Time To Complete the Surveillance Reports

Between June 2011 and June 2012 we evaluated 14 CERs. Table 3 indicates the CERs that were evaluated and the priority they received, the date of their release, and the date of when either the SCEPC or UOEPC sent the assessments to AHRQ. In addition, Table 3 presents the number of days to complete each of the 14 surveillance reports. The mean was 86 days and the median was 74 days, with the majority of reports being completed in 65 to 102 days.

Table 3

Fourteen completed CER topics.

The assessments of 14 CERs that were submitted by the SCEPC or UOEPC can be found on AHRQ's Web site (www.ahrq.gov) where the specific CER is located.

Assessment Findings

The characteristics of the 14 CERs we assessed and the corresponding surveillance assessment results are presented in Table 4. Briefly, the number of Key Questions (KQs) across the 14 CERs ranged from three (CER No.s 16, 19) to seven (CER Nos. 23, 26). The median number of included studies in the CERs was 107 (range: 14–436). The number of newly identified studies deemed relevant for inclusion in the CERs ranged from 0 to 33, with a median of 15 studies.

Table 4

Characteristics of 14 comparative effectiveness reviews and their updating surveillance assessments.

Of the 14 CERs, 4 (29 percent) were up to date in the 12 to 59 months following their original search date (CER No.s 22, 23, 25, 35). For the remaining 10 CERs (CER Nos. 13, 15, 16, 17, 18, 19, 20, 21, 26, 30), at least one conclusion within a KQ changed in status from “up-to-date” to “probably/possibly out of date” or “out of date.” In 4 of these 10 CERs (CER Nos. 13, 18, 19, 20), all conclusions within a KQ changed their status from “up-to-date” to “probably/possibly out of date” or “out of date.”

Of the 14 CERs, 2 (14.3 percent) were assigned to high, 3 (21.4 percent) were assigned to medium, and 9 (64.4 percent) were assigned to low priority for updating. Of the 6 CERs released prior to 2010 (meaning more than 18 months prior to start of the Surveillance Program), 2 were judged as being high priority, 2 were judged as being medium priority, and 2 were judged as being low priority for updating. Of the remaining 8, only one was judged as being medium priority for updating. All but 1 of the CERs released within the year prior to the start of the Surveillance Program were judged as being low priority.

The percentage of experts asked who actually responded was also noted. The response rate ranged from 20 percent to 100 percent, with a median of 34.5 percent.

None of the 14 CERs for which we performed the surveillance assessments had an FDA black box warning associated with an agent, device, or procedure that was a topic of the CER (the strongest FDA warning, which indicates a significant risk of a serious or even life-threatening adverse effect). Five CERs had safety communications, adverse effects, and/or label change alerts (CER Nos. 16, 18, 23, 25, 26), none of which was sufficient to impact the updating priority of those CERs. CER No. 16 had a total of six label change alerts and three drug safety communication alerts, CER No. 18 had three alerts, and 3 CERs (Nos. 23, 25, 26) had one alert each.