Use the links below to skip to the chapter you wish to view:
CALIFORNIA COULD IMPROVE ITS DETECTION OF WORKERS’ COMPENSATION FRAUD
As we indicate in the Introduction, a key step in combatting workers’ compensation fraud is its detection. Nonetheless, we identified weaknesses in the processes CDI, Industrial Relations, and insurers use to detect fraud. For example, CDI does not currently take advantage of a key indicator that could help it identify and audit insurers that may not be adequately reporting potential fraud. Specifically, although state law requires insurers to investigate suspected fraud and refer to CDI and district attorneys’ offices those claims that show reasonable evidence of fraud, insurers vary significantly in the number of referrals they submit. Of the 21 insurers that we examined, eight submitted one or fewer referrals per $10 million in earned premiums for at least one of the two years we examined.6 Low referral rates could indicate that insurers are not referring suspected workers’ compensation fraud, leaving this potential fraud uninvestigated. However, CDI does not use the rate of insurers’ submissions of referrals as a tool to assess risk when identifying those insurers it will audit.
In addition, Industrial Relations has not fully documented its procedures for implementing a critical tool for combatting workers’ compensation fraud by providers. Provider fraud cases can continue unnoticed for years, and a single case can cost an insurer millions of dollars. To more quickly uncover such fraud, Industrial Relations is in the process of implementing data analytics, which will allow it to examine large volumes of data. However, Industrial Relations is still in the beginning stages of its implementation and has not yet fully documented how it will identify potential fraud and use the results of such examinations. Because data analytics has the potential for high rates of return, Industrial Relations should fully document its data analytics efforts as soon as possible.
Finally, California can further improve its fraud detection efforts related to workers’ compensation by requiring insurers to periodically issue explanation of benefits statements (EOB statements) to injured employees. These statements itemize the types of services rendered, the dates employees received the services, and service fees paid on their behalf. EOB statements provide injured employees with the opportunity to review the services for which providers bill and to identify potentially fraudulent charges.
Some Insurers Are Significantly Less Likely Than Others to Report Suspected Workers’ Compensation Fraud
Despite a requirement that insurers refer to CDI and district attorneys’ offices those claims that show reasonable evidence of fraud, the number of referrals insurers submit varies significantly, leading us to question whether some insurers are reporting all suspected fraud. By law, every insurer must have or use a special investigative unit to pursue instances of suspected fraud. State law further requires that within 60 days of having a reasonable belief that a claim may be fraudulent, an insurer must submit a referral to both CDI and the district attorney’s office where the loss occurred. We expected that those insurers with relatively higher amounts of earned premiums—indicating that they likely process more workers’ compensation claims—would also have generally higher frequencies of referring suspected fraudulent claims to CDI and the district attorneys’ offices. However, our review found that referral rates varied significantly.
According to the 2016 Annual Report of the Commissioner, referrals are CDI’s primary source of leads for workers’ compensation fraud investigations. CDI can receive referrals from anyone: insurers, employers, employees, medical providers, and the general public. Referrals can be for any type of workers’ compensation fraud: employee, employer, medical provider, and others. Referrals most often involve employee fraud; the numbers of employer and provider fraud referrals are also considerable. Table 5 provides more information regarding referrals for fiscal years 2013–14 through 2016–17.
In our April 2004 audit report Workers’ Compensation Fraud: Detection and Prevention Efforts Are Poorly Planned and Lack Accountability , Report 2002‑018, we concluded that some insurers appeared to underreport suspected workers’ compensation fraud while others appeared to regularly refer suspected fraudulent claims. We found that although five of the 23 insurers we reviewed during that audit referred more than one claim per $1 million in earned premiums, some of the remaining 18 insurers might have been failing to fulfill their responsibilities to refer suspected fraud, including some that did not submit a single referral during one or more of the years in our audit period. We also identified barriers that might prevent insurers from consistently referring suspected fraud and recommended that CDI take steps to address these barriers.
CDI Receives Thousands of Fraud Referrals Each Year
Fiscal Years 2013–14 Through 2016–17
|Fraud Type||Referrals||Percent of Total||Referrals||Percent of Total||Referrals||Percent of Total||Referrals||Percent of Total|
|Medical Provider *||236||4.1||240||4.0||253||4.7||228||5.5|
Source: California State Auditor’s analysis of data obtained from CDI’s Fraud Integrated Database System.
Note: Due to rounding, the percent columns may not equal exactly 100 percent when added.
* Medical Provider also includes referrals related to pharmacies.
During our current audit, we continued to see significant variation in the rates at which insurers submitted referrals, leading us to believe that some insurers may be underreporting fraud. As Figure 6 shows, we calculated the referral rates for 21 insurers that each had more than $150 million in earned premiums for 2015 and 2016. In 2016 these insurers collectively earned almost $8 billion in premiums and represented 62 percent of workers’ compensation earned premiums in California. For the two years we reviewed, the insurers’ referral rates ranged from a high of 11 referrals per $10 million in earned premiums to a low of no referrals, while the actual number of referrals ranged from more than 350 to zero. Of the 21 insurers, eight submitted one or fewer referrals per $10 million in earned premiums in at least one of the two years we examined. Because of the high amount of estimated fraud in the workers’ compensation system, the rates we observed for these eight insurers seemed low and could indicate that they are not referring suspected workers’ compensation fraud. These eight insurers collectively had $3.9 billion in earned premiums in 2016, which represented 31 percent of workers’ compensation premiums in California.
Sources: California State Auditor’s analysis of referral data from CDI’s case management system and of premium data from CDI’s market share reports.
We believe that when an insurer with over $150 million in annual earned workers’ compensation premiums submits few or no referrals during a year, it should at least prompt CDI to make an inquiry. Because insurer referrals are the primary method the State uses to initiate investigations into suspected fraudulent workers’ compensation claims, we asked CDI whether it was aware of the relatively low referral rates by these eight insurers, whether it knew why the referral rates were so low, and whether it was reasonable for the referral rates to be so low. The manager who oversees CDI’s compliance office (compliance manager) indicated that a low number of referrals by itself does not mean that the insurer is not detecting, investigating, and then referring suspected fraud to CDI. Instead, a low number of referrals may be attributable to other factors, such as the insurer having few California claims. The compliance manager further stated that the reasons for low referral rates may vary based on lines of business and specialized insurance products. However, when we interviewed a senior executive with an insurer, he stated that the cost of the special investigative units is a factor affecting the quantity of referrals and that certain insurers invest only enough to comply with regulations, while others are vigorous in fighting fraud. The compliance manager agreed that insurers’ attitudes toward fraud may play a role, stating that some are committed to combatting it while others accept it as a cost of doing business.
Other entities have called attention to certain insurers’ actions by publishing reports on the insurers’ performances. For example, Texas law requires that the Texas Department of Insurance publish a periodic report card that evaluates specified workers’ compensation health care provider networks on the cost and the quality of medical care provided to injured workers. Similarly, California law requires Industrial Relations to publish the result of its Profile Audit Review in an annual report that lists the insurers it audited in that year, identifies how each scored, and ranks how each performed based on the audit. We believe a comparable public report that rates insurers’ antifraud efforts could motivate insurers with minimal compliance to improve and could also better inform consumers about insurers’ fraud‑fighting efforts—or lack thereof.
Although California regulations require insurers to submit annual reports to CDI regarding the performance of their special investigative units, CDI does not currently screen these reports for low referral rates relative to other insurers. The regulations require these reports to include the number of claims the insurers processed, the number of claims they referred to their special investigative units, and the number of incidents of suspected insurance fraud they reported to CDI and district attorneys’ offices for the past calendar year. The annual reports also provide overviews of the special investigative units’ organizational arrangements of antifraud personnel; descriptions of the units’ staff expertise and how that expertise meets CDI’s requirements; descriptions of the units’ methods of investigation and written procedures for detecting, investigating, and reporting suspected fraud; and the units’ plan for initial and ongoing training for integral antifraud personnel. CDI’s compliance office’s procedures indicate that staff analyze these reports for discrepancies and noncompliance issues. However, the compliance manager stated that CDI does not evaluate the reports for low referral rates relative to other insurers.
In addition, CDI periodically examines insurers’ special investigations units; however, it does not ensure that it selects large insurers with low referral numbers when planning its audits. According to CDI’s compliance review program, the compliance office is responsible for reviewing over 1,100 insurers and their special investigative units. CDI has staffed its compliance office with four to six auditors for the last 10 years. According to the compliance manager, CDI’s management decides which insurers to audit. He also indicated that because of limited staff, CDI uses a risk‑based approach when selecting insurers for review. Some of the risk factors CDI considers are the length of time since an insurer’s last audit, the insurer’s market size, any complaints CDI has received, and information in the insurer’s annual report for its special investigative unit. The compliance manager supplied a schedule showing that the compliance office has averaged roughly 12 audits per year for the last four years. Because of the large number of special investigative units and the small number of CDI audit staff, the compliance office could benefit from using an audit selection criterion that compares large insurers’ referral rates with those of their peers. Since fiscal year 2013–14, CDI’s compliance office has examined four of the eight insurers we selected that had a rate of one or fewer referrals in either 2015 or 2016. Following CDI’s disclosure of the results of these examinations, two of the insurers increased the number of referrals they made, while the other two did not meaningfully change the numbers of their referrals.
Industrial Relations Has Not Yet Fully Documented the Procedures for Its Provider Fraud Data Analytics Efforts
Provider fraud cases can continue unnoticed for years, and a single case can cost insurers millions of dollars. To better fight this type of fraud, the State is in the process of implementing data analytics to predict which providers may be committing workers’ compensation fraud. Two consultants, which CDI and Industrial Relations commissioned, specifically recommended in 2008 and again in 2017 that Industrial Relations explore and implement data analytics. According to one of the consultants, data analytics is a rapidly developing field of information science that involves intensive examination of large volumes of data to discover deeper insights, make predictions, and generate recommendations.
Industrial Relations’ Anti‑Fraud Unit recently began using data analytics both to support new laws enacted in 2016 that related to workers’ compensation liens and to uncover previously unidentified provider fraud. Its efforts related to uncovering new provider fraud are still in the nascent stages. According to Industrial Relations’ documentation, the Anti‑Fraud Unit has a team responsible for implementing data analytics. The Anti‑Fraud Unit performs both descriptive and predictive analytics. Descriptive analytics is a tool that can help identify patterns of past behavior among providers—in other words, what happened—while predictive analytics is a tool that can help identify possible patterns that indicate provider fraud—in other words, what could happen. Industrial Relations told us that as of October 2017, it had provided two lists of potentially fraudulent providers that its data analytics effort had identified to CDI. CDI is then responsible for matching these potentially fraudulent providers to its current investigations. The assistant chief of CDI’s Fraud Division told us that CDI already had investigations underway for most providers on the first list and that CDI had forwarded this list to its regional offices for their review. She also stated that although CDI was still checking the second list against its internal information, it was confident that the second list would uncover previously unknown provider fraud.
Despite the potential value of the lists it has already produced, we believe that Industrial Relations could do more to ensure the success of its data analytics efforts. Specifically, it has not yet fully documented the procedures for these efforts, resulting in a lack of specificity about how it intends to move forward. For example, when we requested a plan for its data analytics efforts, Industrial Relations provided only draft processes for the Anti‑Fraud Unit that did not include any specifics related to data analytics; an undated, one‑page draft schematic of the feedback loop for when data analytics identify suspicious activity; and a list of seven indicators that other providers convicted of fraud exhibited. Industrial Relations later provided a final protocol manual for its Anti‑Fraud Unit and its activities. However, this protocol manual included only limited information about data analytics and did not explain how Industrial Relations intends to refine its data analytics through discussions with CDI or include timelines for that refinement.
Data analytics is a promising tool with a potentially high rate of return. Moreover, Industrial Relations’ effective implementation of data analytics is critical because provider fraud imposes serious financial costs on consumers, businesses, and government. Thus, Industrial Relations should better document its procedures for its data analytics efforts as soon as possible to better ensure the success of those efforts. Its procedures should include a description of how it will adjust its protocols as necessary, depending on the results of its efforts.
By Issuing EOB Statements, Insurers Could Increase Detection of Workers’ Compensation Fraud
California can further improve its fraud detection efforts by requiring insurers to periodically issue EOB statements to injured employees after the employees receive workers’ compensation‑related services. By failing to provide such statements, insurers are missing opportunities to involve injured employees in their antifraud efforts. EOB statements itemize the types of services providers rendered, the dates the patients received the services, and service fees the insurers paid on the patients’ behalf. Thus, EOB statements would provide injured employees with the opportunity to review the services for which providers have billed insurers and identify potentially fraudulent charges. Nonetheless, as of October 2017, California did not require insurers that cover workers’ compensation to send or otherwise make EOB statements available to injured employees.
By requiring insurers to periodically provide EOB statements to injured employees, California could enlist those employees in its battle against workers’ compensation fraud. As the Los Angeles County District Attorney’s Office (LA District Attorney) stated, a number of vulnerabilities in the workers’ compensation system are readily identifiable, including the lack of review by the patients who purportedly received the services, equipment, or medications for which providers submit claims. This lack of review creates the potential for serial billing, in which providers bill multiple insurance carriers for the same services. The LA District Attorney concluded that instances of serial billing are likely to continue unless red flags are identified that might lead to greater scrutiny of or even denial of the billed charges. The periodic provision of EOB statements would allow injured employees to provide this type of red flag.
Certain government agencies and some insurers outside of the workers’ compensation program already use EOB statements to help fight fraud. For instance, the U.S. Centers for Medicare and Medicaid Services provides quarterly EOB statements to beneficiaries under its Original Medicare programs at least in part to fight health care fraud.7 According to health care antifraud literature, Medicare beneficiaries have discovered fraud through reviewing their EOB statements. The concerns that the beneficiaries raised have resulted in prosecutions, convictions, and the recovery of funds. Similarly, California law requires insurers providing disability insurance, including those providing health insurance, to provide EOB statements to people submitting insurance claims (claimants).
In addition, some California employers that self‑insure see the value of providing EOB statements to their injured employees. Disneyland Resort (Disney) stated in a presentation on workers’ compensation that EOB statements can help uncover provider billing mistakes, billing mischief, or fraud. The manager of workers’ compensation for Disney (Disney manager) stated that although some people believe that no one reads or understands EOB statements and that they cost too much, these are misconceptions. The Disney manager asserted that the expense of EOB statements is worthwhile because they promote transparency, awareness, communication, and goodwill. Further, Disney spends only 50 cents per EOB statement, and the Disney manager stated that, to ensure costs remain low, it sends an EOB statement only when there was a billing payment in the prior month.
According to key players within the health care system, not providing EOB statements to patients gives providers who want to commit fraud an easy means of doing so. According to the Ponemon Institute’s 2015 Fifth Annual Study on Medical Identity Theft, in 2014 the third most common method through which victims discovered medical identity theft—the use of an individual’s identity either to fraudulently receive medical services or drugs or to commit fraudulent billing—was through detecting errors in their EOB statements.8 Further, the LA District Attorney stated within its grant application for the Workers’ Compensation Insurance Fraud Program that it actively advocates the value that quarterly EOB statements provide to injured employees. The LA District Attorney also stated that outreach and training to inform the public about the workers’ compensation system and the information available through EOB statements are essential.
Although the Legislature can require insurers to provide EOB statements, both CDI and Industrial Relations have expressed concerns about the statements’ usefulness and cost‑effectiveness. Specifically, CDI stated that while providing EOB statements to injured employees could help reduce billing for services that were not provided, the impact would depend on the injured employees’ interest in reviewing the EOB statements. It believes that this interest might be limited by the fact that employees are not liable for any of the costs related to services they receive through workers’ compensation. However, CDI would support further consideration of whether EOB statements would be a cost‑effective means of identifying workers’ compensation fraud and whether they would be duplicative of existing disclosures. Further, CDI’s website states that it encourages consumers to review their EOB statements for other lines of insurance and to report billing for the following: treatment that was not provided, medical tests or evaluations that were not conducted, medical supplies that were not provided, office visits that never occurred, cancellation charges for office visits that were not scheduled, and pharmaceuticals that were never received.
In contrast, Industrial Relations’ director (director) stated that her department examined the viability of requiring insurers to send EOB statements and concluded that it would be more effective to impose notice requirements, such as EOB statements, on those medical providers who may treat injured employees without the employers’ or insurers’ knowledge. She explained that this type of medical care may be more vulnerable to fraud and abuse and that it would help the workers’ compensation system to require providers rendering treatment in this manner to promptly issue notices to all parties (including employees, employers, insurers, and Industrial Relations). The director further stated that existing controls—including fee schedules, independent medical and bill reviews, and utilization reviews—provide effective controls for care provided within the system under accepted claims, and that such care would not be improved by the issuance of EOB statements. The director also stated that the cost of EOB statements would place a burden on those operating in compliance with the system and might lead to an increase in premiums. Industrial Relations, however, did not provide evidence sufficient to support the director’s statements. Further, we question whether providers that are more apt to commit fraud would issue accurate and complete EOB statements to all parties.
Despite CDI’s and Industrial Relations’ concerns, we believe that EOB statements could be an effective tool to help fight provider fraud in the workers’ compensation system and that the benefits of these statements would likely outweigh any perceived drawbacks. For instance, insurers may believe that EOB statements are prohibitively expensive. However, fraud already harms employers by contributing to the increasingly high cost of workers’ compensation, and the amount of chargeable provider fraud—as we show in Table 1 in the Introduction—has grown from about $130 million in fiscal year 2013–14 to over $812 million in fiscal year 2015–16, an increase of 525 percent. Further, insurers could keep their costs down by providing EOB statements only periodically, such as once a quarter, and by consolidating all claims in that period. Insurers may also argue that EOB statements are confusing and that injured employees will consequently ignore them. To address this, the literature we examined suggested that insurers could format EOB statements in ways that make them easier for employees to understand. For example, the EOB statements should clearly state that they are not bills, should include simple language, should explain medical codes, and should offer question‑and‑answer formats. The EOB statements should also state that the insurers use them to combat fraud and should identify whom to call if the injured employees suspect fraud.
Our research suggests that many injured employees may prove eager to assist in the fight against fraud. For example, a 1998 report by the Office of the Inspector General cited a Medicare survey that revealed that 74 percent of Medicare beneficiaries said they always read their EOB statements. The same survey found that if beneficiaries knew more about Medicare fraud, 89 percent of them would report it when they saw it. Similarly, a senior deputy district attorney for Orange County stated that she often finds that injured employees have no idea that they have been involved in workers’ compensation fraud schemes. She indicated that even though the insurers incur the fraudulent charges, the injured employees are still concerned and object to the use of their identities for others’ fraudulent gain. According to the senior deputy district attorney, many of these individuals stated that had they been aware of the fraudulent charges, they would have reported them.
To better ensure that the payments insurers issue to providers for workers’ compensation claims are based on valid services, the Legislature should require workers’ compensation insurers to periodically provide EOB statements to injured employees.
To reduce insurers’ potential underreporting of workers’ compensation fraud, CDI should do the following by June 30, 2018:
- Create a public report that ranks workers’ compensation insurers based on the effectiveness of their antifraud efforts, including the rate at which they submit fraud referrals.
- Add a requirement that it consider rates of fraud claim referrals when selecting insurers to audit and that it give priority to those insurers with high volumes of premiums and very low numbers of referrals.
To ensure the growth and effectiveness of its data analytics efforts to identify provider fraud, Industrial Relations should better document its data analytics effort within its protocol manual by June 30, 2018.
CALIFORNIA COULD IMPROVE ITS INVESTIGATION AND PROSECUTION OF WORKERS’ COMPENSATION FRAUD
As the Introduction discusses, CDI is the lead state agency for criminal investigations of insurance fraud. However, vacant fraud investigator positions limit CDI’s capacity to investigate suspected fraudulent workers’ compensation claims. CDI has a high number of vacancies for its fraud investigators partly because its salaries for these investigators have historically been lower than those for similar investigative positions at other state agencies, contributing to its inability to retain investigators and to hire new investigators quickly enough to outpace attrition. It also lacks a retention plan. Furthermore, because district attorneys’ offices depend on CDI’s investigators to bolster their investigative and prosecutorial efforts, CDI’s vacancy rate directly impacts district attorneys’ offices’ ability to investigate and prosecute cases.
Further, CDI’s vacancy rate has resulted in it underspending the workers’ compensation fraud assessment funds it has budgeted for personnel to investigate workers’ compensation fraud. State law mandates that CDI must receive a minimum of 40 percent of the total workers’ compensation fraud assessment each year. Although CDI could have received a higher proportion, in recent years the insurance commissioner and the Fraud Assessment Commission (Fraud Commission) have awarded CDI only this minimum allotment—$24 million per year in fiscal years 2015–16 and 2016–17. Nonetheless, CDI was unable to spend $2.4 million (10 percent) of that amount in fiscal year 2015–16. Instead of redirecting CDI’s unspent funds to district attorneys’ offices that could use it to investigate and prosecute more cases of workers’ compensation fraud, the insurance commissioner and the Fraud Commission used the unspent funds to reduce the assessment amounts the State collected from employers in a subsequent year. In effect, the insurance commissioner and the Fraud Commission chose to reduce the total amount of funds available to fight fraud rather than to redirect the funds to the district attorneys’ offices, which could have used it.
Ongoing Vacancies in Fraud Investigator Positions Have Reduced CDI’s Antifraud Efforts
CDI’s capacity to investigate workers’ compensation fraud in California has been limited by ongoing vacancies in its fraud investigator positions. As a result of these vacancies, CDI has closed a substantial number of referrals without investigation and potentially jeopardized the effectiveness of district attorneys’ offices’ efforts to investigate and prosecute workers’ compensation fraud. For example, for fiscal year 2016–17, the state budget authorized a total of 232 fraud investigators for CDI’s Enforcement Branch, which is responsible for five insurance fraud programs, including workers’ compensation. These positions are a combination of investigators and supervising fraud investigators we collectively refer to as fraud investigators. However, according to the Strategic Vacancy Report we received from CDI, it had 63 vacant fraud investigator positions as of February 2017, resulting in a vacancy rate of 27 percent.
Although fraud investigator vacancies caused CDI to spend less money than it budgeted for personnel costs, they also contributed to a decrease in the number of referrals CDI assigned for investigation and an increase in the number of referrals it closed due to insufficient resources. According to its July 2014 criminal activity report, CDI assigned 654 referrals for investigation by fraud investigators during fiscal year 2013–14. However, our analysis of its case management system indicates that it closed more than 1,600 (28 percent) of the roughly 5,700 referrals it had received during this period because of insufficient resources. In addition, CDI’s July 2016 report showed that the number of new referrals CDI assigned for investigation in fiscal year 2015–16 fell to 488, while the percentage of referrals it closed due to insufficient resources increased to 54 percent—2,911 out of 5,366. Although the percentage of referrals it closed due to insufficient resources decreased to 36 percent in fiscal year 2016–17, it only assigned 551 referrals for investigation that year, a 16 percent decrease compared to fiscal year 2013–14 levels. In total, CDI received 21,178 referrals from fiscal years 2013–14 through 2016–17. As Figure 7 shows, CDI closed 8,500 (40.1 percent) of these referrals due to insufficient resources. The total losses insurers reported paying related to these 8,500 referrals was about $160.8 million, or an average of about $18,900 per referral.
CDI Closed 40 Percent of the Fraud Referrals It Received Due to Insufficient Resources
Source: California State Auditor’s analysis of data obtained from CDI’s Fraud Integrated Database System.
Our analysis found that 80.4 percent of the referrals closed due to insufficient resources involved employee fraud and that the total losses insurers reported paying related to these referrals were about $66.8 million, or an average of about $9,800 per referral. Although only about 7.9 percent of the referrals CDI closed due to insufficient resources involved provider fraud, the total losses insurers reported paying related to these referrals totaled about $48.2 million, or an average of about $71,800 per referral. This average illustrates how costly provider fraud can be to the system. As we mention in the Introduction, employers bear the cost of the workers’ compensation system. Because fraud‑related losses can result in insurers raising the premiums employers pay, businesses may in turn increase the prices they charge consumers.
In addition, CDI’s high number of vacant fraud investigator positions can affect the ability of district attorneys’ offices to prosecute workers’ compensation fraud cases. For example, in its fiscal year 2016–17 application to the Fraud Commission for funding to fight workers’ compensation fraud, the LA District Attorney asserted it could not adequately process its caseload unless CDI had a sufficient number of fraud investigators to handle the cases. From fiscal years 2013–14 through 2016–17, Los Angeles County experienced the highest number of suspected fraudulent workers’ compensation claims of any county in the State. However, CDI’s South Los Angeles County regional office had a 53 percent vacancy rate for fiscal year 2015–16. In its funding application, the LA District Attorney stated that because it could not compensate for CDI’s resource limitations, it might have to decline new referrals, establish a minimum‑loss qualifying criterion, or close cases due to a lack of investigative resources. The LA District Attorney added that none of these options serve the public interest and explained that failing to investigate referrals, extending the time it takes to investigate cases and risking evidence spoilage and destruction, or permitting those engaged in fraud to continue to steal for longer periods of time are all unacceptable outcomes that can and should be avoided.
Finally, the vacancies in its fraud investigator positions have had ramifications beyond limiting CDI’s ability to combat workers’ compensation fraud. In a budget change proposal for fiscal year 2017–18, CDI stated it had the resources available to investigate just 5 percent of the annual referrals it received across all types of insurance, including automobile; disability and health care; property, life, and casualty; and workers’ compensation. We estimate that if CDI were fully staffed, it could potentially investigate an additional 200 to 300 workers’ compensation referrals per year.
Although CDI Has Taken Certain Steps to Address Fraud Investigator Vacancies, It Has Yet to Develop a Retention Plan
CDI has acknowledged its continuing high vacancy rate is a problem and has attempted to resolve it. For instance, CDI recognized that the salaries the State authorized it to pay its fraud investigators were less than those some other state agencies paid for their investigative positions. In response, it sought and received increases that have reduced these pay gaps as of July 2017. In addition, CDI has taken steps to create a specific team responsible for recruiting activities, it authored both a recruitment plan and strategic vacancy report, and it established a goal of achieving a vacancy rate of 5 percent or less. However, because many of CDI’s recruiting efforts are either in the planning stage or are just now being implemented, we do not feel that enough time has passed to evaluate their effectiveness. Further, we have concerns regarding CDI’s lack of a retention plan and a departmentwide process for performing exit interviews and surveys.
CDI acknowledges that the salaries the State authorized it to pay its fraud investigators were lower than those offered by other state agencies with similar investigative positions, which may have contributed to difficulties in both keeping staff and attracting candidates to fill vacant positions. In fact, CDI attributes its high vacancy rate primarily to the fact that many fraud investigators chose to leave CDI because the pay it could offer was significantly lower than that offered for similar sworn investigative positions at the California Department of Justice and the California Department of Corrections and Rehabilitation. As Figure 8 shows, our analysis of the State Controller’s Office’s payroll data indicates that CDI lost 98 fraud investigators from fiscal years 2013–14 through 2016–17. Of that number, 31 joined the California Department of Justice, 14 joined the California Department of Corrections and Rehabilitation, and 41 left state service.9 Of the fraud investigators that left state service, 27 were age 50 or older and potentially eligible for retirement.
From Fiscal Years 2013–14 Through 2016–17, CDI Lost 98 Fraud Investigators, 57 of Whom Accepted Other State Positions
Source: California State Auditor’s analysis of payroll data maintained in the State Controller’s Office’s Uniform State Payroll System.
* Although our analysis of payroll data indicates that these employees stopped receiving regular paychecks from the State, we did not assess whether these employees were merely inactive during that time and later returned to duty.
To help address this issue, the California Department of Human Resources (CalHR) established a pay differential that reduced the pay gap between CDI’s fraud investigators and those at other state law enforcement agencies, effective July 1, 2017. In fiscal year 2016–17, before the salary increase, the high end of the salary range for CDI’s investigators was about $7,100 per month, excluding overtime. As of August 2017, the high end of the salary range for CDI’s fraud investigators was about $7,800 per month, which narrowed the gap with other agencies. The top of the ranges for comparable positions at the California Department of Justice and California Department of Corrections and Rehabilitation were about $8,200 per month and $9,800 per month, respectively.
Further, in fiscal year 2016–17, CDI recognized the need to dedicate resources to recruiting in order to attract qualified applicants and reduce its vacancy rate to its goal of 5 percent or less. From fiscal years 2013–14 through 2016–17, CDI’s ability to hire new fraud investigators did not keep up with the rate at which its fraud investigators left. As Figure 9 demonstrates, we found that although CDI lost 98 fraud investigators during this period, it hired only 54 to replace them. According to a human resources analyst with CDI, the background investigation process for hiring a fraud investigator can take several months for some applicants. It includes both an internal component at CDI and an external component within CalHR, and CDI stated each process currently takes about three to six months.
CDI Lost More Fraud Investigators Than It Hired
Fiscal Years 2013–14 Through 2016–17
Source: California State Auditor’s analysis of payroll data maintained in the State Controller’s Office’s Uniform State Payroll System.
* Although our analysis of payroll data indicates that some of these employees stopped receiving regular paychecks from the State, we did not assess whether they were merely inactive during that time and later returned to duty.
To be more effective in addressing its vacancy problem, CDI created a Recruitment and Background Investigations Team in August 2016 to be responsible for its recruiting efforts. This team created both a recruitment strategic plan and strategic vacancy report. The recruitment strategic plan focuses on building new recruiting efforts by developing personalized relationships with applicants before they are hired and on structuring CDI’s internal organization and growing its budget to support more streamlined processes for background investigations and hiring. Because CDI has yet to fully implement the strategies it identified in its recruitment plan, we cannot yet tell whether they will have their intended effect: the reduction of vacancies.
We believe CDI could potentially increase its candidate pool by amending its recruitment plan so that it focuses in part on retired law enforcement officers. Although CDI hires candidates with this type of experience, its plan focuses on recruiting recent college graduates, attending career fairs, and developing recruiting opportunities at peace officer academies. However, we found that other states and two of the three district attorneys’ offices we visited hire retired or experienced law enforcement officers as investigators. The captain of CDI’s Recruitment and Background Investigations Team stated that the plan’s lack of recruiting activities for experienced and retired law enforcement officers was an oversight and that in practice, CDI strives to hire from a diverse applicant pool. Further, he stated that CDI values the qualities that retired and experienced law enforcement officers can bring, including having less need for training and being able to serve as mentors for less experienced investigators. The captain indicated, however, that CDI investigator positions may be more attractive to law enforcement officers who have retired from systems other than CalPERS, the State’s retirement system. He stated that law enforcement officers who have retired from CalPERS would have to be reinstated into the system, which would affect their retirement.
Despite CDI’s recent progress toward improving its recruiting efforts, we are unsure whether other factors exist that may affect its retention of fraud investigators because CDI has not created a retention plan to address fraud investigator separations and alleviate causes that are not related to pay. CDI could develop a retention plan based on the results of routine interviews of separating employees as well as surveys of its current employees to assess their job satisfaction. According to a human resources management textbook published by the University of Minnesota, the first step an entity should consider in developing a retention plan is a formal method to assess the satisfaction level of employees through exit interviews or surveys. From these types of data, CDI could begin to create its retention plan, making sure it is tied to organizational objectives. The plan should include analyses of the exit interview and survey results, the strengths and weaknesses of any prior retention efforts, the goal of the retention plan, and the specific strategies CDI plans to implement.
The chief of CDI’s human resources management division (human resources) stated that although human resources sent exit surveys to recently separated staff to voluntarily complete, it did not complete written analyses of the survey results. The chief asserted that she reviewed all exit surveys and followed up with program management and executive staff as warranted. Nevertheless, the captain of the Recruitment and Background Investigations Team stated his branch did not have access to the results due to confidentiality. In fact, although employees returned fewer than 15 surveys between 2015 and 2017, CDI could have used the results of the surveys to create a retention strategy to address common causes for separation that were unrelated to pay. Our analysis of the exit surveys found that six fraud investigators who separated from CDI highlighted dissatisfaction not only with pay, but with investigative training, growth opportunities, policies and procedures, and promotion potential.
District Attorneys’ Offices Could Have Used CDI’s Unspent Antifraud Funds
As the Introduction explains, state law requires that CDI receive at least 40 percent of the total fraud assessment amount each year, after incidental expenses. The assessment amounts for fiscal years 2015–16 and 2016–17 were each about $59 million, of which CDI received about $24 million. As Table 6 shows, CDI spent about $112,000 more than its allotment in fiscal year 2016–17, but it failed to spend roughly $2.4 million (10 percent) of its allocation in fiscal year 2015–16. In the proposed budget CDI presented to the Fraud Commission in September 2017 to determine the total assessment amount for fiscal year 2018–19, CDI divided its costs into three categories: salaries and benefits, operating expenses and equipment, and administrative support (that is, indirect costs). In the three most recent fiscal years, CDI has expended less for salaries and benefits than it proposed to the Fraud Commission and spent more than it proposed in one or both of the remaining categories. However, because CDI did not present this information side‑by‑side to the Fraud Commission, the Fraud Commission may be unaware of these trends. Additionally, information CDI presented to the Fraud Commission was not always consistent. For instance, CDI excluded encumbrances from a spending total for one year while including them in other totals for the same year.10
Sources: The Fraud Division’s Report to the Fraud Assessment Commission for fiscal years 2014–15 through 2016–17 and CDI’s internal expenditure documents.
Note: CDI presents proposed expenditures to the Fraud Commission at its September meeting when the Fraud Commission determines funding levels for the following fiscal year. In other words, proposed expenditures for fiscal year 2016–17 were presented at the September 2015 meeting.
* We used the positions calculated in the Fraud Division’s Report to the Fraud Assessment Commission because CDI uses a different methodology to calculate these figures than the data we used elsewhere in the text to report vacancy rates. Also, CDI based the calculated number of positions here on staffing levels for fraud investigators and support staff at its headquarters and regional offices.
CDI should be held to stricter reporting standards in order to increase transparency and help the Fraud Commission make more informed funding decisions. For example, when the district attorneys’ offices initially apply for grant funds, they have to provide expansive grant applications, detailed statistical reports on program activities, financial audit reports prepared by independent auditors, and carry‑over utilization requests for unexpended funds. However, because the State does not require CDI to apply for funds to investigate fraud, it is not subject to the same rigorous application requirements as district attorneys’ offices. In addition, district attorneys’ offices must provide detailed proposed budgets to the Fraud Commission, whereas CDI—as we previously mention—only provides its three major spending categories. In fact, at the Fraud Commission’s most recent meeting in September 2017, a commission member requested additional detailed budgetary information related to CDI’s staffing and personnel costs. CDI’s presentation to the Fraud Commission included an overview of the program successes and a request for additional funding. It did not, however, describe whether the additional funds were needed to maintain current staffing levels or to fully staff the program.
Finally, if district attorneys’ offices wish, for example, to spend more money on personnel and less on equipment than they originally presented, they must submit budget modification requests to CDI for approval. No similar requirement applies to CDI, as it asserts it is not technically a grantee, and thus it may underspend or overspend in categories without the Fraud Commission’s approval or knowledge. As a result of this lack of reporting requirements, the Fraud Commission may not have the information necessary to understand how CDI spends its assessment funds and what funding CDI needs, which is critical information for determining the appropriate total assessment the State needs to collect.
In addition, the insurance commissioner and the Fraud Commission missed an opportunity to increase the amount of money available to district attorneys’ offices when they decided to use CDI’s unspent funds to offset—or reduce—a subsequent year’s collections from employers. Although we found no direct evidence of a specific decision by the insurance commissioner and the Fraud Commission regarding the use of CDI’s unspent funds, a March 2017 letter from the Fraud Commission to Industrial Relations reduced the total assessment for fiscal year 2017–18 by an offset amount that appears to include CDI’s unspent funds from fiscal year 2015–16. State law gives the insurance commissioner and the Fraud Commission the option of using unspent CDI funds to offset or augment future program funding. The Fraud Commission’s chair stated that he suggested years ago that the Fraud Commission consider redirecting any unused CDI funds to district attorneys’ offices rather than offsetting future assessments; however, the Fraud Commission encountered issues that prevented it from redirecting the funds at that time, and it has not reconsidered this approach in subsequent years.
Had the insurance commissioner and the Fraud Commission redirected CDI’s unspent fiscal year 2015–16 allocation, district attorneys’ offices could have used the funds to further support their antifraud efforts. State law in effect caps the allotment that district attorneys’ offices can collectively receive at 60 percent of the assessment funds, after incidental expenses. From fiscal years 2013–14 through 2016–17, the insurance commissioner and the Fraud Commission awarded participating district attorneys’ offices—37 representing 44 counties for fiscal year 2016–17—the maximum funding allowable under the law. Although two of the three district attorneys’ offices that we visited underspent grant funds at least once from fiscal years 2013–14 through 2015–16, both offices received the required approvals from the insurance commissioner to carry over money to the subsequent year. Table 7 summarizes the three offices’ spending. During this time period, the district attorneys’ offices that applied for funding collectively requested more grant funding than was available for distribution, often citing the need for more investigative staff. The amounts the district attorneys’ offices requested suggest they could have used the unspent funds. Had the insurance commissioner and the Fraud Commission decided to redirect CDI’s unspent funds, they could have partially covered the deficit in requested funding for district attorneys’ offices. Table 8 summarizes the budget funding requested and approved for the three counties we visited.
Sources: The district attorneys’ offices’ grant applications, audit reports,
expenditure reports, and requests for use of unexpended funds; CDI's Fraud
Division’s reports to the Fraud Commission; CDI’s internal documents; and
interviews with staff of the Orange County District Attorney’s Office.
Note: Due to rounding, totals may differ slightly.
* This row includes both filled investigative and prosecutorial positions and was calculated using the number of positions and percentage of time devoted to the workers’ compensation program.
† The district attorneys’ offices’ indirect costs, including those not readily itemized but necessary to local program operations, may not exceed 10 percent of their personnel salaries (excluding benefits and overtime) or 5 percent of total direct program costs (excluding equipment).
Sources: District attorneys’ offices’ grant applications and CDI’s disbursement documents.
Note: A district attorney’s office’s requested grant amount may not be indicative of the total amount required to operate its program; that is, a district attorney’s office cannot request funding above a certain amount for indirect costs, even though its actual indirect costs may be greater than that amount.
The LA District Attorney provides an example of how the district attorneys’ offices might have used the additional funding. In its fiscal year 2016–17 grant application, the LA District Attorney—which consistently received the highest number of referrals in the State for the years we examined—requested additional funding for more personnel. It asserted that it would use the additional funds to add four workers’ compensation investigators and explained that it had the staffing resources available to fill these positions internally. However, the Fraud Commission approved only $6.7 million for the LA District Attorney—more than $1 million less than the amount it requested. Given that CDI has struggled to fill its fraud investigator positions, distributing its unspent funds to district attorneys’ offices to use for their fraud‑fighting efforts seems logical, particularly when doing so will likely enable the offices to increase the number of cases they investigate.
If the insurance commissioner and the Fraud Commission decide to use unspent CDI funds to augment funding to district attorneys’ offices, CDI will need to establish processes for doing so. CDI’s deputy general counsel indicated that before deciding to reallocate unspent funds, CDI must first address policy considerations and practical hurdles. He stated that deciding whether unspent assessment funds should be returned to employers as offsets against subsequent years’ collections or be used to augment the existing budget awards for district attorneys’ offices is a significant policy consideration. He added that employers might object if the State does not reduce the assessment even though it failed to spend the prior‑year’s funding. The deputy general counsel also asserted that CDI would need processes for deciding how much to reallocate to individual district attorneys’ offices and for transferring the funds from CDI to the offices. Currently, CDI does not have a process for either. Finally, CDI’s deputy general counsel stated that it might be preferable to allocate unspent funds to district attorneys’ offices that had not received their full allocations but had spent their awards. Given that state law requires district attorneys’ offices to submit to CDI independent audit reports—which would expose any historical pattern of underspending—and that CDI has information showing which district attorneys’ offices have consistently received less funding than requested, we find it reasonable that CDI develop a process to award and distribute any unspent CDI funds.
Although the District Attorneys’ Offices We Reviewed Have Their Own Approaches to Fighting Workers’ Compensation Fraud, All Three Coordinate Their Efforts With CDI
Our review of three district attorneys’ offices found that each had its own approach to fighting workers’ compensation fraud because each structures its investigative and prosecutorial efforts to reflect the individual characteristics of fraud in its county. Although each county fights all types of workers’ compensation fraud, we reviewed each of the three county district attorney office’s applications for assessment funds to gain an understanding of the elements that influence its approach. For example, the San Diego County District Attorney’s Office asserted in its fiscal year 2017–18 application that premium fraud associated with the underground economy especially plagues the county and that the size of its population and its physical proximity to an international border lead to a high volume of workers’ compensation fraud cases. The LA District Attorney’s application, on the other hand, stated that it focused primarily on both provider and employer fraud but had opted to forego pursuing misdemeanor employer fraud cases due to the lack of investigative resources at CDI’s Southern Los Angeles County Regional Office. Lastly, the Orange County District Attorney’s Office stated in its application that it prioritizes provider fraud in particular because it has determined that, among other factors, the mix of a large workforce coupled with the skyrocketing growth of the health care industry in the county creates the essential demographics to make it prone to provider fraud. In general, we found that the district attorneys’ offices made choices depending on the type and magnitude of fraud affecting the counties and the available resources. This approach appears reasonable.
In their applications for assessment funds, all three of the district attorneys’ offices we visited submitted joint plans with CDI that described their efficient use of joint resources. The request for applications requires that all applicants submit joint plans that create the framework for effective communication and resources management in the investigation and prosecution of fraud. Both the county prosecutor and the captain of each CDI regional office that is responsible for that county must agree upon the plan. For example, we found that joint plans for all three district attorneys’ offices we visited included processes for assessing whether cases merit opening before the offices use their investigative and prosecutorial resources. The LA District Attorney’s joint plan indicates that the office conducts a preliminary review to determine the feasibility of asking the referring party to make a case presentation for any suspected fraudulent claim that it believes is based on sufficient evidence. Subsequently, it will determine whether the case merits opening. The Orange County District Attorney’s Office and the San Diego County District Attorney’s Office both conduct similar preliminary reviews.
In addition, the three district attorneys’ offices we visited use a process generally referred to as vertical prosecution, which aids in balancing their efforts between investigation and prosecution. As we discuss in the Introduction, a CDI fraud investigator, an investigator at a district attorney’s office, or both may investigate a case before it is prosecuted by the district attorney’s office. To balance these efforts, the vertical prosecution process requires a case investigator to communicate with the assigned prosecutor at the beginning of the investigation so that they can work together to build the case from inception through final adjudication. For example, according to the San Diego County District Attorney’s Office’s most recent assessment application, it assigns a prosecutor to a case when CDI opens an investigation, thereby providing CDI’s investigator with a legal resource should any issues arise. For some cases, the investigator and prosecutor will hold regularly scheduled meetings and share case updates throughout the investigation. This enables the prosecutor to know the facts of the case, and it also ensures that CDI uses its investigative resources for work that is necessary to the case’s prosecution. The LA District Attorney asserts that vertical prosecution is an essential component of developing and implementing an effective and efficient investigative prosecution plan.
To better address vacancies in its fraud investigator positions, CDI should take the following actions by June 30, 2018:
- Develop and implement a retention plan. This plan should be based on the results of in‑person exit interviews with separating staff or similar tools, such as satisfaction surveys, to identify and address potential causes for separation other than pay. CDI should share the results of any trends arising from its exit interviews as well as its analyses of survey responses with the appropriate units as it deems necessary.
- Revise its recruiting plan to include the recruitment and hiring of retired local law enforcement officers.
To better enable the Fraud Commission to determine an appropriate amount for the total annual fraud assessment, CDI should, within 60 days and periodically thereafter, meet with the Fraud Commission and agree upon specific information to include in the Fraud Division’s report to the Fraud Commission. Additional information could, for example, include a comparison of proposed, projected, and actual expenditures by category for a specific fiscal year, calculated using a consistent methodology.
To better ensure the timely and effective use of fraud assessment funds to fight workers’ compensation fraud in California, CDI should, by June 30, 2018, develop and implement a process to use its unspent funds to augment funding to district attorneys’ offices rather than to offset collections from employers for subsequent years.
We conducted this audit under the authority vested in the California State Auditor by Section 8543 et seq. of the California Government Code and according to generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives specified in the Scope and Methodology section of the report. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.
ELAINE M. HOWLE, CPA
December 12, 2017
Mike Tilden, CPA, Audit Principal
Dale A. Carlson, MPA, CGFM
Daniel Mitchell, MBA, CFE
David A. Monnat, CPA
Brigid Okyere, MPAc
Sean Wiedeman, MBA
Ben Ward, CISA, ACDA, Audit Principal
Derek J. Sinutko, PhD
Mary K. Lundeen, Sr. Staff Counsel
For questions regarding the contents of this report, please contact
Margarita Fernández, Chief of Public Affairs, at 916.445.0255.
8 According to its website, the Ponemon Institute conducts independent research on privacy, data protection, and information security policy to enable public and private organizations to have a clear understanding of the trends in practices, perceptions, and potential threats that will affect the collection, management, and safeguarding of personal and confidential information about individuals and organizations. Go back to text
9 The remaining 12 went to a variety of other state agencies. Of the 41 who left state service, several took positions with district attorneys’ offices or other local law enforcement entities, some of which also offered higher pay. Go back to text