Opinion
January 24, 2008.
Present: Greco, P.J., Curtin Brant, JJ.
Insurance, "Personal injury protection" benefits; Chiropractic services; "Reasonable" charges.
Consumer Protection, G.L.c. 93A; Unfair and deceptive acts or practices.
Opinion vacating judgment for defendant and returning case for new trial. Action heard in the Lawrence Division by Geary, J.
Francis A. Gaimari for the plaintiff.
Paul F. Lynch for the defendant.
Plaintiff Michael Davekos ("Davekos"), a chiropractor, commenced this suit against defendant Liberty Mutual Insurance Company ("Liberty") to recover G.L.c. 90, § 34M Personal Injury Protection ("PIP") payments for medical treatment he provided to an individual insured by Liberty, and G.L.c. 93A, § 11 damages for Liberty's alleged unfair and deceptive practices in refusing to make PIP payments in compliance with statutory mandates. After a jury-waived trial, judgment was entered for Liberty.
It was undisputed that Davekos rendered chiropractic services to Renee DiThomas ("DiThomas"), who sustained injuries in an automobile accident while riding as a passenger in a vehicle insured by Liberty. Davekos' bills for that treatment totaled $6,063.28. Although Liberty received Davekos' bills and reports in May, 2004, it did not issue its first PIP payment until October 12, 2004. When Liberty failed to make any further G.L.c. 90, § 34M payments, Davekos filed this action on February 14, 2005. After suit was commenced, Liberty made additional PIP payments on March 1, 2005 and April 8, 2005. The parties agreed that the unpaid balance of Davekos' bills at the time of trial was $394.77. They also stipulated that all of the treatment provided by Davekos was medically necessary.
Thus, the sole issue at trial on Davekos' § 34M claim for PIP payments was whether the amounts he charged for his services were reasonable. Davekos charged $175.00 for his initial examination of DiThomas (CPT code 99204) and $50.00 for manual therapy (CPT code 97140), which he performed forty-seven times. Liberty "down-coded" the initial examination to CPT code 99203 and paid only $147.73 for that service. Liberty also reduced Davekos' manual therapy fee by $6.44, paying him only $43.56 for each treatment provided.
CPT or "Current Procedure Terminology" is a system by which the American Medical Association categorizes all medical services by five-digit CPT numerical codes, and defines the requirements for the particular service identified by each CPT code.
To establish that his unpaid bills were fair and reasonable, Davekos introduced into evidence his medical bills and records certified under G.L.C. 233, § 79G. Davekos also testified that he based his fees, in part, on what other chiropractors in his area charged for the same services. Over Davekos' objection, Liberty introduced statistical summaries and graphs prepared by a company called "Ingenix" (see discussion, infra) of billing patterns Ingenix derived from its own database of CPT code service charges, and testimony by a Liberty employee that what Ingenix labeled as the 80th percentile fees for an initial examination (CPT code 99203) and manual therapy (CPT code 97140) constituted the "reasonable" charges for Davekos' performance of those services.
Davekos attempted to testify that he was aware of his colleagues' billing rates from speaking with them at seminars and seeing their bills; and that he compared his fees with those in his own zip code as listed in a nationwide industry publication called "Fee Facts." The trial judge sustained Liberty's objection to that evidence.
Davekos argues on this appeal that the admission into evidence of the Ingenix materials constituted reversible error. Ingenix, a wholly owned subsidiary of United Health Corporation, is a self-styled nationwide "health care information company" that sells "customized fee analyzers" to medical providers, health care insurers and automobile liability insurance companies. The "customized" analysis or "medical module" purchased from Ingenix by Liberty is supposed to chart the billing pattern for any CPT code service in a "geo-zip" area based on the number of times that service was performed and the frequency of the dollar amounts billed. As indicated, over Davekos' objection, Liberty introduced into evidence Ingenix data, summaries and graphs purportedly showing the billing patterns in Davekos' geographical area for initial examination and manual therapy charges at the time of his treatment of DiThomas.
The Ingenix summaries and spreadsheets introduced by Liberty purport to show, for the two individual medical services at issue herein (CPT code nos. 99203 and 97140), the low and high fees as well as the "mean" (average), median (50th percentile) or mode (most commonly occurring charge) of at least those billings contained in Ingenix's database. Utilizing Ingenix analyses, Liberty generally reimburses PIP claimants in amounts consistent with what Ingenix has identified as the 80th percentile of billings for the service in question, deeming that amount to constitute a reasonable charge for the service.
Ingenix divides all states, including Massachusetts, into "geo-zips" composed of cities and towns sharing three-digits of postal zip codes, which are grouped together by not only geographical proximity, but also by what Ingenix decides are "data similarities." Davekos, who practices in Peabody (zip code 01960), falls within "geozip 018," which includes all 018 and 019 zip code billings in the Ingenix database.
We recognize that where a general objection to the admission of evidence is overruled, there is no error if the evidence was admissible for any purpose. See Karjavainen v. Buswell, 289 Mass. 419, 430 (1935). Arguably, the records objected to in this case may be relevant to Davekos' G.L.c. 93A claim in that they show the process undertaken by Liberty to establish an appropriate fee. However, counsel for Davekos specifically objected to the hearsay nature of the records and to their reliability. In overruling the objection, the trial judge ruled that they were reliable and that the witness was qualified to testify about them. In this context, it is clear that the objection related to admissibility of the records to establish what in fact medical providers in the area were billing.
However, there is nothing in the record to establish the accuracy or reliability of Ingenix's raw data and, thus, its statistical extrapolations. Ingenix does not collect its own billing data, and did not undertake any verification of the accuracy of the particular data Liberty introduced in this case. Instead, Ingenix relies on its "data contribution program" in which some, but not all, of only those health insurers that are Ingenix clients submit information, on a purely voluntary basis, about the amounts they happen to have been billed by an undisclosed number of unidentified health care providers for specific CPT code services. While Ingenix requires those of its health insurer clients that elect to participate to certify that the number and amount of the CPT code billings they submit are accurate, there is no Ingenix mechanism to enforce or validate the client certificates. Although Carolyn Gee ("Gee"), Ingenix Manager of Research and Development, mentioned occasional audits by Ingenix of the data it receives, she had no knowledge of when such an audit had last been made. Indeed, Ingenix itself prints the following disclaimer on its products: "The database is provided for informational purposes only and Ingenix disclaims any endorsement, approval or recommendation of data in the database."
Further, the billing data volunteered by some of its clients to Ingenix lists only the medical service CPT code number, the bill amount, the date of the service, and the provider's zip code. The data does not name, or otherwise identify, the medical providers who billed Ingenix insurer clients and whose bill amounts were then passed along to Ingenix. Nor does the information disclose the total number of providers whose charges may make up the Ingenix database at any point. Ingenix cannot guarantee that all of the bills received for a particular CPT code service at any given time have been reported, much less accurately reported, by its volunteer insurers. Nor can Ingenix ascertain if the bills that are listed constitute the unnamed providers' usual and customary charges for the service, or, instead, a discounted rate required by the agreements one or more of the providers may have had with health care insurers. While Ingenix requests that the CPT code billing data be accurate and complete, Gee conceded at trial that Ingenix remains "at the mercy" of its voluntary data contributors with respect to that result.
In Massachusetts, courts ask two questions to determine whether a hearsay document is admissible under the business record exception. Wingate v. Emery Air Freight Corp., 385 Mass. 402, 408 (1982) (Liacos, J., concurring). See also Azran v. Potter, 1996 Mass. App. Div. 202, 203. The first question is whether the writing itself qualifies as a business record under G.L.c. 233, § 78. Wingate, supra at 408. The second question, which arises only upon an affirmative answer to the first, is whether the material and information contained in the document is within the scope of the exception. Id.
The Ingenix documents in this case appear to satisfy the first requirement. Section 78 of G.L.c. 233 requires that an entry, writing or record be made "[1] in good faith [2] in the regular course of business [3] before the beginning of the civil . . . proceeding . . . and [4] that it was the regular course of such business to make such . . . record at the time of such act, transaction, occurrence or event or within a reasonable time thereafter." See also Beal Bank, SSB v. Eurich, 444 Mass. 813, 815 (2005). On the face of the record, the graphs and summaries introduced into evidence by Liberty were apparently compiled or made by Ingenix in good faith, in the regular course of its business, before the commencement of this action and in accordance with its regular business practice to make such records at or about the time it received the billing data in question. However, the Ingenix statistical analyses are derivative only, and the record indicates that the Ingenix raw data itself, at least as explained by Gee, lacks the requisite indicia of reliability to be admissible. As such, they do not fall within the broadened business record exception for scientific studies containing primarily factual data. See Vassallo v. Baxter Healthcare Corp., 428 Mass. 1, 18-19 (1998).
Although the data, graphs, and summaries qualify as Ingenix's business records, the records are not admissible to establish the truth of the underlying hearsay they contain. While § 78 makes clear that "personal knowledge by the [preparer] is a matter affecting the weight (rather than the admissibility) of the record," Wingate, supra at 406, it does not follow that the preparer may rely on statements that are not themselves part of the regular course of the preparer's business.
The preparer's hearsay sources must carry the same indicia of reliability arising from regularity and business motives, that bring his own act of recording the information within the statutory exception. Thus, unless the statements on which the preparer relies fall within some other exception to the hearsay rule, the proponent must show that all persons in the chain of communication, from the observer to the preparer, reported the information as a matter of business duty or business routine (emphasis supplied).
Wingate, supra at 406. See also Ford Motor Credit Corp. v. United States Fid. Guar. Corp., 1994 Mass. App. Div. 46, 48. Liberty made no such showing that its underlying data from unidentified medical providers was submitted to Ingenix as a matter of normal business practice by its data volunteers. The opposite was, in fact, established. Ingenix conceded that those of its health insurer clients that elect to participate in its data program submit billings from unnamed providers on a purely voluntary basis. No business duty to report the data to Ingenix, much less a duty to report it accurately, exists. Ingenix remains "at the mercy" of its participating insurer clients.
Apart from any hearsay objection, a significant portion of the Ingenix data introduced by Liberty could have been excluded on the basis of relevance. The Davekos bills at issue were for the treatment he provided to DiThomas from May, 2003 through January, 2004. Liberty introduced Ingenix summaries, however, of old billing data from 2002.
Even assuming that the Ingenix materials were admissible under G.L.C. 233, § 78, it remains unclear to what extent they could be deemed probative of the G.L.c. 90, § 34M issue before the trial court, namely, the fair and reasonable charges for Davekos' initial examination and manual therapy services. The Ingenix analyses were supposed to permit the trial court to compare Davekos' professional rates with other health providers' bills for the same services in Davekos' geographical area. Gee conceded, however, that the Ingenix database does not include data on the total number of times a particular CPT code service was actually performed in a given area at any time, or what the actual dollar range of provider fees was for the service. Gee testified that Ingenix billing data is neither a "universal sample" of professional services and charges, nor even a random sample of what other professionals were charging for CPT code 99203 and 97140 at the time of Davekos' treatment of DiThomas. At best, the Ingenix database includes the bills of an unspecified number of medical providers who, within a specific period of time, happened to have billed only those health insurers that were not only Ingenix clients, but also Ingenix clients that elected to participate in its voluntary data contribution program. Thus, an Ingenix statistical conclusion as to the specific dollar amount of, for example, the 80th percentile of billings for manual therapy, was not evidence of the 80th percentile of actual charges for that service in Davekos' area. The conclusion is, instead, proof of nothing more than the amount of the 80th percentile of only those charges that happen to be included in Ingenix's particular database. As stated, it is unknown how many providers' bills for an initial examination or manual therapy may have been included in the Ingenix database at the time of DiThomas' treatment, or at any given point. Thus, the Ingenix materials introduced by Liberty did not constitute evidence of the dollar amount of the usual and customary charges in Davekos' area for the services he provided, much less proof of what could be considered fair and reasonable charges for those services.
Moreover, the Ingenix charts and spreadsheets admitted into evidence did not show actual bill amounts submitted to Ingenix for the CPT code services in question which the trial judge could have readily compared to Davekos' bills. While we decline to describe Ingenix's statistical methodology in detail, it is sufficient to note that Ingenix groups the bills for an individual CPT code service in a given area with billing data submitted for six to eight different CPT code services. The combined data is then divided by the "relative value" Ingenix has assigned to reach CPT code from its own proprietary "relative value scale." The conversion factors determined are then arrayed, and Ingenix relative values are applied again to determine, for example, the 80th percentile bill amount. The dollar amounts for any service necessarily vary depending on the relative value assigned. Thus, the bill amounts charted by Ingenix were simply dollar amounts resulting from its statistical extrapolations from whatever actual bills were included in its particular database. Gee conceded that Ingenix has never tested its results to determine if its statistical conclusions bear any relationship to the actual high, low, median or 80th percentile of actual marketplace CPT code service rates charged by health providers in any given area.
The Ingenix relative value scale is its own product and differs from the relative values assigned to CPT codes by, for example, the American Medical Association or Medicare.
Judgment for the defendant is vacated, and the case is returned to the Lawrence Division for a new trial.
So ordered.