Claims Data

  • Atherosclerosis
  • Age 75 and older. The risk of advanced age-related macular degeneration increases from 2% for those ages 50-59 to nearly 30% for those over 75.
  • High density lipoprotein cholesterol (HDL-C)

Diabetic Retinopathy EHR Data

  • Race
  • History of cigarette smoking
  • Serum Cystatin C

Genetic Data

  • Genetic predisposition to ARMD - at least 34 genetic loci and 52 gene variants are linked to the disease.
    Black JR, Clark SJ. Age-related macular degeneration: genome-wide association studies to translation. Genet Med. 2016 Apr;18(4):283-9. [PMC free article] [PubMed]

4. Procedures

Look for the procedures associated with diagnosing and treating ARMD within claims data.


  • Ophthalmologic examination (ICD-10 z01.01, CPT code 92002-92014)
  • Fundus Imaging (CPT code 92250 )
  • Fluorescein Angiogenisis (FA) (CPT code 92235, 92242)
  • Indocyanine Green Angiography (ICGA) (CPT code 92240)
  • Optical Coherence Tomography (OCT) (Procedure codes 92133, 92134)
  • Optical Coherence Tomography (OCT) + AI (Artificial Intelligence)
  • OCT angiography (OCT-A) (Procedure Code 92134)

    One could dig even deeper into the findings of these evaluation modalities. For example, when using Optical Coherence Tomography (OCT), a noninvasive modality that captures detailed images of the retina, wet ARMD is characterized by the following findings:
  • Subretinal and intraretinal fluid (ICD-10  H35.81)
  • Serous retinal pigment epithelial detachment (PED) (ICD-10  H35.721, H35.721, H35.723, H35.732)
  • Fibrovascular PED (FVPED)
  • Hemorrhagic PED (ICD-10 H33.75)
  • Disciform scarring (ICD-10 35.3233, H35.3213)

5. Treatments and Site of Care

Look for specific therapies used to treat ARMD in medical and pharmacy claims, specifically the HCPC and J codes for medically administered therapies and the site of care for that  appear on the medical claim.

  • Ranibizumab injections (Brand names Susvimo, Lucentis, Byooviz, Cimerli) (CPT codes 67028, HCPCS codes J2778, J2779, J3590;  NDC: 50242-078-12, 50242-078-55, 50242-080, 64406-019,70114-0441-01, 70114-0440-01 )
  • Aflibercept injections (Brand name Eylea) (CPT 67028, HCPCS code J0178; NDC 61755-0005-02)
  • Brolucizumab injections (Brand name Beovu) (CPT 67028, HCPCS code J0179, J9035, J3490, J3590, and J7999;  NDC 0078-0827,)
  • Photodynamic therapy (PDT). PDT can be combined with ranibizumab. (CPT code 96573, 96574)

Providers administering these injections (PDT) often practice in multiple locations. Pinpointing the site of care where a patient encounter takes place offers unique insight into segmenting where and when services were delivered.

6. Providers

Not only can ARMD patients be identified within RWD, providers treating ARMD can too. Typically, patients with ARMD are seen by various healthcare providers before being referred to a retina specialist. So, primary care providers, optometrists, general ophthalmologists, and retina specialists should be included when identifying providers treating rare patients. It should be noted that in many regions of the US, retina specialists may not be accessible; therefore, widening the lens to include general ophthalmologists is essential.

In addition to specific providers, the data should let you drill into specific health care organizations down to the site of care level and segment providers by a number of variables, including:

  • Number of unique patients versus new patients
  • Prescribing and treatment behaviors, including newly written prescriptions  
  • Patient journey information such as patients who are newly diagnosed, starting therapy, and procedure outcomes

Using More Data Dimensions for More Accurate Results

With remastered RWD, Life Sciences and Biotech users can analyze a combination of symptoms, diagnosis, procedures, risk factors, therapies, precise site of care, and referral behaviors to create a more complete, nuanced view of diagnosed and pre-diagnosed patients, not just patient counts. This holistic, granular view more accurately identifies the targeted patient population and helps build a model to predict patients who may have a future diagnosis.  

Connect with Kythera Labs to learn more about using Wayfinder, our big data platform, to access 15+ Terabytes of remastered claims data and integrate data like EHR or genetics to pinpoint your rare disease patients and their providers.

Photo by Markus Spiske on Unsplash

Find Patients in Healthcare Data: 6 StepsLinkedIn

Kythera Labs


Find Patients in Healthcare Data: 6 StepsLinkedIn