Report Number: 2016-128

In-Home Supportive Services
The State Could Do More to Help Providers Avoid Future Payment Delays

Use the links below to skip to the section you wish to view:

• Social Services Has Not Provided Adequate Direction or the Appropriate Tools for Counties to Effectively Monitor and Resolve Provider Timesheet Exceptions


• Social Services and OSI Have Not Analyzed Key Information That Could Aid in Monitoring Timesheet Exceptions and Payments


• The Complexity of Certain Time‑Reporting Protocols Hinders Providers’ Ability to Accurately Complete Their Timesheets


• Other Areas We Reviewed


• Scope and Methodology

---

Social Services Has Not Provided Adequate Direction or the Appropriate Tools for Counties to Effectively Monitor and Resolve Provider Timesheet Exceptions

Key Points:

• Social Services has not given adequate guidance to counties on how to handle timesheets with exceptions, resulting in inconsistent methods of addressing these issues. Without proper guidance, at least one county handles certain exceptions by printing and sending blank replacement timesheets to providers, while other counties initially review and attempt to correct exceptions before resorting to issuing replacement timesheets.


• CMIPS II’s inability to provide detailed reports makes it difficult for counties to identify providers that consistently make errors on their timesheets or experience payment delays. If counties were able to identify providers with recurring timesheet problems or delays, they would be able to more effectively target their support efforts to ensure that subsequent timesheets are submitted accurately and payment is issued promptly.


• Social Services could more effectively handle communications to providers by developing a system that sends them automated notifications about the status of timesheets and paychecks. Currently, providers seeking this information must contact their county or the State, resulting in large call volumes at the counties and limiting the ability of county staff to focus on those providers who continue to have difficulties preparing and submitting valid timesheets.

Social Services Has Not Issued Adequate Guidance to Counties That Could Help Them Reduce Delayed Payments to Providers

Although providers submit their timesheets directly to the TPF, the TPF sends any timesheets with exceptions to the counties for handling. Roughly 1.2 million—more than 3 percent—of the 34.4 million timesheets that the TPF processed from November 2013 through July 2016 had at least one exception that caused payroll delays. As discussed in the Introduction, the TPF scans and processes timesheets and then imports the timesheet data into CMIPS II. For timesheets without exceptions, CMIPS II processes and sends the payment information to SCO so that it can issue paychecks. However, the TPF sends timesheets with exceptions—such as those missing signatures or containing unreadable entries—to the applicable county office. The county is then responsible for taking action to enable the provider to be paid. Because Social Services has not issued adequate guidance to counties for handling timesheet exceptions, counties have developed their own processes, which can result in delayed payments if these processes do not handle the exceptions in an efficient manner.

Social Services has issued some direction to counties on how to handle timesheet exceptions, but the guidance does not explicitly instruct counties to handle exceptions by first attempting to make a timesheet correction. In some cases, guidance has been conflicting. For example, the CMIPS II User Manual issued by Social Services directs county staff to review the timesheet and take appropriate action based on their respective county’s processes, but also states that workers must issue a replacement timesheet in almost every instance in which a timesheet is determined to contain an error. However, the systems and administrative branch chief at Social Services explained to us that counties should be handling exceptions by first attempting to read the timesheet and contacting the provider to confirm the entry, and then making a payment correction and processing payment. If the county cannot determine the correct entry or is unable to contact the provider, it should send a replacement timesheet to the provider. Because Social Services has not issued these specific instructions to counties, there is inconsistency in the manner in which counties handle timesheet exceptions. Specifically, four of the five counties we visited—Sacramento County, San Bernardino County, San Diego County, and the City and County of San Francisco—stated that they attempt to resolve certain exceptions by initiating timesheet corrections. The fifth, Los Angeles County, does not require county staff to review timesheets for corrections they could make and instead directs its staff to issue blank replacement timesheets to the affected providers and have them correct their own exceptions.

In addition to each county’s process for resolving exceptions, the length of time a paycheck is delayed can vary depending on the type of exception. Some exceptions inherently take longer to resolve, such as when a timesheet is missing a signature. In such cases, the county generally mails the provider a blank replacement timesheet to complete and mail to the TPF, thereby prolonging the payment process. From November 2013 through July 2016, more than 189,000 timesheets statewide were missing either a recipient signature, a provider signature, or both signatures. However, as shown in Figure 3, the most prevalent timesheet exception category that caused a delay was an unreadable entry, which the county may be able to resolve quickly by reviewing the scanned image of the timesheet or calling the provider to determine the correct entry and then making the adjustment directly in CMIPS II. When the four counties made corrections within CMIPS II to resolve timesheet exceptions, payments to providers were processed in four business days, on average. In contrast, the average for all five counties was more than 14 business days to resolve timesheet exceptions in those instances when they issued replacement timesheets.

Figure 3
Categories of Exceptions in the Case Management, Information and Payrolling System That Caused Payroll Delays From November 2013 Through July 2016

Source: California State Auditor’s analysis of data obtained from the California Department of Social Services’ In‑Home Supportive Services program, as maintained in the Office of Systems Integration’s Case Management, Information and Payrolling System.

* Some timesheets contained multiple exceptions.

^† This category includes exceptions with fewer than 100,000 occurrences, such as timesheets missing entries, multiple entries in a time entry box, and provider or recipient eligibility issues.

In one example we reviewed for Los Angeles County, the TPF processed a provider’s timesheet as an exception with an unreadable entry and forwarded it to the county for review. In reviewing a copy of the timesheet, we were able to read the entry that the TPF had rejected. However, in accordance with its policy, Los Angeles County sent a blank replacement timesheet to the provider for resubmission, rather than having its staff review the timesheet or attempt to correct the entry. This process essentially restarted the timesheet process and resulted in the provider’s paycheck being issued 13 business days after the TPF received her initial timesheet—nine days longer than the average time we noted when counties made payment corrections to resolve timesheet exceptions.

Not only does requiring the provider to complete a replacement timesheet delay payment, but it also increases the potential for the provider to make an inadvertent error on an unrelated portion of the timesheet that was initially correct. Exceptions resulting from these additional errors may require the need for completion of yet another replacement timesheet, which would compound the delay even further. From November 2013 through July 2016, the five counties we visited had more than 333,000 original timesheets with exceptions for which at least one replacement timesheet was issued to resolve the exception. Of these, roughly 90 percent—or 299,000 original timesheets—were processed at the TPF after the provider submitted a single replacement timesheet. However, the remainder—roughly 34,000 timesheets—were processed at the TPF only after the provider submitted two or more replacement timesheets, which resulted in delayed payments to providers averaging 24 business days, or nearly five weeks. Providers may incur a financial hardship if payments are delayed. Specifically, several recipients and providers testified at legislative hearings in 2016 describing hardships that providers have faced when their paycheck was delayed, such as not being able to buy food or pay rent. Beyond such financial difficulties, delays in payment processing may force providers to seek other employment with more stable methods for receiving payment, resulting in recipients having to find other caregivers.

For timesheets that have unreadable entries or missing signatures, the system may still be able to recognize and validate some of the data recorded on the document, and CMIPS II could be modified to have the ability to reprint these valid data on a new replacement timesheet. The provider would then need to complete only those sections that contained the unreadable data or missing signature, minimizing the chance of making an additional error. Nonetheless, without adequate guidance from the State, counties may continue to send replacement timesheets to providers without requiring their staff to review the scanned image of the timesheet or to work with the provider in an attempt to make corrections, thus contributing to payment delays for providers.

Social Services’ guidance is also unclear about the length of time providers must wait before reporting a paycheck lost or stolen. From November 2013 through July 2016, counties reported more than 30,000 lost, stolen, or damaged paychecks statewide. The Social Services IHSS program manual states that counties shall make a request for a replacement paycheck expeditiously, but no sooner than five days from the date the original paycheck should have been received. Additionally, Social Services instructed HPE to respond to paycheck inquiries by informing providers that their paychecks should arrive within 10 business days of the issue date and to refer them to the county after this time frame. However, the counties we visited provide differing information to providers regarding the length of time they must wait before reporting a lost or stolen paycheck. For example, the acting division manager at Sacramento County stated that the county requires providers to wait 10 calendar days from the issuance of the check before requesting a replacement; whereas, San Bernardino County informs providers that they must wait 10 business days, which generally equates to an extra four calendar days.

Social Services could not provide documentation to support why it instructed HPE to inform providers to wait 10 business days, nor could it adequately justify this waiting period. SCO will place a stop payment on a check and issue a duplicate check after seven business days, and it does not have a separate wait time for providers. According to Social Services’ system and administrative branch chief, the wait time before reporting a paycheck lost or stolen is intended to account for the mail transit time of issued paychecks. However, according to information from the U.S. Postal Service’s website, locations within California should generally receive their first‑class mail within two to three business days. As a result, we believe a more reasonable amount of time for providers to wait before reporting a paycheck lost or stolen is five business days, which allows the average two to three business days for mail delivery, plus two extra days for any unforeseen circumstances. We recognize the inherent risk that a provider may receive his or her paycheck after reporting it as lost or stolen, and that SCO’s process of issuing a replacement paycheck requires it to issue a stop payment on the original paycheck. This subsequent step will result in the provider experiencing an additional delay while waiting for the replacement paycheck and may be frustrating for providers if they receive the original paycheck after requesting a replacement. However, we believe Social Services should allow providers to request replacement checks expeditiously and inform them of the risk they incur in relinquishing their rights to the original paychecks.

The CMIPS II Reporting Function Does Not Effectively Assist Counties in Reducing Timesheet Errors or Payment Delays

CMIPS II does not report information at a level of detail and in a usable format to allow counties to effectively identify providers who routinely make timesheet errors or consistently experience delays in receiving paychecks. CMIPS II was designed to support counties in managing the IHSS program, to monitor and control program activities and expenditures, and to specifically identify problem areas. Data from the system are an important resource for accomplishing these objectives, and the CMIPS II User Manual states that CMIPS II allows for the on‑demand printing of reports, and for data in the reports to be exported, filtered, and sorted as needed. However, representatives from the five counties we visited believe that the reports do not adequately perform this function. For example, CMIPS II generates a timesheet exceptions report that is intended to help county staff assess their workload when processing timesheets with exceptions, identify areas where additional training on timesheet completion for providers and recipients may be needed, and assess potential fraud or authorization issues for submitted timesheets. However, the counties informed us that the CMIPS II timesheet exceptions report does not allow them to filter the data in the manner they need to manage staff workload. Additionally, the CMIPS II timesheet exceptions report does not allow counties to identify trends in exceptions over time, because it removes exceptions from the report within one pay period after they are resolved. Consequently, providers could conceivably continue to make the same errors on their timesheets without drawing the attention of the county. Representatives from the counties we visited confirmed that in order to track providers who have recurring problems with their timesheets they would have to develop their own customized reports or track this information manually. In response to our inquiries, Social Services’ systems and administrative branch chief told us that Social Services has recently requested that HPE report additional data to track providers who repeatedly incur exceptions on their timesheets.

CMIPS II also has the capability for counties to download timesheet data daily or monthly to conduct additional analyses. However, we observed that the format of these data downloads does not allow counties to sort and filter the data to perform specific analyses as needed. In fact, the five counties we visited had developed their own processes to sort the exception data from CMIPS II in ways that enable them to follow up with specific groups of providers and to manage staff workload. Examples of the types of sorting they perform include sorting by exception type, preferred language of the provider, and county staff member assigned to the exception. San Diego County’s provider services manager exports the CMIPS II data into a spreadsheet and manually reformats it to produce a report that can be sorted and used to track exceptions over time. The City and County of San Francisco creates a custom report of a similar nature. In two other counties—Los Angeles County and San Bernardino County—staff collect and track some exception data manually outside of CMIPS II. Sacramento County’s payroll manager told us that because she was unable to use the exception data to balance her staff’s workload, her staff manually track payment corrections used to address exceptions and she uses these data instead. However, all five counties informed us that the manual processes are time‑consuming and that having an automated report in CMIPS II would improve accuracy. Moreover, time spent creating these reports takes county staff away from their objective of directly assisting providers who have difficulty completing timesheets or who have consistently experienced payment delays. The deputy director of the adult programs division at Social Services agreed that current CMIPS II reports do not allow counties to sort and filter data sufficiently to enable them to manage staff workload or to target providers with certain types of exception problems.

If counties were able to identify providers with recurring timesheet problems or delays, they could more effectively target their support efforts to ensure that subsequent timesheets are completed accurately and paychecks are issued promptly. According to the systems and administrative branch chief, Social Services has not had sufficient resources for implementing changes to improve the timesheet exceptions report because resources have been devoted to legislative mandates, such as overtime rules, since CMIPS II’s implementation. Although we recognize that these mandates have required a significant investment of resources, making these changes to the reporting function would allow counties to be more effective in tracking and analyzing trends in timesheet exceptions and paycheck delays.

CMIPS II Could More Effectively Communicate to Providers the Status of Their Timesheets and Paychecks

During the audit, we noted that providers seeking information about the status of their timesheets or paychecks generally contacted their county or the State, resulting in a significant number of phone calls. For example, San Diego County processed 15,022 phone calls in September 2016 alone, which the public authority deputy director stated were mostly from providers checking on the status of their paychecks. Each of the five counties we visited, and most of the 10 other counties we spoke with, stated that the majority of the phone calls they receive from providers pertain to inquiries about the status of their timesheets or paychecks. Although the five counties we visited did not maintain records on the specific reasons for provider calls, Santa Cruz County did track this information and found that more than 62 percent of the calls received from January 2013 to November 2016 involved status checks on provider timesheets or paychecks. Additionally, the State maintains a call center operated by HPE specifically designated for providers to obtain information about the status of their timesheets and paychecks, and this call center averaged more than 50,000 calls per month from July 2015 to July 2016. These high call volumes demonstrate that providers desire frequent status updates and suggest that the State could be more efficient at communicating these updates.

As previously mentioned, state law authorizing the development of CMIPS II required the system to incorporate technology that could be readily enhanced and modernized for the expected life of the system. Automatic status updates have become a standard function of modern technology, such as tracking the status of a postal delivery or a food delivery. Social Services could incorporate similar functionality into CMIPS II that would notify providers via their preferred method of communication—email, text message, or automated phone call—when their timesheet is received at the TPF, when payment information is approved and sent to SCO for processing, and when their paycheck is printed. The notification could also communicate to providers whether any exceptions were noted on the timesheet that prevent payment from being processed and could direct them to a specific point of contact at the county with whom they could work to address the exceptions.

According to a chief deputy director at Social Services, implementing automatic status notifications within CMIPS II could be a costly upgrade and would soon be irrelevant, because Social Services is planning to implement electronic time reporting in June 2017. This feature will allow providers to submit their timesheets using a website and receive immediate notifications regarding whether Social Services has accepted and processed their timesheets for payment. However, automated notifications have become a standard function in modern technology, and Social Services should be seeking opportunities to cost‑effectively incorporate such technology into its system, given the volume of inquiries about the status of timesheets and payments. We further discuss Social Services’ plans for electronic time reporting in the Other Areas We Reviewed section.

Automatic status notifications would provide an ongoing benefit to providers who continue using paper timesheets, such as providers without Internet access. To provide some context on the proportion of providers who may continue using paper timesheets, we contacted the Washington State Department of Social and Health Services, which implemented electronic timesheets in March 2016 for its in‑home care program. Its representative informed us that as of January 2017, approximately 83 percent of providers have opted to use electronic timesheets, meaning that 17 percent of its providers still use paper timesheets. In California, if 17 percent of providers continue to use paper timesheets after electronic time reporting is implemented, roughly 78,000 providers would still be exclusively using the paper method. These providers would likely benefit from automated status updates. Additionally, providers who take advantage of electronic time reporting may still want to know the status of their paychecks, even after submitting their timesheets electronically, and having this functionality in CMIPS II could meet that need.

Representatives from four of the five counties we visited and most of the 10 counties we interviewed stated that they would generally be in favor of this additional functionality in CMIPS II. Automatically sending these notifications to providers would have the added benefit of likely reducing the number of calls that counties receive about the status of timesheets and paychecks, which would allow the counties to focus their resources on other aspects of timesheet processing, such as identifying common timesheet issues and developing training for providers to address those issues.

Recommendations

To ensure that counties are handling timesheet exceptions consistently and minimizing delays, Social Services should develop and issue procedures by July 2017 to require the counties to first attempt to correct timesheet errors for specific types of exceptions before mailing blank replacement timesheets to providers. For example, counties should attempt to correct timesheets with unreadable entries or entries that exceed 24 hours in a day by reviewing the timesheet, contacting the provider if necessary to clarify the intended entry, and making a timesheet correction in CMIPS II. Additionally, Social Services should review a random sample of exceptions at least quarterly to ensure that the counties are following its new procedures.

To reduce the likelihood of inadvertent errors on replacement timesheets, Social Services should create functionality within CMIPS II to allow replacement timesheets to be printed with data that had been submitted correctly on the original timesheet. Social Services should develop a plan by August 2017 that outlines actions, such as assessing the cost and seeking funding from the Legislature if necessary, that will be taken to create the functionality.

To ensure that counties follow a consistent and expeditious policy for responding to providers who report lost or stolen paychecks, Social Services should issue a policy by September 2017 that allows providers to request replacement paychecks after five business days from the issue date of the lost or stolen paychecks.

To assist counties in resolving exceptions efficiently and in managing their workload, Social Services should by December 2017 develop a timesheet exceptions report in CMIPS II that enables county staff to categorize common exceptions, identify providers with recurring exceptions, and track timesheet processing workload over a period of time. Social Services should also train county staff on the most effective use of these reports.

To effectively communicate information to providers and reduce call volumes at counties, Social Services should implement functionality within CMIPS II by December 2017 to provide automated notifications to providers about the status of their timesheets and paychecks, including when timesheets are received and processed, when paychecks are processed, and whether there are exceptions on timesheets that would delay processing paychecks and whom to contact at the county to address those exceptions.

Back to top

---

Social Services and OSI Have Not Analyzed Key Information That Could Aid in Monitoring Timesheet Exceptions and Payments

Key Points:

• Social Services and OSI have not been able to effectively monitor the IHSS program because they have not received key data from HPE on the time it takes to process timesheets, the number of timesheets with exceptions for each county, and the time required to resolve them, even though the agreement with HPE specifically requires this information.


• Social Services does not have a process for monitoring whether EDD is printing and mailing timesheets, and whether SCO is printing and mailing paychecks, within their contracted time frames.


• Social Services lacks a formal process for systematically reviewing common themes among concerns raised by providers and recipients regarding IHSS timesheets and payments.

Social Services and OSI Do Not Monitor the Timeliness of Timesheet Processing

Social Services and OSI have not monitored how long the CMIPS II contractor, HPE, takes to process timesheets at the TPF or how long counties take to resolve exceptions. As a result, since achieving statewide implementation of CMIPS II in November 2013, neither entity has been able to determine whether HPE is meeting its contractual obligations, resulting in potential missed opportunities to identify deficiencies in the timesheet process. As described in the Introduction, Social Services is the primary state agency responsible for the IHSS program and has an agreement with OSI to contract with HPE and to jointly ensure successful operation of IHSS and CMIPS II. OSI’s contract with HPE requires HPE to process timesheets within five business days of receipt, and for OSI to assess penalties for timesheets that are not processed promptly. The contract also requires HPE to report to OSI and Social Services the number of timesheets with exceptions that it sent to each county for follow‑up, and the time it took to resolve them. Therefore, we would expect that Social Services would exercise its oversight role to ensure that OSI was enforcing the contract and would review the exception data from HPE’s reports to identify problems with timesheet protocols or with county processes. We would further expect that OSI would monitor these reports to ensure that HPE processes timesheets promptly and submits the required data, and that OSI would assess penalties when appropriate.

However, neither OSI nor Social Services has been enforcing these key provisions. Because Social Services is the state agency ultimately accountable for the IHSS program, its responsibility for determining whether the TPF meets or exceeds its five‑day time limit to process timesheets is critical to ensuring that providers are paid promptly. Social Services confirmed that it did not receive or review any reports showing whether HPE was meeting the five‑day time frame or whether OSI was enforcing the time frame. The chief deputy director at Social Services stated that OSI has primary responsibility as the project manager for the contract with HPE to monitor whether the TPF processes within the five‑day time frame the timesheets it receives. However, while Social Services’ agreement with OSI specifies that OSI has primary responsibility for managing HPE, it also states that Social Services is responsible for overseeing and participating in the review of HPE’s activities and for actively monitoring status reports from OSI. Without regularly monitoring the processing of timesheets, Social Services cannot ascertain whether HPE is meeting contract requirements and whether the State should assess any penalties. Specifically, OSI does not review any data on the number of timesheets HPE processed within the five‑day limit. Although the CMIPS II project director at OSI acknowledged the oversight, she stated that her management team thought these data were included in the information they were monitoring. However, we question how the management team at OSI thought it could be monitoring information that HPE had not reported. Additionally, according to its contract with HPE, OSI may assess penalties to HPE of $100 per timesheet per day for any timesheet that is not processed within the five‑day time frame, not to exceed $10,000 per day. Although we did not identify any instances during the period from November 2013 through July 2016 that would have warranted OSI assessing HPE a penalty, OSI should monitor whether the TPF is processing timesheets within five business days so that it can appropriately assess penalties if warranted in the future.

Social Services and OSI also have not ensured that HPE provides required timesheet exception data that could identify problems in processing timesheets. As mentioned previously, under its contract with OSI, HPE is required to report to OSI and Social Services the number of timesheets with exceptions that were directed to each county and the length of time it took to resolve those exceptions. Although HPE submits to OSI a monthly operations management report (management report) containing data on other required performance metrics, such as system availability and the total number of exceptions per month, the management report does not contain details on timesheet exceptions by county or the time taken to resolve them. Social Services’ systems and administrative branch chief acknowledged that, until we brought it to her attention, she was not aware of the detail that should be in these management reports or of the contract requirement that HPE provides this timesheet exception information. OSI’s CMIPS II project director told us that the management report does not contain this information because it was not intended to monitor county work efforts. Nevertheless, we would expect that OSI and Social Services would require HPE to include this information in its reports because OSI’s contract with HPE specifically requires it, and because timesheet exception data is relevant and necessary for assessing the performance of the timesheet and payment process. For example, obtaining this information could have assisted Social Services in detecting discrepancies among different counties’ processes for handling timesheets with exceptions, as we describe in the previous section, and correcting its policies sooner.

Social Services Lacks a Process for Monitoring the Printing and Mailing of Timesheets and Paychecks Performed by Its Contractors

Social Services does not have a process for monitoring whether EDD and SCO are meeting the required time frames for printing and mailing timesheets and paychecks. As shown in Figure 1 on page 8, CMIPS II sends timesheet data to EDD and payment data to SCO each business day. The agreement with EDD stipulates that it is responsible for printing timesheets and mailing them to providers within two business days of receiving timesheet data from CMIPS II. Although EDD sends an email to Social Services to confirm receipt of the timesheet data shortly after receiving it but before printing or mailing the timesheets, according to the chief of the printing facility, this automated email is the only notification that EDD sends to Social Services, and Social Services has not requested any additional information. The agreement with EDD does not require EDD to provide information showing how long it took to mail timesheets, and according to the Social Services systems and administrative branch chief, this omission was the result of an oversight during the development of the agreement. However, without this information, Social Services cannot ensure that EDD is printing and mailing timesheets within the required time frame, which prevents Social Services from holding EDD accountable for any delays within its span of responsibility. Providers who experience delays in receiving their timesheets are at risk of not having sufficient time to complete and submit them at the end of the pay period, which can result in paycheck delays.

Payroll staff from some of the counties we visited or interviewed expressed concerns that timesheets were not always mailed promptly to providers. They noted that many of the inquiries they receive from providers pertain to delays in receiving timesheets. To gain some assurance that EDD was meeting its required two‑day time frame, we selected three dates in October 2016 during which EDD printed and mailed a high volume of timesheets, and reviewed its internal records to identify processing times. Although we found that EDD had met its two‑day obligation to print and mail timesheets for these three dates, we were unable to conduct a comprehensive review to determine whether EDD was consistently meeting its obligation because Social Services does not require EDD to track this information. In response to our concern, Social Services initiated discussions with EDD in January 2017 to add a requirement in the next renewal of their interagency agreement—estimated to begin July 2017—for EDD to provide reports on printing and mailing dates.

SCO’s agreement with Social Services requires SCO to mail paychecks and electronic funds transfer notices within one business day of receiving payment information from CMIPS II, as well as to send electronic data files to Social Services confirming that it has processed the payments. Social Services does not receive these files directly, although HPE receives them and enters the data into CMIPS II, thereby making this information available to Social Services. Our review of CMIPS II data found more than 24,000 paychecks issued on a single day in June 2014 that did not meet the one‑day time frame specified in SCO’s contract with Social Services. However, according to records from Social Services’ fiscal systems and accounting branch, this situation occurred because Social Services took an extra day to issue its authorization to SCO, which SCO requires before it can issue the paychecks. Specifically, the fiscal systems and accounting branch authorized payment a full day after SCO received the payment data from CMIPS II, resulting in SCO issuing these paychecks in two days, rather than one. Social Services’ systems and administrative branch, which oversees CMIPS II, was unaware that another branch within Social Services had caused this delay because it had not monitored whether SCO was meeting its time frame. The branch chief stated that her staff does not review reports to confirm how long it takes for SCO to issue paychecks, but instead relies on HPE to alert Social Services of any anomalies. Nevertheless, as the state entity responsible for the operation of IHSS, Social Services should review this information regularly to determine whether there are any delays in SCO’s issuance of payments and, if so, their cause and how to resolve them.

Social Services Is Not Using Information It Receives From Providers and Recipients to Identify Issues With IHSS Timesheets and Payments

IHSS providers and recipients provide feedback to Social Services on issues with IHSS, including concerns about timesheets and payments. Social Services maintains a communications page on its website by which providers and recipients can express their concerns by entering information into an online form. Providers and recipients may also use the mail or send emails and faxes about concerns they have regarding IHSS timesheets and payments directly to Social Services; the California Health and Human Services Agency, which oversees Social Services; and the Governor’s Office. Letters and emails received by the California Health and Human Services Agency or the Governor’s Office are typically forwarded to Social Services for a response. Although many providers contact their respective counties regarding the status of timesheets and payments, some choose to direct their correspondence to the State with the intent of having the issue be addressed throughout the program.

Social Services has a process for receiving and responding to complaints from its stakeholders, but it does not aggregate and analyze these concerns in a manner that would allow it to address systemic issues within the IHSS program. Its process for responding to complaints generally involves assigning an analyst to use CMIPS II to research the concerns reported and to prepare a response letter. At least two managers then review the response before the letter is sent. However, we found that Social Services does not always document its research and analysis of specific issues—including those related to IHSS timesheets and payments—nor does it compile information pertaining to these types of issues. Social Services could more effectively manage the IHSS program if it systematically reviewed common themes raised by recipients and providers. For example, identifying numerous concerns in a specific county about replacement timesheets could alert Social Services to the need to provide strategic support in that region. Moreover, a high‑level approach to identifying and addressing concerns could help Social Services ensure consistency in its responses.

We also noted that Social Services does not appear to be ensuring that it is thorough and prompt in addressing complaints. In particular, the contents of many of the response letters we reviewed seem to indicate that staff did not follow up with the individuals initiating the complaints to seek additional information or to clarify the issues they raised. Although the unit manager responsible for reviewing response letters informed us that he reviews the research performed by his analysts, we noted that this review does not always ensure all issues in the complaints are addressed, as evidenced by three of the files we examined. Furthermore, Social Services did not reply in a timely manner to the 10 complaints we reviewed. Specifically, it did not send a letter within 20 days in response to any of the complaints we reviewed, and it did not reply within 40 days for half of these complaints. The manager attributed the delays to Social Services’ other responsibilities that have taken precedence over addressing complaints, such as implementing the FLSA requirements. We expected Social Services would at least acknowledge receipt of the complaint within 10 days and provide an estimate of when the complainant could expect a full response. However, by the time Social Services responded to several of the complaints, the delays in timesheet processing or payments had been addressed, effectively diminishing the value of the reply from Social Services.

Recommendations

To ensure that HPE is meeting its contractual obligation for processing timesheets, OSI should monitor whether the TPF is processing timesheets within five business days, assess penalties when warranted, and report the results of this monitoring to Social Services on a monthly basis.

To ensure that HPE is meeting its contractual requirements, Social Services should review timesheet processing data and reports and follow up with OSI to make sure it is taking corrective action if HPE exceeds the agreed‑upon processing time frames.

To ensure that OSI is adequately monitoring HPE and to allow for more proactive management of the IHSS program, Social Services should work with OSI to enforce the contract provision requiring HPE to submit monthly data on the number of timesheets with exceptions by county and the time taken to resolve those exceptions. Moreover, Social Services should develop a process for regularly reviewing these data to detect any discrepancies among the counties’ processes for handling timesheets with exceptions.

To ensure that the reports it receives from HPE are complete and allow it to better manage CMIPS II and support the IHSS program, OSI should enforce its agreement requiring HPE to submit monthly data on the number of timesheets with exceptions by county and the time taken to resolve them.

To enable it to track whether EDD is meeting its contractual time frame for printing and mailing timesheets, Social Services should either modify its current agreement or require in the renewal of its agreement a method for tracking the time required to print and mail timesheets. Social Services should also perform monthly reviews of the activities performed by EDD and SCO to ensure compliance with the time frames for each agreement. Additionally, Social Services should implement a process to regularly test EDD and SCO processes to ensure that they are within the required time frames.

To more effectively address common problems reported by providers and recipients, Social Services should develop a formal process to document and address patterns of concerns conveyed through complaints. Specifically, the process should include a method for Social Services to identify and aggregate the complaints it receives, to analyze that information to determine whether there are common themes or broader issues to address within IHSS, and to obtain sufficient information to substantiate responses to the complaints. The process should also include steps to clarify ambiguous issues raised in the complaints and define clear deadlines and the steps to take when responding to complainants if those deadlines cannot be met.

Back to top

---

The Complexity of Certain Time‑Reporting Protocols Hinders Providers’ Ability to Accurately Complete Their Timesheets

Key Point:

State law requires that providers record and track work time both semi‑monthly and weekly. However, these time frames do not align for most of the year and create obstacles to providers’ ability to accurately plan or report their in‑home services.

State Law Requires That Providers Follow Two Conflicting Time‑Reporting Protocols

The IHSS program requires providers to adhere to conflicting time‑reporting protocols, resulting in the potential for providers to unknowingly exceed their number of authorized hours, which can jeopardize their continued employment in the program. Providers are required to follow certain rules that limit the number of hours they may claim in a workweek. State law defines the IHSS workweek as starting on Sunday and ending on Saturday. A provider must adhere to the workweek hour limits outlined in the text box to avoid receiving a violation from the IHSS program. For example, a recipient with a monthly authorization of 100 hours may initially ask his or her provider to work 25 hours per week. The recipient and provider may later agree that the provider will work six additional hours in an upcoming week, and then work six fewer hours in another workweek in the same month. In this example, no workweek hour limit has been violated because the provider would claim no more than 31 hours in a week and thus would not claim any overtime. Conversely, if the recipient and provider agreed that the provider would work 16 additional hours rather than six, the recipient would need to obtain county approval because the provider would claim 41 hours in that week, which includes one hour of overtime.

In another example, a recipient with a monthly authorization of 200 hours may initially agree with his or her provider on 50 hours of work per week. Under this scenario, the provider may not claim more than 10 hours of overtime in a workweek if doing so would cause the provider to ultimately claim more than 40 hours of overtime for the entire month, without the recipient first seeking county approval for an exception to the workweek maximum of 50 hours.

Adding complexity to the workweek issue, providers complete two timesheets each month for pay periods that generally do not coincide with the start of the workweek. State law establishes for IHSS providers a structure consisting of two pay periods per month. As specified by state regulations, one pay period starts on the first day of the month and ends on the 15th, and the other begins on the 16th and ends on the last day of the month. However, in most cases, the first day of a pay period does not coincide with the first day of the IHSS workweek. Specifically, during the 2016 calendar year, only two pay periods began on Sunday—one starting May 1 and the other starting October 16—while the other 22 pay periods started on other days of the week. This more frequent situation results in an additional burden, requiring the provider to track the amount of time worked in a workweek that crosses over two pay periods.

Figure 4 illustrates a situation in which a provider could inadvertently exceed the workweek limit because of confusion resulting from the workweek crossing over between two pay periods. In this example, the provider reported working 21 hours and 30 minutes during the four‑day period from June 12 to June 15, 2016. On the subsequent timesheet, the provider reported working 20 hours for the three‑day period from June 16 to June 18, representing the remainder of that workweek. Although the provider may have interpreted that each period constituted an individual workweek and that the hours reported adhered to the weekly limit of 35 hours, the provider may not have realized that the total time reported of 41 hours and 30 minutes—which covered the workweek from June 12 through June 18—exceeded the 40‑hour threshold outlined in the text box on the previous page, resulting in a violation.

Figure 4
Excerpt of Sample Timesheet Illustrating Misalignment Between Workweeks and Pay Periods

Sources: California State Auditor’s analysis of workweek and pay period rules defined in the Welfare and Institutions Code and the California Department of Social Services In‑Home Supportive Services (IHSS) provider timesheet.

* The IHSS workweek runs from Sunday through Saturday. June 2016 had two pay periods: June 1 through June 15 and June 16 through June 30.

Social Services’ policy pertaining to workweek limits states that a provider who receives three violations for exceeding workweek maximums will be suspended for 90 days from providing IHSS services, and a provider who receives a fourth violation will be suspended for one year from the IHSS program. To be eligible after a one‑year suspension, the provider would need to complete the enrollment process again, which we describe in the Introduction. According to the CMIPS II research and data analysis unit manager at Social Services, approximately 21,000 providers statewide received violations in the first six months after Social Services began issuing overtime violations in July 2016, of which nearly 3,200 have received their second or third violation.

The deputy director of the adult programs division informed us that Social Services is aware that recipients may be adversely affected if providers are suspended. The deputy director indicated that Social Services identified approximately 100 providers who had received a third violation in November 2016. In the interest of promoting compliance with the requirements, Social Services made a one‑time decision to remove the third violation from these providers’ records to avoid a suspension and also encouraged the providers to review instructional materials pertaining to workweek limitations. Although another nearly 50 providers were subsequently suspended after receiving a third violation in December 2016, the deputy director indicated that Social Services will continue to instruct counties to work with providers to educate them on minimizing their risk of being suspended.

Staff at each of the five counties we visited told us that understanding the relationship between workweeks and pay periods was difficult for many providers. Some of the counties we visited developed a series of small‑group seminars to assist providers in learning how to properly complete their timesheets. However, county staff stated that despite the availability of explanatory materials, timesheet seminars, and in‑depth phone explanations, many providers continue to struggle with this time‑reporting protocol.

To address these concerns, a coalition of county organizations and provider and consumer advocates created a proposal in February 2016 requesting that the workweek and pay period issue be simplified. The proposal pointed out that despite the collective efforts of stakeholder groups to educate providers on the implementation of overtime rules, the current time‑reporting rules were too complex for recipients and providers. The coalition recommended that the State change the pay periods to two‑week periods that would directly align with the workweek defined in state law. According to a chief deputy director, Social Services commented on the proposal at legislative hearings, and the Legislature ultimately decided not to take action on the proposal. Given the interest voiced by legislators and the approval of this audit by the Joint Legislative Audit Committee (Audit Committee), we believe it may be time for the State to reconsider this proposal to simplify the use of timesheets.

State law enacted in 2014 to implement the overtime rules required by the federal FLSA defined the workweek for determining and tracking weekly authorized hours and required that the monthly authorization of services for the recipient be converted to a weekly authorization.² Identification of the recipient’s weekly authorization is necessary to determine whether any of the hours a provider works within a workweek would be eligible for overtime pay and subject to any limitations. However, using monthly hours as a starting point to determine the weekly authorized hours results in complexities in determining the statutory limit on the number of hours a provider may work in a workweek. To facilitate compliance with state law, Social Services issued guidance to recipients to determine the limit by dividing the recipient’s monthly authorization by four. Social Services specifies that this maximum weekly amount is a guideline to ensure that the recipient receives the full amount of monthly services and to assist providers with properly reporting their hours on their timesheets. However, providers who use this standard conversion are susceptible to reporting more hours in a month than are authorized to the recipient. For example, a provider who works for a recipient who is authorized to receive 150 hours in services each month would receive notification from Social Services that the maximum number of hours per week would be 37 hours and 30 minutes. If the provider adhered to this limit by claiming those hours each workweek in the month of June 2016 (a 30‑day month), the provider would end up claiming 165 hours, or 15 hours more than the 150 hours authorized, because of the additional days beyond the four weeks in that month. Further, the provider would be paid only for the 150 hours the recipient was authorized to receive, meaning that 15 hours of work would go unpaid.

Moreover, we found that providers have historically claimed more hours on their timesheets than their recipients are authorized. Although overtime rules did not take effect until February 2016, the TPF identified more than 3.1 million timesheets during the period from November 2013 through July 2016 in which the hours claimed on a timesheet exceeded the recipient’s remaining hours authorized.

To address this issue, in conjunction with the workweek and pay period issue outlined here, the Legislature would need to amend state law to normalize the tracking and reporting of time worked on a weekly basis. We believe that, rather than using a monthly authorization of hours, a simpler approach would be for the county to establish a weekly authorization of hours for the recipient. Figure 5 presents an excerpt of a proposed revised timesheet that illustrates how providers would report the hours they worked using a weekly authorization on a timesheet that aligns the workweek and pay period. In addition, our illustration shows the spaces for the recipient and provider signatures moved from their current placement on the back of the timesheet to the front, which could result in the added benefit of a reduction in the number of exceptions pertaining to missing signatures that were overlooked because of their placement.

Figure 5
Excerpt of Sample Timesheet Illustrating Two‑Week Pay Periods

Source: California State Auditor’s analysis based on California Department of Social Services’ Standards Manual and Standard Arrears timesheet.

According to the chief deputy director at Social Services, recipient advocates historically supported monthly authorizations because they allow recipients the flexibility to adjust the timing of their providers’ hours within the month. State law specifically allows a recipient to request approval from the county to adjust the hours his or her provider works during a workweek as long as the recipient’s actual hours for the month remain within the total hours that are authorized. State law also permits recipients to authorize a provider to work hours in excess of the weekly maximums without notifying the county as long as the adjustment does not cause providers to work more than 40 hours per workweek or to exceed the recipient’s monthly authorization. However, we believe that same flexibility could be afforded to recipients under a weekly authorization structure. For example, state law could allow recipients to request approval from their county to move hours from one week to another, so long as the average number of hours per week during the two‑workweek pay period does not exceed their weekly authorization.

The chief deputy director of Social Services indicated that the cost of modifying CMIPS II to align the workweek and pay period and to use weekly authorizations would be significant. In April 2016, HPE developed a high‑level proposal that estimated the changes to CMIPS II to accommodate this alignment would cost between $34 million to $36 million to implement, noting that nearly all aspects of CMIPS II would be affected by the proposed alignment. Although this cost estimate does not include expenses for state operations to manage and communicate the changes throughout the State, we believe that the benefits to providers in making their timesheets easier to understand and complete would outweigh the costs in the long run.

Recommendations

To facilitate providers’ efforts to report their time, and to reduce the potential for providers to be inadvertently suspended from the IHSS program, the Legislature should amend state law to define the pay period as two workweeks. Moreover, the Legislature should modify state law to require weekly hours as the basis for authorizing services but continue to allow flexibility for recipients to adjust the hours their providers work across workweeks in a manner similar to the provisions of the current law. Until state law is changed, Social Services should inform providers of the weekly maximum number of service hours for each variation in the length of the month, rather than using a standard conversion that results in providers claiming more hours than their recipients are authorized.

If the Legislature amends state law as we recommend, Social Services should modify the timesheet format to incorporate the weekly authorization for services and the new two‑workweek pay period. Social Services should also reconfigure its timesheet to require that all information be entered on one side of the document, including the signatures of the provider and recipient.

Back to top

---

Other Areas We Reviewed

To address the audit objectives approved by the Audit Committee, we also reviewed and assessed timesheet protocols, information provided by Social Services and counties to aid providers with timesheets and payments, the method in which data on reported problems with CMIPS II were collected and resolved by Social Services, the current plans for improvements to CMIPS II, and the plans in place to address a breakdown in the CMIPS II infrastructure. Table 1 shows the results of our review of these issues.

Table 1
Other Areas Reviewed as Part of This Audit

Timesheet Protocols

We reviewed and assessed IHSS protocols for providers to report their time, including correcting errors on their timesheets, and we generally found that these protocols were reasonable. For example, providers are required to complete a timesheet using black ink because it results in the most readable scanned image of the timesheet at the TPF. Both the provider and recipient are required to sign and date the timesheet to declare that the information the provider entered is true and correct and to declare their understanding of penalties if they are convicted of fraud. Other protocols that appear reasonable include requiring providers to submit their timesheets no sooner than the last day of the pay period, to enter the hours and minutes worked in the spaces specified on the timesheet, and to mail the timesheet in the envelope provided by the State through the U.S. Postal Service first‑class mail option. Payment protocols for correcting errors include crossing out the erroneous entry and entering the correct information in the box, or requesting a replacement timesheet. Another protocol is that providers should not make any additional markings or use correction liquid or tape on their timesheets. Social Services generally instituted these payment protocols to ensure that timesheets could be quickly processed and paid using the scanning technology at the TPF. We determined that this rationale appears reasonable. We did find some protocols used to manage recipients’ authorized hours difficult for providers to understand and that, if misinterpreted, could jeopardize their continued status in the IHSS program. We discuss our assessment of these specific protocols beginning here and recommend how they could be improved here.

Information to Aid Providers With Timesheets and Payments

The State and counties present providers with three venues for seeking assistance with timesheets, payments, or technical problems: online, over the phone, or in person. Social Services includes on its website many resources, such as provider guides, educational videos, and contact information where providers can call to get additional help. Similarly, county websites also contain information for providers, including informational documents, videos, and contact information. As we describe here, some of the counties we visited developed a series of small‑group seminars to assist providers in learning how to properly complete their timesheets.

Tracking and Resolution of Reported Problems With CMIPS II

We were asked to determine how data on reported problems with CMIPS II are collected and, to the extent possible, identify the rate of complaint resolution. We determined that Social Services and OSI have a process for collecting service requests from users and a process for resolving those problems identified as system defects. OSI monitors the resolution of service requests from users through its monthly management reports from HPE. We examined a selection of service request data and determined that HPE generally resolved service requests from users within reasonable time frames. For the purposes of our audit, we reviewed service requests closed during a six‑month period from May through October 2016 that HPE identified as being related to payroll or timesheets and affecting a provider’s pay or eligibility, and found that HPE resolved them within five business days on average, with the majority being resolved in two business days or less. We also looked at the subset of those service requests that HPE determined to be system defects, which it tracks separately. HPE, OSI, and Social Services can also report defects internally. HPE ranks reports of defects in CMIPS II by severity, which assigns a target resolution time frame. OSI uses a weekly report to track the defects and the expected time frame for resolution based on the timing of new system releases. According to the CMIPS II project director, OSI reviews test results of its releases to verify that the corrections for these defects have been included in the releases. Another way to identify potential problems with CMIPS II is through written complaints that Social Services receives. However, as we describe starting here, Social Services lacks a formal process for reviewing the issues it receives through written complaints and addressing systemic problems with IHSS timesheets and payments. We recommend a solution here.

Current Plans for Improvements or Upgrades to CMIPS II

We were asked to identify and assess any current plans for improvements or technology upgrades to CMIPS II and whether those plans included timelines. Social Services currently has plans and a timeline to upgrade CMIPS II to include an electronic timesheet function, which will create the capability for providers to submit their timesheets online. Social Services anticipates that this upgrade will improve timesheet and payment processing times by reducing the number of exceptions. The optional online timesheet submission capability will be available alongside the current paper timesheet format. OSI’s CMIPS II project director stated that OSI plans to implement this function statewide by June 2017. Social Services’ staff also explained that they have been monitoring the federal 21st Century Cures Act, which became law in December 2016 and mandates a system of electronic visit verification for in‑home care providers under Medicaid. The 21st Century Cures Act defines electronic visit verification as a system under which visits—conducted to perform personal care or home health care services—are electronically verified with respect to the type of service, the recipient of the service, the provider of the service, the date and location of the service, and the starting and ending time of the service. In 2016 the State of Texas implemented electronic visit verification for home care services through the use of a telephone and computer‑based system that requires providers to use the recipient’s in‑home landline telephone or an approved small alternative device to log the start and end time of the service being performed. Although the deputy director of the adult programs division stated that Social Services does not yet have a formal plan for responding to this recent mandate, which will not take effect until January 2019, Social Services has been following this mandate through the legislative process and has identified several potential benefits for the IHSS program. For example, management at Social Services believe that electronic visit verification will provide real‑time submission for provider payroll information and potentially eliminate the need for paper timesheets.

Contingency Planning and Response

We were asked to determine whether there is a plan in place to address a breakdown in the CMIPS II infrastructure or process. In response to a May 2015 U.S. Postal Service delivery issue that delayed the TPF’s processing of thousands of timesheets, Social Services implemented two changes to CMIPS II. First, it developed the Daily Timesheet Processing Volume Report (timesheet volume report) intended to monitor whether the daily number of timesheets received from each county at the TPF is in line with past pay periods to more quickly identify delivery delays. Second, Social Services enhanced the functionality within CMIPS II to issue replacement timesheets in bulk when needed. However, Social Services has not been monitoring the timesheet volume report. According to CMIPS II design documents, the timesheet volume report will highlight any instances in which there is a decrease of 25 percent or more in a county’s volume of timesheets for a pay period from its average volume for the three previous pay periods. However, this report is produced only on demand and, according to OSI’s CMIPS II project director, the report does not automatically alert Social Services if the timesheet volume falls below the predetermined threshold. According to Social Services’ systems and administrative branch chief, Social Services has not developed procedures to monitor the daily timesheet volume report because its resources have been devoted to implementing changes from legislative mandates, such as the overtime requirements. We also reviewed the disaster recovery plan and procedures for CMIPS II and found that they appear reasonable to address infrastructure and process breakdowns. Recommendation To ensure Social Services can quickly identify potential concerns with the number of timesheets received at the TPF, it should develop procedures to review its timesheet volume report on a daily basis. Alternatively, Social Services could work with OSI to modify the reporting function within CMIPS II to require automated notifications to management when the timesheet volume report identifies an instance when the volume of timesheets falls below the threshold specified.

Source: California State Auditor’s analysis of the records identified in this table.

Back to top

---

Scope and Methodology

The Audit Committee directed the California State Auditor to conduct an audit of the timesheet and payment systems for IHSS providers, including the roles and responsibilities of state and local agencies with respect to those systems. The audit analysis the Audit Committee approved contained 10 objectives. We list the objectives and the methods we used to address them in Table 2.

Table 2
Audit Objectives and the Methods Used to Address Them

Audit Objective		Method
1	Review and evaluate the laws, rules, and regulations significant to the audit objectives.	Reviewed the applicable laws, rules, and regulations for each objective.
2	Identify the roles the various state agencies have in processing IHSS provider timesheets and payments.	Reviewed Social Services’ contracts with other state agencies to identify the roles of those other state agencies. Reviewed Social Services’ guidance to identify the roles of counties regarding IHSS provider timesheets and payments. Selected five counties—Los Angeles County, Sacramento County, San Bernardino County, San Diego County, and the City and County of San Francisco—for audit based on their number of IHSS providers and ratio of timesheet errors per provider. Reviewed contracts between the five counties we visited and their respective public authorities to identify the roles of the public authorities in processing provider timesheets and payments. Interviewed staff at Social Services, OSI, EDD, and SCO and reviewed related documentation to confirm our understanding of their roles in processing timesheets and payments. Interviewed staff at the five counties we visited and reviewed related documentation to confirm their roles in processing IHSS provider timesheets and payments.
3	Determine how CMIPS II is intended to function according to its original design and from the IHSS provider/user perspective.	Reviewed the OSI contract with the CMIPS II contractor, HPE, for the development of CMIPS II, and the CMIPS II system design documentation, project plans, request for proposals, and manuals. Interviewed staff at Social Services, OSI, and EDD to confirm our understanding of the system’s design, development, and functionality. Interviewed staff at the five counties we visited to understand functions of the system from the user perspective.
4	Assess the rationale for payment protocols that includes a review of the following:
	a. The options available to a provider to ensure they receive timely payment after realizing they have a timesheet error.	Reviewed Social Services’ training materials directed to IHSS providers to develop a listing of all payment protocols, including those for correcting errors. Also reviewed Social Services’ All County Letters, policy manual, and training materials for county staff. Determined the rationale for payment protocols through discussions with Social Services’ staff and reviews of documents, including design documents from CMIPS II. In anticipation of California’s planned implementation of alternate methods of reporting hours worked, we identified other states with similar in‑home services programs and interviewed staff at those states or reviewed their websites to understand the performance of electronic timesheets and electronic visit verification. Interviewed staff at the counties we visited to determine whether they have established any additional payment protocols. Reviewed supporting documentation to substantiate interview discussions. Interviewed county staff to identify the options available to providers to correct timesheets when providers realize their timesheets have errors.
	b. The reasonableness of the length of time providers must wait before reporting a lost or stolen paycheck.	Interviewed staff at Social Services and each of the five counties we visited to determine the process for providers to report a lost or stolen paycheck. Reviewed the State Administrative Manual and SCO’s Payroll Procedures Manual regarding the statewide process for reporting lost or stolen paychecks issued by SCO. Used information available from the U.S. Postal Service to determine the approximate mailing time for paychecks from SCO to providers throughout the State and assessed the reasonableness of Social Services’ policy based on the average mailing time. As part of Objective 8(a), we also reviewed data to determine the length of time it takes for SCO to process paychecks for all timesheets.
5	Identify whether any significant updates have been made to protocols or information technology to improve CMIPS II since its original design and launch, and determine whether any required changes to the system have led to improvements or caused new problems.	Reviewed CMIPS II contract documentation, system design documentation, project plans, and record of changes to determine updates to CMIPS II that were significant contract and design changes. Interviewed staff at Social Services and OSI to confirm our understanding of significant updates to CMIPS II. Interviewed staff at the five counties we visited to identify updates to CMIPS II protocols or information technology that were significant to county users, and how they affected counties and providers. Reviewed a selection of change requests to determine the reasons for the changes and their effect on the program. Our review identified several significant changes to CMIPS II, which we discuss in the Introduction. These changes to CMIPS II include those related to the recent implementation of overtime rules in statute that now allow providers to be paid overtime, but also affect the way providers report the hours they worked. We address these changes here.
6	Assess how well the system is working for those state agencies that interact with IHSS provider timesheets and payments. In doing so, determine whether new, completed, rejected, and replacement timesheets are being distributed in the required amount of time.	Reviewed stakeholder meeting documentation for concerns about timesheets and payments. Interviewed staff at Social Services, EDD, and SCO to determine how CMIPS II is working for state agencies that use data from CMIPS II to print and mail timesheets and paychecks. We address Social Services’ management over the methods in which timesheets and paychecks are distributed here. Interviewed staff at the five counties we visited and 10 additional counties we contacted by telephone to assess how well CMIPS II is working for staff and to determine how new and replacement timesheets are distributed. These additional counties are Alameda County, Butte County, Fresno County, Humboldt County, Imperial County, Lake County, Orange County, Riverside County, Santa Clara County, and Santa Cruz County. Generally, these 15 counties indicated that CMIPS II either worked well or was an improvement from the original CMIPS, but expressed an interest in new reporting functions within CMIPS II and options for providers seeking status information about their timesheets or paychecks. We address the issues related to replacement timesheets here.
7	Identify the options or tools that are currently in place to assist IHSS providers with timesheet, paycheck, and technical problems. To the extent the data is available, identify the utilization rates of these tools.	Reviewed online resources from Social Services and the five counties we visited. Interviewed Social Services and county staff to identify additional options and tools currently in place for providers and to clarify our understanding of how the counties make use of statewide tools. Where available, reviewed county‑level utilization rates of options and tools for providers. In general, county staff informed us that they do not track utilization of their resources because detailed tracking is not needed or the effort involved in tracking would not be cost‑effective.
8	For the period of time since the implementation of CMIPS II, perform the following:
	a. Identify the magnitude of reported problems, such as the number of timesheets that resulted in delayed payment, which have been documented or reported to the applicable departments.	Analyzed data from CMIPS II for the period from the implementation of CMIPS II in November 2013 through July 2016. Specifically, we reviewed the length of time it takes for the TPF to process timesheets and for SCO to issue paychecks; the number of lost, stolen, or damaged paychecks; and for the five counties we selected, the average processing time of payment corrections issued to resolve timesheet exceptions and the average processing time for timesheets with exceptions that were resolved by replacement timesheets.
	b. Determine how data on reported problems with the system are collected and, to the extent possible, identify the rate of complaint resolution.	Reviewed the tracking log for letters or complaints received by Social Services’ executive office and the CMIPS II branch for timesheet‑ and payment‑related issues and randomly selected 10 cases for further analysis. Analyzed the selected letters or complaints to determine the nature of the problems, how long Social Services took to address those letters or complaints, and the steps Social Services took to resolve each letter or complaint. Reviewed CMIPS II design documentation and contract to determine how service request information is collected and tracked. Analyzed data obtained from OSI to determine the average length of time it takes to resolve service requests and to assess trends in those service requests and their resolution. Interviewed staff at Social Services, OSI, and HPE to understand how they manage service requests. Reviewed data on system defects provided by OSI to determine the extent to which OSI was monitoring the length of time it took HPE to resolve them.
9	Identify and assess any current plans for improvements or technological upgrades to CMIPS II. In doing so, perform the following:
	a. Determine the timeline for any such plans.	Interviewed staff at Social Services, OSI, and HPE to identify any current plans for improvements or upgrades to CMIPS II. Reviewed change requests related to electronic submission of timesheets and identified the scheduled implementation timeline.
	b. If applicable, determine the reasons why there are no plans to improve identified problems.	We found this objective not applicable, as we identified plans for improvements and upgrades that respond to problems identified by users.
	c. Determine whether there is a plan in place to address a breakdown in the CMIPS II infrastructure or process, such as the May 2015 missing timesheet incident, to ensure that providers can still be paid and have access to the timesheets for the next pay period.	Reviewed Social Services’ contingency response plans to determine whether sufficient plans are in place for infrastructure or process failures. Assessed whether changes made to CMIPS II in response to the May 2015 U.S. Postal Service delays were sufficient to address potential future delays.
10	Review and assess any other issues that are significant to the audit.	Visited the TPF and EDD printing and mailing facility to develop an understanding of the processes for receiving and processing timesheets, and for printing and mailing timesheets.

Sources: California State Auditor's analysis of the Audit Committee's audit request number 2016‑128, the planning documents, and analysis of information and documentation identified in the column titled Method.

Assessment of Data Reliability

In performing this audit, we obtained electronic data files extracted from the information system listed in Table 3. The U.S. Government Accountability Office, whose standards we are statutorily required to follow, requires us to assess the sufficiency and appropriateness of computer‑processed information that we use to support findings, conclusions, or recommendations. Table 3 describes the analyses we conducted using data from this information system, our methods for testing, and the result of our assessment. Although this determination may affect the precision of the numbers we present, there is sufficient evidence in total to support our audit findings, conclusions, and recommendations.

Table 3
Methods Used to Assess Data Reliability

INFORMATION SYSTEM	PURPOSE	METHOD AND RESULT	CONCLUSION
Social Services’ timesheet data as maintained by the OSI CMIPS II as of August 21, 2016	To determine the number of timesheets and paychecks as well as to calculate various statistics related to processing times for the period from November 2013 through July 2016.	We performed data‑set verification and found no errors. Further, we performed electronic testing of key data elements and did not identify any significant errors. To gain some assurance of the completeness of the data, we traced a haphazard selection of 29 timesheets to the data and found no errors. We did not perform accuracy testing on these data because the system is a partially paperless system. Alternatively, we could have reviewed the adequacy of selected system controls that include general and application controls, but we determined that this level of review was cost‑prohibitive.	Undetermined reliability for this audit purpose. Although this determination may affect the precision of the numbers we present, sufficient evidence exists in total to support our audit findings, conclusions, and recommendations.

Sources: California State Auditor’s analysis of various documents, interviews, and data from the entities listed in this table.

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.

Respectfully submitted,

ELAINE M. HOWLE, CPA
State Auditor

Date:
March 16, 2017

Staff:
Linus Li, CPA, CMA, Audit Principal
Vance W. Cable
Terra Bennett Brown, MPP
Michael Henson
Kurtis Nakamura, MPIA

IT Audits:
Michelle J. Baur, CISA, Audit Principal
Lindsay M. Harris, MBA, CISA
Richard W. Fry, MPA, ACDA
Reed Adam, MPAc

Legal Counsel:
Scott A. Baxter, Sr. Staff Counsel

For questions regarding the contents of this report, please contact
Margarita Fernández, Chief of Public Affairs, at 916.445.0255.

---

Footnote

² Although the Legislature passed SB 855 in June 2014 to apply the federal requirement, state law was not implemented until February 2016 due to delays caused by court challenges to the federal overtime rules. Go back to text

Back to top