Summary

REPORT BY THE STATE AUDITOR OF CALIFORNIA A Review of the Reports Submitted By the Department of Corrections and the Department of the Youth Authority on the Early Intervention Program 93011 January 1994 93011, January 1994 California State Auditor Bureau of State Audits January 11, 1994 93011 The Governor of California President pro Tempore of the Senate Speaker of the Assembly State Capitol Sacramento, California 95814 Dear Governor and Legislative Leaders: Summary The Bureau of State Audits presents its review and evaluation of the reports submitted by the California Department of Corrections (CDC) and the California Youth Authority (CYA) on the accomplishments of the early intervention pilot program for workers' compensation injuries. The types of worker injuries that are most likely to benefit from early intervention are defined as those injuries resulting in 30 days or more of disability from work, or cases in which the injury was alleged to be the result of psychological stress or its physical manifestation. In response to Chapter 1233, Statutes of 1988, we obtained the reports prepared by the directors of the CDC and the CYA on the accomplishments of the early intervention pilot program. Complete copies of the CDC's and the CYA's reports are attached to this letter as Attachments A and B, respectively. Our report focuses on our review of the two departments' reports in terms of the accuracy of the data compiled, its completeness, and the reports' compliance with statutory mandates. This review was originally required of the Office of the Auditor General (OAG). However, the OAG was closed in December of 1992 due to budget reductions, but in May 1993 the Bureau of State Audits began operation pursuant to Government Code Section 8456.8 and assumed responsibility for this review. During our review of the reports submitted by the CDC and CYA, we noted the following conditions: Letter Report 93011 Page 2 January 11, 1994 Effects of Other Workers' Compensation Reforms Events other than the 1989 implementation of the early intervention program, such as the passage of major workers' compensation legislation in 1989 and 1993, are currently blurring the CDC's and the CYA's attempts to measure the effectiveness of early intervention. No Mechanism Exists To Effectively Collect Certain Data The CDC and the CYA were unable to report on certain data requested by the Legislature because no mechanism currently exits to effectively collect the data requested. The statute requires the reporting of the number of workplace injuries meeting early intervention criteria that are reported on a specific state form, the Cal-OSHA Log 200. However, the Cal-OSHA Log 200 is not designed to effectively identify injuries meeting the criteria for early intervention. Therefore, both the CDC and the CYA relied on information supplied by the State Compensation Insurance Fund (SCIF) as the more accurate source for this data. Some Data Not In Conformance With Law Some of the data presented by the CYA and the CDC does not completely conform to statutory requirements. Specifically, the CYA used a data source other than the one requested by statute in reporting on one data element and did not include certain cost data in reporting on another data element. In addition, the CDC used a survey approach in reporting one of its data elements instead of reporting on its entire workforce as required. Some Data Could Not Be Validated We were unable to validate the accuracy of some of the source data used by both departments in compiling their reports. Both the CDC and the CYA derived the cost information appearing in their respective reports from data included in a two-year study conducted by the SCIF. This study included data compiled from cases meeting the criteria for early intervention. However, because the SCIF staff gathered the cost information included in the study from 14 of its district offices on various occasions over a two-year period rather than as of a baseline date, it was not possible for us to validate this cost data. Letter Report 93011 Page 3 January 11, 1994 Some Data Was Inaccurate Some of the data reported by the two departments was not accurate. We found inaccuracies in some of the source data used by both the CDC and the CYA in compiling their reports. Furthermore, some of the data elements that both departments reported and stated came from a specific source did not agree with that source. The recent passage of major workers' compensation legislation is blurring the CDC's and the CYA's attempts to measure the effectiveness of early intervention. For this reason, the Legislature may wish to consider deferring further attempts to evaluate the accomplishments of the early intervention pilot program until sufficient time has elapsed to accumulate data unaffected by competing workers' compensation legislation. Background The Department of Corrections (CDC) and the Department of the Youth Authority (CYA) currently operate a four-year old pilot program which seeks to ensure that the parties involved in the two departments' workers' compensation programs are fully informed of available options and that decisions on compensation for injured employees are reached and implemented quickly. The goal of this program, known as the early intervention program, is to minimize the potential financial and personnel losses to the two departments by taking steps to return injured employees to work as soon as possible. The program also seeks to identify those employees who will not be able to return to their regular jobs. The early intervention program encourages those employees who cannot return to their normal duties to explore alternative job placement either within their respective departments or elsewhere in State service or, if they qualify, to seek the services offered by vocational rehabilitation. Chapter 1233, Statutes of 1988, requires the directors of the CDC and the CYA to report to the Auditor General by July 1, 1992, on the accomplishments of early intervention, including the following specific data elements for the calendar years 1987 through 1991: (a) The number of injuries reported on Cal-OSHA Log 200. (b) The number of days elapsed from the date of injury to the date that the workers' compensation benefits were provided to the employee. Letter Report 93011 Page 4 January 11, 1994 (c) The number of days that the employee was off work due to the injury. (d) The number of employees who returned to work. (e) The number of days from the date of injury to the date the employee was referred to vocational rehabilitation. (f) The early intervention counseling costs. (g) The number and costs of medical-legal consultations as described in Section 4620 of the Labor Code. (h) The total claim costs of closed injury cases. (i) The number of employees who received industrial disability retirement. (j) The costs for replacing industrially injured employees. (k) The recruitment and training costs for replacement personnel. (l) The costs of industrial disability retirement. In order to gauge the effectiveness of the early intervention effort this legislation requires that the directors of the CDC and the CYA compare data for each of the above elements to workers' compensation statistics covering the three calendar years preceding the December 31, 1989, implementation of the early intervention program. Furthermore, the statute requires that the data reported for elements (a) through (i) be compiled for every workers' compensation case that resulted in 30 or more days of disability from work, or cases in which the injury was alleged to be the result of psychological stress or the physical manifestations of stress. Finally, the statute directs the Auditor General to review and evaluate the information reported by the directors of the CDC and the CYA in terms of the accuracy of the data, its completeness, and compliance with the statutory mandates of the act. The Auditor General's review, along with the directors' reports, were required to be submitted to the Legislature by December 31, 1992. However, the OAG was closed in December 1992 due to budget reductions, but in May 1993 the Bureau Letter Report 93011 Page 5 January 11, 1994 of State Audits began operation pursuant to Government Code Section 8456.8 and assumed responsibility for this audit. Scope and The focus of this audit was to have the Bureau of State Audits review Methodology and evaluate the reports prepared by the CDC and the CYA on the accomplishments of early intervention in terms of the accuracy of the data compiled, its completeness, and compliance with statutory mandates. To evaluate the completeness of the data reported, and that the data reported met the requirements of the statute, we reviewed the reports submitted by the CDC and the CYA comparing the data elements reported to those mandated in the act. To evaluate the accuracy of the data compiled, we interviewed employees from both departments to determine the methodologies used in compiling the information contained in each report. However, the CDC did not document all the methodologies it used in deriving various figures included in its report. The person primarily responsible for collecting and analyzing the information included in the CDC report is no longer an employee and was unavailable for consultation in reconstructing those methodologies that were not documented. Consequently, we were unable to completely validate the accuracy of some of the information presented in the CDC report. In addition, we validated the reliability of the source data used by the departments in compiling each of its reports to verify the accuracy of the information presented. Both the CDC and the CYA primarily relied on data provided by a State Compensation Insurance Fund (SCIF) study as the source for several of the data elements presented in the two reports we reviewed. The SCIF study compiled data concerning cases alleged to have resulted from stress and all other claims where the injured worker received thirty or more days of compensation. The SCIF study took two years to complete and encompassed a six year time period between 1986 and 1991. We obtained a copy of the SCIF study and tested a sample of the data included in it. Specifically, we tested the SCIF files supporting 30 cases involving CDC employees and 37 cases involving CYA employees included in the SCIF study to determine whether the SCIF had properly included employee workers' compensation cases based on the criteria established Letter Report 93011 Page 6 January 11, 1994 for early intervention and whether the data in the study agreed with the applicable SCIF case files. We also tested 13 workers' compensation cases not included in the SCIF study to determine that they were properly excluded because they did not meet the established early intervention criteria. To further test the accuracy of the information provided in the CDC and CYA reports, we traced a sample of the data elements required to be reported to the source documentation the two departments used. We independently tabulated a sample of data elements appearing in the SCIF study and compared the results to the same data elements presented in the CDC and CYA reports to verify the accuracy of the data in the reports. We tested two data elements in the CDC report and three data elements in the CYA report for the 1987 and 1991 calendar years. In compiling the data for certain elements contained in their reports, both the CDC and the CYA relied on information obtained from other entities. For instance, the CDC obtained certain information from the Department of Personnel Administration and the Public Employees' Retirement System, while the CYA obtained certain information from its various institutions. We did not visit those entities to validate the reliability of the information supplied by them. Effects of Other Events other than the 1989 implementation of the early intervention Workers' program are currently blurring the CDC's and CYA's attempts to measure the effectiveness of early intervention. This is because Compensation factors other than those directly relating to early intervention have Reforms intervened to improve the effectiveness of the workers' compensation program overall. In fact, in its report, the CDC stated that in addition to the implementation of the early intervention program, the Margolin-Green Workers' Compensation Reform Act of 1989 and the 1990 addition of return-to-work coordinators at each institution have had a beneficial impact on the CDC's workers' compensation statistics included in its report. For example, the CDC stated that the decrease in average Letter Report 93011 Page 7 January 11, 1994 elapsed time from the date of injury until benefits are provided to injured workers may reflect the effect of the Workers' Compensation Reform Act and the addition of its return-to-work coordinators as well as the efforts of the early intervention program. Additionally, in a cover letter accompanying its study of cases meeting the criteria for early intervention, the SCIF cautioned against over-reliance on the data it provided in the study. In warning users of its study to exert caution in evaluating the results, the SCIF cited two issues. First, is the issue of the maturity level of the injury claim. The SCIF gave as an example the fact that a serious knee injury occurring in 1986 would have much higher incurred and paid costs than the same injury sustained in 1991. This is because the time elapsed for incurring additional medical treatment, compensation, vocational rehabilitation, and litigation costs is so much greater that it makes comparing 1986 injury cases with 1991 injury cases very difficult. Second is the issue that, right in the middle of the six year period covered in the SCIF's study, a major reform package was implemented, making the comparison of injury cases occurring before and after the reform legislation misleading. In our view, these are reasonable admonitions that have only been heightened by the recent adoption of another major workers' compensation reform package in 1993. For these reasons, we recommend that the Legislature defer further attempts to evaluate the accomplishments of the early intervention pilot program until sufficient time has elapsed to accumulate meaningful data unaffected by competing workers' compensation legislation. Chapter 1233, Statutes of 1988 requires specific data be compiled and No Mechanism reported concerning the accomplishments of the early intervention Currently Exists program. For example, the statute requires the presentation of the To Effectively number of workplace injuries reported on a specific form, the Collect and Cal-OSHA Log 200. The statute further requires the data be compiled Report Certain for every workers' compensation case that has resulted in 30 days or Data Requested more of disability from work, or in which the injury was alleged to be the result of psychological stress or the physical manifestation of stress. By the These two conditions are the criterion for which an injury case qualifies Legislature for the early intervention program. Although provisions in the statute require the reporting of the number of injuries recorded in the Cal-OSHA Log 200 in order to evaluate the Letter Report 93011 Page 8 January 11, 1994 early intervention program, the data recorded in the Cal-OSHA Log 200 does not specifically distinguish between early intervention and non-early intervention injury cases. Further, the data compiled in the Cal-OSHA Log 200 may not accurately reflect the number of stress-related injuries because in many instances, a stress-related injury will not be recorded in the log when it is a secondary injury and is not the initial injury reported. According to the CDC, to determine which cases recorded in the log might have been stress related injuries, or those which resulted in 30 days or more of disability, it would have been necessary to first read the description of each reported injury and then manually count each case meeting the criteria from 21 different Cal-OSHA Log 200s. Neither of the departments has in place a procedure for extracting this specific information. As a result, the CDC and the CYA relied on a study that the State Compensation Insurance Fund (SCIF) completed in April, 1992. In this study the SCIF compiled all claims meeting the early intervention criteria to provide a more accurate picture of the early intervention program. Some Data In our review of the CDC and CYA reports, we noted that certain data Reported Does reported did not conform to what was required by statute. Specifically, the CYA reported on two data elements that did not Not Completely conform to statutory requirements and the CDC's data did not conform Conform to for one element. Statutory Requirements Provisions in Chapter 1233, Statutes of 1988 requires the reporting of the number of injuries recorded in the Cal-OSHA Log 200. However, the total injury figures reported by the CYA (See Attachment B, page B-3) were not derived from the Cal-OSHA Log 200. Instead, the CYA used the SCIF 3067 First Report of Accident/Injury form to compile the data presented. Using this source, the CYA presented complete data regarding the annual number of injuries meeting the early intervention criteria. As noted in the previous section, the Cal-OSHA Log 200 is not designed to effectively distinguish between early intervention and non-early intervention injury cases. Consequently, the CYA used the SCIF form because this form better reflected early intervention types of injuries. Statutory provisions also require the reporting of the recruitment and training costs incurred in replacing personnel with reported injuries. The CYA derived the annual figures it reported (See Attachment B, Letter Report 93011 Page 9 January 11, 1994 page B-6) by first estimating the cost of a five-week training course at the CYA academy. This cost was multiplied by the number of industrial disability retirees for the year to estimate the cost for replacement personnel. However, the CYA understated the estimated costs associated with the recruitment and training of replacement personnel because it did not include an estimate for recruitment costs. In addition, statutory provisions require the reporting of the total number of employees who returned to work subsequent to early intervention types of injuries. However, the CDC relied on a survey of three of its institutions to derive these figures (See Attachment A, page A-10). Because the CDC only gathered data from three institutions, the figures presented in its report only reflect a sample of the employee population, rather than aggregate data based on the CDC's entire workforce. According to the CDC, the sample survey approach it took was necessary to generate the required information since this information is not easily gathered. The SCIF does not collect this data and the department currently has no automated means of gathering this information. In order for the CDC to have collected the information for its entire employee population it would have needed to manually search thousands of employee case files from its 21 institutions. Therefore, the department relied on a representative survey from a sample of its institutions to provide the information for this data element. Nevertheless, by reporting on a sample of the workforce rather than the entire population, the CDC's data did not conform with statutory requirements. Some Data Could We could not validate cost data used by the CDC and the CYA in Not Be Validated compiling certain figures included in their respective reports because we could not duplicate the methodology the SCIF used in developing for Accuracy the data included in its study. The cost information included in the SCIF study, and used by both departments for various elements in its reports, is based on a survey the SCIF staff conducted during visits to 14 SCIF district offices over a two-year period. The SCIF staff collected the data for all cost elements as of the date of their visit to a particular district office rather than as of a uniform baseline date. Because the SCIF staff did not record the various dates of their visits and the elements of information collected on such dates, it was not possible for us to re-create any of the cost data included in the SCIF study and subsequently used by both the CDC and the CYA in preparing their respective reports. Letter Report 93011 Page 10 January 11, 1994 Some Data We found some inaccurate data included in the SCIF study as well as Reported Was the early intervention reports submitted by the CDC and the CYA. Not Accurate As stated earlier in this report, both the CDC and the CYA relied on data included in the SCIF study when compiling certain data elements included in each of their reports. Therefore, we traced certain information shown in the SCIF study to supporting documents found in employee case files to determine if the data was accurate. We found that 3 of 30 (10 percent) CDC cases included in the SCIF study that we tested contained inaccuracies. In two cases, the date of benefit shown in the SCIF study did not agree with the supporting documentation. In the third case, the employee should not have been included in the SCIF study because the injury did not meet the criteria for an early intervention case according to the source documents we reviewed. In addition, 2 of 37 (5.4 percent) CYA cases included in the SCIF study that we tested contained inaccuracies. In both cases, either the date of injury or the date of benefit did not agree with supporting documents. Furthermore, we found instances when the data elements presented in the CDC and CYA reports did not agree with the same data elements shown in the SCIF study, even though both the CDC and the CYA stated that the SCIF study was the source for those specific elements. We tested some of those elements to determine that the CDC's and the CYA's reports accurately reflected the SCIF data. For example, for the CDC, we tested the accuracy of the elements concerning the number of days elapsed between the date of injury and the date that benefits were provided and the number of days elapsed between the date of injury and the date the employee was referred to vocational rehabilitation for injuries meeting the early intervention criteria. In testing these two data elements as reported by the CDC for two different years, 1987 and 1991, we found that the totals reported differed from the corresponding SCIF data by 5 percent or more in two cases. For instance, the CDC reported that in 1987, a cumulative total of 85,141 days elapsed between the date of injury and the date benefits were provided (See Attachment A, page A-7). However, according to the SCIF data, we calculated that during 1987, a total of 94,769 cumulative days elapsed between the date of injury and the date benefits were provided, a difference in cumulative elapsed days of more than 10 percent. Letter Report 93011 Page 11 January 11, 1994 We also found that the CDC inaccurately reported the number of days between the date of injury and the date an employee was referred to vocational rehabilitation. The CDC reported that in 1987, a total of 108,659 cumulative days elapsed between the two dates (See Attachment A, page A-11). Using the SCIF's data, we calculated that during 1987, a total of 118,814 cumulative days elapsed between the two dates, a difference of more than 8.5 percent. Furthermore, in testing three of the elements reported by the CYA that used the SCIF data as the source for the years 1987 and 1991, we found that one of the totals did not reconcile with the SCIF data. The elements we tested were the early intervention counseling costs, total claim costs after finalization, and the total elapsed time between the date of injury and the date that benefits were provided. The CYA reported that in 1991, a total of 14,835 cumulative days elapsed between the date of injury and the date that benefits were provided (See Attachment B, page B-3). However, using the SCIF's data, we calculated that during 1991, a total of 9,013 cumulative days elapsed between the date of injury and the date benefits were provided, a difference of 65 percent. Conclusion Statutory provisions require the reporting of the number of injuries reported in the Cal-OSHA Log 200 in order to evaluate the early intervention program. However, the data recorded in the Cal-OSHA Log 200 does not specifically distinguish between early intervention and non-early intervention injury cases. Further, the data compiled in the Cal-OSHA Log 200 may not accurately reflect the number of stress-related injuries because in many instances, a stress-related injury will not be recorded in the log when it is a secondary injury and not the initial injury reported. Furthermore, in our review of the CDC and CYA reports, we noted that certain data reported did not conform to what was required by statute. Specifically, the CYA reported on two data elements that did not conform to statutory requirements and the CDC's data did not conform for one element. In addition, we could not validate some of the cost data used by the CDC and the CYA in compiling certain figures included in their respective reports because we could not duplicate the methodology the SCIF used in developing the data included in its study. Letter Report 93011 Page 12 January 11, 1994 We also found some inaccurate data included in the SCIF study as well as the early intervention reports submitted by the CDC and the CYA. As stated earlier in this report, both the CDC and the CYA relied on data included in the SCIF study when compiling certain data elements included in each of their reports. Therefore, any inaccuracies included in SCIF's source data for those elements would also be reflected in the corresponding data elements that the CDC and the CYA reported that relied on the SCIF study as the source of the data. Finally, we found instances when the data elements presented in the CDC and CYA reports did not agree with the same data elements shown in the SCIF study, even though both the CDC and the CYA stated that the SCIF study was the source for those specific elements. In testing two data elements as reported by the CDC for two different years, 1987 and 1991, we found that the totals reported differed from the corresponding SCIF data by 5 percent or more on two occasions. In addition, in testing three data elements reported by the CYA that used the SCIF data as the source for the years 1987 and 1991, we found that one of the totals reported for 1991 varied from the total shown in the SCIF data by 65 percent. Events other than the 1989 implementation of the early intervention program are currently blurring the CDC's and CYA's attempts to measure the effectiveness of early intervention. This is because factors other than those directly relating to the early intervention program have intervened to improve the effectiveness of the workers' compensation program in general. Both the CDC and the SCIF recommended using caution in interpreting the early intervention program data, in part because of the effect that other legislation enacted since 1989 and aimed at workers' compensation reform might have had in making such data misleading. In our view, these are reasonable admonitions that have only been heightened by the recent adoption of another major workers' compensation reform package in 1993. For these reasons, we recommend that the Legislature defer further attempts to evaluate the accomplishments of the early intervention pilot program until sufficient time has elapsed to accumulate meaningful data unaffected by competing workers' compensation legislation. We conducted this review under the authority vested in the State Auditor by Section 8543 et seq. of the California Government Code Letter Report 93011 Page 13 January 11, 1994 and according to generally accepted governmental auditing standards. We limited our review to those areas specified in the audit scope of this letter report. Respectfully submitted, KURT R. SJOBERG State Auditor Attachments Staff: Steve M. Hendrickson, Audit Principal Douglas Cordiner, Audit Supervisor Arn Gittleman, Staff Auditor Paul Navarro, Staff Auditor The response from the Youth and Adult Correctional Agency is attached to this report.