[Main Page of Report | Contents of Report]
In this chapter, we report on why MCOs collect personal health information, how they use it, what types of information are commonly collected and how this varies across plans, and the various ways in which the information is collected.
The most common reason for collecting personal health information is to support utilization review. In this review, an MCO determines the medical necessity of the request and the appropriate level of care. Typically, the client makes the initial request for treatment. The client generally speaks to a care manager at the MCO, who discusses the nature of the problem and the symptoms and makes a referral to a provider for the minimum level of care deemed appropriate (Edwards 1997). Once this initial authorization is exhausted, the provider must request authorization again if the patient continues to need treatment. The process for requesting re-authorizations varies from company to company; some plans conduct reviews by telephone, usually following a prepared set of questions, while others require the provider to fax a treatment request form to the company. The frequency of re-authorizations also varies. Some plans require re-authorization every two to three visits, while others may authorize 10 or more outpatient sessions at a time (Hennessy and Green-Hennessy 1997).
There are also differences in the authorization and re-authorization processes for outpatient versus inpatient treatment requests. There is some indication that MCOs may be moving away from requiring extensive information as part of utilization review for outpatient treatment. A number of recent empirical studies have shown that intensive utilization management of outpatient cases may not be cost-effective for managed care firms. One study found that it is 50 percent more expensive to administer managed care than fee-for-service (Meyeroff and Meyeroff 1999). Another study found that the majority of patients receiving outpatient behavioral health treatment voluntarily terminated treatment after a limited number of sessions (Hennessy and Green-Hennessy 1997). These authors suggested that MCO efforts to manage care do not appear to have had a significant impact on overall outpatient utilization and that MCOs might find that it is not cost-effective to intensively manage all cases.
We pursued this issue in interviews with providers and managed care firms, asking them whether they had observed a trend in MCOs requiring less patient information. Most confirmed that some MCOs seem to be requiring less patient information for utilization review. Magellan has recently introduced a treatment request form that requires only very basic patient information. In addition, several plans we spoke with have instituted interactive voice response (IVR) systems, through which a provider calls an MCO and provides basic patient information—such as demographics, diagnosis, and services requested—into an automated system. The treatment is automatically approved as long as the request meets certain basic parameters. Case managers review a small sample of the cases from the IVR system.
Some plans are also requiring patient information less frequently than in the past. One plan we spoke with has, within the past year, decreased the frequency of their reviews from every 10 sessions to every 20 sessions for psychiatric treatment and from every 20 sessions to every 40 sessions for substance abuse treatment.
One provider believes managed care firms are requesting less information partly in response to provider and patient pressure but also because plans are beginning to find that the costs of hands-on management through authorizations are not worthwhile relative to the cost of treatment because most patients only need short-term treatment. This view is consistent with the experience of one managed care plan we spoke with, which stated that the firm has reduced the amount of patient information it collects because “99 percent of cases are managed fine” without the plan having to manage each one. However, several providers pointed out that not all MCOs have streamlined their requests.
Although MCOs collect the patient information contained in outpatient treatment requests (OTRs) primarily for utilization review, many MCOs also use this information for quality management. Several MCOs we spoke with use the information submitted in OTRs to identify outlier cases in which the diagnosis appears to warrant more extensive treatment than what is being received. Examples include a patient with schizophrenia who is not on medication or a patient who is actively suicidal but for whom appropriate levels of care have not been indicated. OTRs are also used to track patient progress. For example, if a person with an adjustment disorder has been in treatment for several years with no apparent improvement, the MCO would want to flag the case and then call the provider for an explanation. One MCO stated that it hopes to have to follow up on no more than 10 to 15 percent of cases. Procedures for quality management differ at one staff-model HMO, where the provider’s supervisor and other authorized personnel in the behavioral health department randomly review charts to ensure that appropriate care is received.
MCOs do audits primarily to make sure that clinicians are actually performing the services for which they are billing. In addition, several providers stated that MCOs may need to review records in order to comply with accreditation requirements such as those developed by the National Committee for Quality Assurance (NCQA).
An MCO might also request a medical record because of quality-of-care concerns, whether expressed by patients, other providers, or other sources. One MCO in our study also reviews the full record when a patient chooses to go to an out-of-network hospital. All of the providers we spoke with stated that the MCOs include in their provider contracts the right to access the full medical record at any time. MCOs may audit the provider’s chart system on site or they may simply ask the clinician to send in a sample of charts.
MCO requests for complete charts can be problematic because, in practice, many therapists do not separate psychotherapy notes from the general medical record. These notes reflect the therapist’s thoughts and opinions during treatment and may also contain information on patients’ family members who probably had not agreed to have their information disclosed to the therapist, let alone the MCO. Once the information is in the chart, anyone at an MCO that handles the chart, including data clerks, could have access to that information. Releasing sensitive information in charts can have serious consequences because an MCO clerk could be required to testify in court as to what he or she saw in a chart. To illustrate the severity of this possibility, one provider used an example of a patient who was a physician being treated for substance abuse. The physician had been writing fraudulent prescriptions and consuming the drugs himself. If this is recorded in a file and an MCO clerk later sees it and reports it, the physician could lose his medical license and face criminal charges.
Several providers mentioned that they do not keep separate charting systems for physical and mental health care because of the administrative hassles. Two providers said explicitly that they do not separate their notes even though they know they should because they have never been audited. If they were to be audited, they would pull out the notes before the MCO came on site.
Case Management. MCOs may also use personal health information for case management and care coordination purposes. MCOs may assign case managers to patients who use a high volume of services, to help them coordinate care between providers and to help them access community services. Case managers may also be “go-to” people that can be contacted in a crisis, or between visits to the therapist. The use of personal health information for this purpose is far less controversial, especially among consumers. Consumer advocates we spoke with generally support the use of patient information for case management and care coordination as long as the patient approves it. One advocate stated that patients distinguish between the sharing of information within and outside the health care system and feel comfortable with information being shared with people such as case managers or clinicians when their roles are to facilitate or participate in treatment.
Care Coordination. It is often beneficial to the patient for information to be shared by the mental health/substance abuse provider and the primary care provider, particularly information that could prevent drug interactions (Simmons 1997). Plans we spoke with generally ask the patient to sign an authorization for treatment information to be shared with the primary care provider. MCOs do not so much collect this information as facilitate communication between providers. Such care coordination is especially common when the mental health/substance abuse provider and the primary care physician work in the same clinic in a staff-model HMO. In some cases, the providers may be able to share records electronically, further streamlining the process.
One mental health care provider we spoke with works with an MCO that automatically shares treatment information with the patient’s primary care provider. She feels this is unnecessary, as some of her patients may have no relationship with their primary care provider, so the information is being shared with a stranger. She would rather that the decision to share information with a primary care physician be made on a case-by-case basis. Most providers we spoke with, however, said that in their experience, patients do sign an authorization for this information to be shared.
Most of the providers we spoke with agreed that there is a great deal of variation in the amount of patient information requested by MCOs for outpatient authorizations. The one exception, a provider who works only with Medicaid managed care firms in her state, has not observed much variation among the plans. The other providers agreed that there is variation from company to company, and even within companies, depending on the type of contract an MCO has with an employer and on state laws that may restrict the types of patient information that can be shared with managed care firms.
To determine what information is shared between providers and payers, we asked both if they would be willing to share copies of their outpatient treatment request forms and telephone review protocols. We collected a total of 10 forms and one telephone protocol. From these, we were able to identify a number of topics that are fairly standard in treatment authorizations and others that vary considerably from company to company. The documents we collected include:
The Maryland Uniform Treatment Plan Form.This form is used to collect the only patient health information that insurers can routinely collect in Maryland, per state law. The form went into effect in October 2000. Self-insured(ERISA-exempt) plans are exempt from this requirement (Appendix B).
· The Magellan Treatment Request Form. Magellan is a national managed behavioral health care organization with an enrollment of approximately 70 million people. Magellan adopted the form in August 2000 (Appendix C).
· The ValueOptions Outpatient Treatment Report. ValueOptions is a national managed behavioral health care organization that manages services for over 23 million people. The ValueOptions Outpatient Treatment Report is available on the firm’s website at www.valueoptions.com/provider/forms.htm.
· Two forms used by other national managed care firms but not publicly available.
· Five forms from small or local managed care firms. Two of these plans serve primarily Medicaid populations.
· One telephone protocol used in at least one market by a large managed behavioral health care organization.
We created a list of the information that was requested by the managed care plans and recorded the frequency with which each item occurs in all the forms or protocols we reviewed (Appendix D). Items that occur in at least six of the forms or protocols are identified in Table II.1. Information on whether the item is requested in a categorical format (usually checklists or yes/no questions), in narrative form, or in both is also included in the table.
While we found wide variation in both the amount of information collected and the processes for collecting that information, we also found some similarities across plans. All plans ask for administrative data, including the patient’s name, date of birth, social security or insurance identification number, and identifying information for the practitioner. Most plans ask for the DSM-IV diagnosis code, including axis five, the Global Assessment of Functioning. Treatment information, including the requested procedures or types of services, the frequency and duration of treatment, and expected outcomes are also fairly standard. Most plans also ask about the patient’s current medications and compliance with the regimen. Finally, many plans ask for information about the practitioner’s coordination with the primary care provider and about the patient’s involvement in other community services.
Some respondents believe that the information requested varies by the type of plan. We reviewed whether the types of information requested varies by whether an MCO does or does not carve out behavioral health (Appendix E). While our data are limited, with only four examples from MCOs and seven from MBHOs, there do not seem to be any differences in the types of patient information requested by the two types of MCOs.
| Requested Informationa | Categorical | Narrative | Both |
|---|---|---|---|
| Demographic/Administrative Information | |||
| Patient’s name | 11 | ||
| Patient’s date of birth | 10 | ||
| Patient’s social security/insurance ID number | 10 | ||
| Practitioner’s name, address, phone | 11 | ||
| Practitioner’s license and/or ID number | 10 | ||
| Initial authorization or continuing | 6 | 1 | |
| Length of treatment/start and end dates | 8 | ||
| Diagnosis | |||
| DSM-IV diagnosis code | 11 | ||
| Current Global Assessment of Functioning (GAF) | 10 | ||
| Highest GAF in past year | 8 | ||
| Patient History | |||
| Previous MH/SA treatment | 2 | 4 | 2 |
| History of substance abuse | 2 | 5 | 1 |
| Presenting Problems | |||
| Symptoms | 7 | 2 | 2 |
| Duration and severity of symptoms | 6 | 1 | |
| Risk assessment suicide/homicide | 5 | 3 | |
| Current substance abuse | 5 | 1 | 3 |
| Family/social relationships | 4 | 2 | 2 |
| Job/school performance | 4 | 2 | 2 |
| Obsessions/compulsions | 4 | 2 | |
| Treatment Information | |||
| Requested procedures/types of services | 8 | 1 | 2 |
| Frequency/duration of treatment | 4 | 7 | |
| Expected treatment outcomes | 4 | 5 | |
| Member notified/concurs with goals? | 5 | 1 | |
| Medications | |||
| Current medications | 2 | 8 | |
| Dosage/frequency | 1 | 7 | |
| Compliance | 3 | 3 | |
| Care Coordination | |||
| Communication with PCP | 6 | 1 | |
| Patient receiving other community services | 4 | 3 | 2 |
| aItems are included in table if they were listed in 6 of 11 examples studied. | |||
We also reviewed whether local MCOs request different types of patient health information from the national firms for which we had information (Appendix F). Two providers noted that, in their experience, local MCOs tend to ask for less detailed information than do national firms. One said that the likely reason for this is that the local MCOs are more familiar with her clinic and have a closer working relationship with the providers. Again, this analysis is limited because we have examples only from five national managed care firms and six local firms. However, there do not seem to be any differences in the types of patient information requested by local or national MCOs. However, beyond these forms, MCOs request follow-up information on certain cases in an informal manner, and it is possible that less follow-up information is requested if a provider develops a strong working relationship with an MCO. Stronger working relationships could, in turn, be easier for providers to develop with local MCOs.
Two of the outpatient treatment request forms we examined were geared specifically toward substance abuse treatment. There were too few of these forms to do a separate analysis, but in comparing them to the others, we found that the only difference is that these two forms do not ask about the patient’s risk of suicide or homicide. Otherwise, information requested in these forms does not differ from information requested for general behavioral health. In addition, two of the forms we looked at were from Medicaid managed care plans. These forms ask for the same types of patient information as commercial managed care plans.
Information on patient history varies considerably from plan to plan. Three plans do not ask for any information on the patient’s history or previous treatment. Several plans ask whether the patient has received treatment for mental health and/or substance abuse; some plans provide the clinician with a checklist of treatment types (i.e. outpatient, partial hospitalization or inpatient), asking the clinician to indicate which ones the patient had received. Three plans ask whether the patient has other family members also receiving treatment; two simply require a yes/no response, while the third asks the provider to provide descriptive information about personal and family history relating to mental health and substance abuse disorders.
Information collected on a patient’s current status and presenting problems also varies greatly by plan. Three of the outpatient treatment forms we examined asked the provider to explain the patient’s current problems and to describe the plan for addressing each one. Other plans provide a checklist of symptoms and ask the provider to indicate which ones the patient has experienced, and in some cases, to also indicate the severity and duration of the symptoms. Some plans also include a separate checklist for level of functioning in such areas as family and social relationships, work/school performance, physical health, sexual functioning, legal problems, financial situation, and activities of daily living.
There is a great deal of variation among plans in the lists of symptoms they ask providers about. The Magellan Treatment Request Form (Appendix C) is the shortest list used by the plans we studied. Magellan simply lists four symptoms: self-injurious behavior, suicidal ideation, homicidal ideation, and substance use problems; the form also requests information on the severity of each (mild, moderate, severe). The Maryland Uniform Treatment Plan (Appendix B), which requests the most detailed information of the forms we studied, includes a checklist of 56 symptoms. It also asks the provider to rate the patient’s level of functioning (mild, moderate, severe) in six areas: family relations, job/school, finances, physical health, legal, friends/social. The ValueOptions Report lists 24 symptoms and asks for information on their duration. It also includes a checklist for level of functioning in 12 areas and asks the provider to rate the severity level on a scale of one to five and to estimate the severity level of each at discharge. A comprehensive list of the symptoms requested by all plans is included in Appendix D along with the number of plans requesting information on each.
Plans also vary considerably in their questions regarding the level of risk of harm to self or others. Three plans do not ask for any information on this topic. Notably, both of the forms geared specifically towards substance abuse do not ask for information on this topic. However, the majority of the plans request information on the patient’s risk of suicide and homicide, asking the provider to indicate whether the patient has exhibited ideation, a plan, or intent with or without means. Two plans request information on other risk behaviors as well, including items such as self-injury, fire setting, family violence, and psychosis. These two plans also ask the clinician to record any additional risk behaviors.
Providers generally request treatment authorization by telephone or by submitting a written form. However, as mentioned in Chapter I, several MCOs have recently implemented interactive voice response (IVR) systems for outpatient treatment authorizations. The provider calls into an MCO and supplies basic patient information—including the patient name, social security number, diagnosis, and services requested—to an automated system. How the system is used varies from company to company. For two of the MCOs we spoke with, the system automatically gives the provider an authorization number for the services; the only reason for a denial would be if the member or provider is not eligible. Master’s level case managers then pull reports off the system and review them retrospectively to ensure that services are being used appropriately. The case managers therefore only need to review a limited number of cases, not every case. (A third MCO requires the provider to supply clinical information to the system, and the approval is granted within five days.) Compared with standard treatment request forms, the IVR system provides somewhat more privacy. Because there is no need for a data clerk to enter information into a database, the systems eliminate the need for an additional person to see patient information.
Despite the growing popularity of IVR systems, most providers requested authorizations for services by completing paper treatment request forms or by speaking to a case manager over the telephone. The providers we interviewed differ somewhat in terms of whether they prefer sharing information over the telephone or in writing. Two providers believe that phone conversations are more intrusive. One said that a form allows providers to clearly state only the necessary information, whereas in a telephone conversation, the case manager might be more likely to ask for additional information. Another provider believes that clinicians might be more likely to reveal more information than they intend to when they are on the telephone. However, a third provider prefers telephone conversations because she feels she has more control over what she says, telling case managers what she thinks they need to know without revealing anything she feels is irrelevant.
Another concern that providers raised regarding telephone reviews is that the MCO staff taking the calls may not be sufficiently trained in mental health and substance abuse treatment, making them less-than-responsive, in the providers’ eyes, to requests for authorization. A number of providers said that it is frustrating to give information to a clerical staff person who simply reads from a script and enters the information into a computer. One provider described a situation in which a patient was actively suicidal and under supervision until an ambulance came, during which time the MCO staff member was reading through a set of questions on the patient’s hygiene that were not relevant to the case. However, at some plans, the case managers are master’s-level clinicians who are knowledgeable about treatment. One substance abuse provider at an inpatient clinic has interacted with case managers who are already familiar with the clients before they enter her facility and take an active interest in the treatment.
Several providers expressed concern regarding the treatment request forms, notably about the security of faxing this highly sensitive information. One provider noted that an MCO she works with asks that forms be faxed without a cover page. Another provider recalled a case in which an MCO had given out the wrong fax number, so information was inadvertently sent to a private residence. Still another provider mentioned that she always calls the MCOs after she faxes forms to make sure that they are properly received.
A few providers mentioned that MCOs are increasingly accepting records electronically. A representative of a large national managed behavioral health organization that has a number of Medicaid managed care contracts said that, in some states, doctors work with electronic medical records that feed directly into the managed care plan’s system. Several providers we spoke with have strong concerns about the security of transmitting confidential patient information in this manner. One provider said that her attorneys have advised her not to transmit records electronically until greater security measures are in place. Another stated that patients should be informed if their medical records are being transmitted in this manner.
Once a provider submits a treatment authorization request, an MCO case manager may call the provider to ask for further clarification. A great deal more information may be shared as a result. It is not clear how often this occurs, as providers and MCOs were not able to give precise figures. As mentioned in Chapter I, one MCO said that it hopes to have to follow up only on 10 to 15 percent of outpatient cases.
If a request for treatment is denied, the patient and clinician have the right to appeal. The appeals process varies from plan to plan. The initial appeal may take the form of a telephone conversation between the clinician and a doctor on staff at the MCO. If the two are unable to reach an agreement on the course of treatment, the case will go to a second round of appeals. At this stage, plans generally invite the clinician to submit the patient’s full medical record. In general, to pursue their appeal, providers need to submit it.
One provider stated that, in lieu of releasing the entire record in the second round of appeals, he can sometimes prepare a summary of additional information that the plan needs. However, other providers we spoke with said that, in their experience, plans always require the full record in order to review the case. One provider mentioned that when he calls an MCO, he usually speaks with a clerical person, not a psychiatrist or psychologist. Since that person does not know what information will be required for the appeal, the provider is simply told to send everything.
We did not review the process for inpatient authorizations systematically, but some providers in the study have worked with both inpatient and outpatient treatment requests and said the two are very different. Inpatient authorizations are much more intrusive, probably reflecting the fact that most of the costs in behavioral health are incurred on the inpatient side. Processes for inpatient reviews vary considerably from plan to plan but, in general, consist of telephone discussions between hospital staff and MCO case managers. Reviews occur frequently, sometimes every day or every couple of days. The questions are usually open-ended and may be tailored to the specifics of the case. In some cases, the MCO case managers are very familiar with the patient’s history and may suggest treatment strategies. At one MCO, case managers may even visit the facility in person to meet with the patient and providers. If an MCO issues a denial, there is an appeals process similar to that for outpatient treatment: a first round with a doctor-to-doctor review and a second round in which the MCO may request the full medical record. One MCO said that inpatient cases reach the second round of appeals more frequently (about one to five percent of the time) than outpatient cases do.