People v. Hard, 2014 Colo. App. LEXIS 1676 (Colo. Ct. App. Oct. 9, 2014): 

First, although the People argue that information from Drugs.com is reliable because Trooper Hancey testified that Drugs.com is "nationally recognized" and that law enforcement officials rely on it when identifying unknown pills, the prosecution offered no foundation for Trooper Hancey's assessment that Drugs.com is nationally recognized. And, in any event, Trooper Hancey's testimony alone was insufficient to establish the website's reliability. See In re Anthony M.S., No. 2010AP1669, 2011 WL 2228538, at *2 (Wis. Ct. App. June 9, 2011) (unpublished opinion) (holding that police officer's testimony that he had used Drugs.com previously in his professional capacity was insufficient to prove the reliability of Drugs.com). Indeed, under the People's proposed interpretation of CRE 803(17), hearsay evidence from nearly [**13]  any Internet database would be admissible, so long as a law enforcement official testifies that the database is used and relied on generally by law enforcement officials. That interpretation is simply too broad. Cf. Pers. Audio, LLC v. CBS Corp., No. 2:13-CV-270-JRGRSP, 2014 WL 1202698, at *4-5 (E.D. Tex. Mar. 20, 2014) (unpublished opinion) (holding a LinkedIn profile inadmissible under Fed. R. Evid. 803(17), and noting that under CBS's interpretation of Rule 803(17), any online information would necessarily be admissible if a journalist testifies that he uses that resource to gather information).

 [*P32]  Second, courts in several other jurisdictions have held that mere visual identification of prescription drugs -- that is, identification of a drug by comparing its physical attributes to a reference material such as Drugs.com -- is not a sufficiently reliable method of proof in a criminal trial. See, e.g., People v. Mocaby, 882 N.E.2d 1162, 1166-68 (Ill. App. Ct. 2008) (holding that evidence was insufficient to show that pills were controlled substances where a forensic scientist had identified the pills by comparing them to pictures in a book but had not performed a chemical analysis); State v. Ward, 694 S.E.2d 738, 740, 743-47 (N.C. 2010) (holding that expert witness testimony establishing that a substance is a controlled substance "must be based on a scientifically valid chemical analysis and [**14]  not mere visual inspection"); cf. State v. Stank, 708 N.W.2d 43, 54-55 (Wis. Ct. App. 2005) (holding that evidence was sufficient to support the defendant's conviction for possession with the intent to distribute a controlled substance where the detective conducted both a preliminary identification of the pills using the Physicians' Desk Reference and a confirmatory chemical analysis of the pills).⁶

6   Some courts have noted in this context that by criminalizing possession of counterfeit controlled substances, legislatures have recognized expressly that counterfeiters can make substances appear, by markings and physical character, to be controlled substances. E.g., State v. Ward, 694 S.E.2d 738, 744-45 (N.C. 2010). The Colorado General Assembly is one such legislature. It has criminalized possession of imitation or counterfeit controlled substances, §§ 18-18-419 to -424, C.R.S. 2014, and in so doing has defined an "imitation controlled substance" as a substance that is not a controlled substance but appears to be by, among other things, "color, shape, size, and markings . . . ." § 18-18-420(3).

 [*P33]  Third, although the People argue that information from Drugs.com is reliable because that website compiles and publishes material from reliable sources, they provide no support for that proposition. And, indeed, courts in several jurisdictions have [**15]  questioned the reliability of several of Drugs.com's sources -- including the Physicians' Desk Reference and Micromedex. See, e.g., Garvey v. O'Donoghue, 530 A.2d 1141, 1145 (D.C. 1987) (holding the Physicians' Desk Reference inadmissible under Fed. R. Evid. 803(17) because the publication contains not only factual statements, but also "directions, opinions, suggestions, and recommendations"); Ward, 694 S.E.2d at 745-47 (holding that an expert witness's reliance on Micromedex was not a sufficiently reliable method of proof for a criminal prosecution); Kahanek v. Rogers, 12 S.W.3d 501, 504 (Tex. App. 1999) (holding the Physicians' Desk Reference inadmissible under Texas's market reports exception because the publication "goes beyond objective information to items on which learned professionals could disagree in good faith"); see also In re Richardson-Merrell, Inc. Bendectin Prods. Liab. Litig., 624 F. Supp. 1212, 1232 (S.D. Ohio 1985) (holding without explanation that excerpts from the Physicians' Desk Reference did not fall within the commercial publications exception of Fed. R. Evid. 803(17)), aff'd, 857 F.2d 290 (6th Cir. 1988).

 [*P34]  Fourth, we find unpersuasive the People's argument that information from Drugs.com is reliable because its "mission statement says that it is the "largest, most widely visited, independent medicine information website available on the Internet,' and it provides 'independent, objective, comprehensive, and up-to-date information . . . for both consumers and [**16]  healthcare professionals.'" That mission statement was not included in the record; thus, we may not consider it. See Fendley v. People, 107 P.3d 1122, 1125 (Colo. App. 2004) ("We are limited to the record presented and may consider only arguments and assertions supported by the evidence in the record."). But even if we assume that the mission statement is properly before us, a website's own self-serving mission statement hardly constitutes objective proof of the website's reliability. And the mission statement is offset by the website's disclaimer of any guarantee as to the accuracy of the information on the site.

 [*P35]  Finally, we are not persuaded by the district court's rationale that information from Drugs.com is sufficiently reliable because numerous courts have cited Drugs.com in published judicial opinions. Nearly all of the decisions we have found that cite Drugs.com do so in the context of providing general information about a particular drug (e.g., its intended uses or side effects). See, e.g., Gentry v. Comm'r of Soc. Sec., 741 F.3d 708, 715 (6th Cir. 2014) (side effects); Blackmon v. Astrue, 730 F. Supp. 2d 867, 870-73 (N.D. Ill. 2010) (intended uses). We have not found a single decision approving evidence from Drugs.com as proof of the identity of an unknown drug.

 [*P36]  For these reasons, we conclude that the People failed to show that information from Drugs.com [**17]  was sufficiently necessary and reliable to be admissible under CRE 803(17). The district court therefore erred in admitting the hearsay evidence obtained from that website.⁷