Replies: 2 comments 6 replies
-
re: "Here's an example: does the in network file in the image below map to each of the plans in the table, even though some of them are Bronze, others are Silver and others Gold? Note that they all have the same 10 character HIOS ID." - they should. The 10 byte HIOS ID defines the product, the 14 byte defines the specific plan being offered. (Early on there was concern/feedback that contracts are negotiated at the product v plan level and that shaped the requirement to 10 byte HIOS ID from 14) To your question on Metal levels, those represent the actuarial value of the plan to the member - based on member cost share vs rates negotiated with providers. Here's a good CMS reference on Product v Plan - https://www.cms.gov/CCIIO/Resources/Training-Resources/Downloads/product-vs-plan-ppt.pdf |
Beta Was this translation helpful? Give feedback.
-
Re: "A product encompasses different metals which encompass different plans" - I would say a plans often have multiple metal levels - same product/benefit offering or benefit package with different share (on average) paid by the health insurance issuer in the case of ACA plans - same plan, 80% gold, 70% silver, 60% bronze AV of the plan. see https://www.qhpcertification.cms.gov/s/Plans%20and%20Benefits%20FAQs and too note the AV of a plan is expressed by bytes 15-16 of the HIOS Plan ID - the max verbosity of an ACA plan is to the 16th digit, not related to the negotiated rates. "Plans are certified at the Standard Component ID level. All cost sharing plan variations, including the Zero Cost Sharing and Limited Cost Sharing plan variations, must meet the QHP certification standards to be certified. On the Cost Sharing Variance worksheet of the Plans and Benefits template, the plan variations are distinguished by adding a code to the Standard Component Plan ID as follows:
|
Beta Was this translation helpful? Give feedback.
-
I noticed that the plans are reported at the 10 character HIOS identifier if available, otherwise by the 5 character HIOS Issuer ID and otherwise by the EIN. This is different to how plans are identified in other datasets. Importantly, I am having enormous trouble understanding how the
plan_id
maps to one or multiple real life plans.Referencing #500 by @shaselton-usds :
Here's my questions:
Here's an example: does the in network file in the image below map to each of the plans in the table, even though some of them are Bronze, others are Silver and others Gold? Note that they all have the same 10 character HIOS ID.
<style> </style>
Understanding this is mission critical as I cannot do any kind of healthcare economics research if I am unable to know what the plan underlying all of this in_network_rate information actually is. I.e. let's say I want to compare HMOs and PPOs, and see the impact on negotiated rates (somebody did this with Massachusetts claims data and saw HMOs tend to be 5% cheaper compared to PPOs, as seen in this paper). To replicate this / empower other researchers I need to tie the in_network rate information to the plan attributes information.
Similarly, if I want to provide a price transparency tool, I must be able to link the information about the plan coming from the user to actual in_network_rates information. I can't do that if the files are reports for products and not plans!
reporting_plan_id
, however, this seems not to be accurate / coherent with 5,10, or 14 digit HIOS - Potential for all to be acceptable? #500. It is areporting_product_id
- which means that many plans are actually represented by the in_network rates file. Am I right? Does this mean I need to look at thereporting_plan_name
to distinguish plans within the same product? I have noticed there are multiple plan_names sharing the HIOS ID or EINs and referencing different files...As a quick note, with the availability of table of contents files that map multiple plans to a single file, it seems that referencing plans directly by the 14 character HIOS ID might be more clear and would not cause redundancy. This data can be incredibly useful but only if we are able to integrate it with other datasets.
Would appreciate clarification to be able to work with the data knowing what it actually means. Many thanks and all the best!
Beta Was this translation helpful? Give feedback.
All reactions