Improving Quality of Primary Care for Anxiety Disorders



Improving Quality of Primary Care for Anxiety Disorders

1. Specific Aims

The specific aims of this study are to:

a) Adapt and integrate existing patient information and self-management materials for PD and GAD into a locally modified program that will enhance patients’ understanding and ability to self-care for their anxiety disorder;

b) Construct an evidence-based EMR algorithm to direct PCPs in the treatment of PD and GAD that is based upon the APA’s anxiety treatment guideline and several other recently published evidence-based treatment algorithms;

c) Perform a randomized clinical trial to determine the 12-month clinical outcomes and treatment costs for 246 primary care patients who are informed that they are experiencing PD and/or GAD along with their PCPs. Patients will be randomized to either: (1) receive additional information about their anxiety disorder, be offered an information/self-management program for treatment of PD/GAD with assistance from a facilitator over the telephone, and be cared for by PCPs who are exposed to our PD/GAD treatment algorithm via EMR; or (2) patient and PCP notification of the PD/GAD diagnosis alone (“usual care”).

Our primary hypothesis is that 60% of patients diagnosed with PD and/or GAD who are randomized to our intervention arm will experience a 50% reduction of anxiety symptomatology and an improved functional status four months following the diagnosis, coinciding with the end of the acute phase of treatment for an anxiety disorder, compared with 35% of patients randomized to “usual care.”

Our secondary hypotheses are that compared to our “usual care” group, intervention patients will: (1) obtain treatment for PD and GAD that is more consistent with accepted guidelines; (2) be less likely to relapse to baseline levels of distress following recovery at eight and twelve months following study entry; (3) increase their knowledge and understanding of anxiety disorders; and (4) achieve at least a similar clinical outcome at a lower average medical cost.

2.0 Background and Significance

Anxiety disorders are prevalent in primary care practice and generate a significant excess of morbidity and health care expenditures.1-3 Unfortunately, they are often unrecognized and ineffectively treated by PCPs who care for the majority of individuals suffering from these psychiatric disorders.4-6

Among the various anxiety disorders in primary care practice, PD and GAD create the largest burden of morbidity. They have a point prevalence of 1-13%7-11 and 5-16%,8-10 respectively, and have a similar adverse impact on self-reported quality of life.12-13 Unfortunately, patients with PD and GAD experience poorer than expected clinical outcomes despite the availability of efficacious treatments that PCPs can provide.9, 14-17 A clinical practice guideline for treating anxiety disorder14 along with several algorithms for PD15, 18, 19 and GAD15, 20 have been developed to improve care. Nevertheless, experience with other guidelines suggests they are unlikely to improve practice if disseminated using traditional methods.21

Patients who perceive themselves as active participants in their medical care pose questions, elicit treatment options, and state preferences during their physician encounters. As a result, they are more likely to receive appropriate medical care 'and experience better health outcomes.22-24 Yet most efforts to change clinicians behavior have not directly involved patients.25 Information self-management programs for PD/GAD can be particularly useful as they familiarize the patient with the illness and its characteristics, treatment options, and possible coping skills. Patients may then probe their physician's clinical reasoning, request certain procedures or medications, or refuse to comply with a recommended test or treatment. Although self-management strategies have been effective for such diverse medical problems as asthma,26 diabetes,23 and hypertension,27 just one randomized trial has evaluated a self-help strategy for treating an anxiety disorder in a primary care setting.28

EMR systems which prompt clinicians to perform certain patient-specific actions can improve the frequency at which recommended medical services are provided and the quality of clinical care.29-34 Recent developments in the capabilities of commercially available EMRs combined with reductions in their cost provide opportunities to implement practice guidelines and rapidly transfer new scientific findings into clinical care. However, little is known about their effectiveness, acceptability, and economic impact in primary care settings, particularly within the domain of mental health.

Despite numerous reports of the distinct effectiveness of patient information/self-management strategies and of delivering patient-specific treatment advice to clinicians via EMRs, little is known whether combining these interventions can improve the quality of primary care for PD and GAD, and produce clinical benefits exceeding those achieved by either intervention alone. PD and GAD are ideal conditions to study within this framework as they are: (1) prevalent in primary care; (2) responsible for sizable amounts of excess morbidity and health services utilization; (3) poorly recognized and managed by PCPs; and (4) effectively treated within primary care settings. Our findings will enhance our understanding of meaningful methods to disseminate CPGs that improve the quality of medical care.

0. Research Design and Methods

1. Drug/Device Information

No new drugs, medical devices, or placebos will be used for this study.

3.2 Research Design

Our protocol is a randomized controlled effectiveness study of the effect of disseminating clinical care guidelines for GAD/PD to both patients and their physician compared “usual care”.

3.3 Methods

Study recruitment will occur at Falk Clinic and Partners In Health-UPMC, two primary care practices of the University of Pittsburgh Medical Center. Patients will be randomized to either the intervention group or the “usual care” group, as displayed in Figure 1, following confirmation of protocol eligibility (see 4.2). Sixty-percent of patients will be randomized to our intervention arm, while the balance will be randomized to receive “usual care” to help ensure sufficient sample sizes of patients are available to test our secondary hypotheses should a sizable number of intervention PCPs disagree with the diagnosis of GAD/PD on the PRIME-MD. An overview of our intervention by group is shown in Table 1.

Figure 1. Patient Randomization Scheme.

Protocol-eligible primary care patients will be informed that they are experiencing a pattern of symptoms consistent with GAD/PD : (1) orally by the research assistant obtaining their informed consent; (2) in a mailed letter from the investigators requesting the patient to follow-up with their PCP regarding the symptoms; and (3) over the telephone by the research assistant performing the baseline assessment. Protocol-eligible patients must consent to allow us to inform his/her PCP of finding PD/GAD on the PRIME-MD to enroll in our study. This is key so their PCP: (1) can be exposed to our anxiety treatment guidelines per the patient’s randomization status (see Figure 1) upon agreement with the PRIME-MD; and (2) is aware the patient may be receiving our information/self-management program. If the patient does not provide permission for us to inform their PCP of the symptoms found on the PRIME-MD, we will not do so against their wishes. Still, regardless of whether a protocol-eligible patient decides to enroll, we will strongly encourage him/her to discuss the results of their PRIME-MD with their PCP as it may have clinical implications for the treatment of any health problems the patient may be experiencing, and we will provide the patient with a general brochure on anxiety disorders for their review such as the ones developed by the Anxiety Disorders Association of America () or the NIMH. Physicians will be informed of the PD/GAD symptomology of patients who enroll in our study through both an e-mail notice and a letter sent to them via the EMR by the investigators. Neither PD nor GAD will be entered into a patient's electronic medical record until the PCP first agrees with the diagnosis (see 4.3).

Subjects who are randomized to the treatment arm of the investigation will receive mailed information about their anxiety. Later, a trained facilitator will call these subjects to follow-up on the mailing and offer an information/self-management program for PD/GAD in either audiotape, videotape,36 or workbook form.37 If the subject accepts the materials, the facilitator will mail them to the subject in the desired format and call back every other week to review the materials. If the subject declines the materials, the facilitator will offer to call back again in two weeks to answer any new anxiety related questions that may arise and offer the materials again. If the subject declines this offer, the facilitator will not call the subject back again. We expect the facilitator will contact subjects who accept the self-management about 5-6 times every other week with each call lasting approximately 10-30 minutes.

|Subjects |“Usual Care” |Intervention |

|Notified of the PD/GAD diagnosis |( |( |

|Provided: |( |( |

|Information on PD/GAD | | |

|Phone #s of local mental health resources/support groups | | |

|Question list (FAQ) to ask PCP about anxiety d/o | | |

|Contacted by self-management program facilitator to: | |( |

|Follow-up on mailing | | |

|Offer anxiety self-management program | | |

|Receive follow-up call(s) from facilitator (total 1-5) | |( |

|Receive reinforcement phone call by facilitator ~2 months after | |( |

|completing acute phase anxiety Tx. | | |

|Anxiety symptoms/progress (w) self-management program fedback to PCP | |( |

|via EMR by facilitator | | |

|Physicians | | |

|Trained in Anxiety Guideline (PD/GAD) |( |( |

|Trained in use and interpretation of PRIME-MD |( |( |

|EMR notification of the PD/GAD Dx. |( |( |

|PD/GAD entered on subject’s problem list |Upon PCP agreement |Upon PCP agreement |

|EMR reminder of PD/GAD Dx. At time of clinical encounter on: |Problem list |Problem list, Encounter form, |

| | |& on line |

|Counseling/educational messages provided | |( |

|at time of subject encounter | | |

|Guidance re: anti-anxiety pharmacotherapy | |( |

|Pt. Instructional handout generated at start of Tx. | |Suggested by Protocol |

|Guidance re: follow-up appointment interval | |( |

|Internet access to additional information on |Upon Request |Suggested by Protocols |

|PD/GAD and guidelines for MHS referral | | |

|Notified of potential drug-drug interaction |Automated |Automated |

|Receive progress note from self-mgt. Program facilitator | |( |

Table 1: Intervention Overview by Group

PCPs whose subjects are randomized to the intervention arm and who agree with the PRIME-MD findings of PD and/or GAD will be presented with information both on-line and on the encounter form printed at each patient visit. Advice will be based upon the American Psychiatric Association's treatment guideline for anxiety disorders14 and several algorithms for PD15, 18, 19 and GAD15, 20 that we will convert to electronic form prior to starting subject recruitment. The telephone facilitator will also provide written feedback to the PCP via EMR on each intervention subject’s progress with the self-management program and suggestions for care by the PCP (e.g., management of side effects from anxiolytic pharmacotherapy) if the facilitator uncovers a problem. These suggestions will be based upon our treatment algorithm and reviewed by the medical and psychiatric co-investigators at routinely scheduled weekly monitoring meetings.

No research visits to the health center will be required of subjects beyond their routine care. Direct research assessments of enrolled subjects will only occur over the telephone using the standardized instruments in the schedule as outlined in Table 2 (see Appendix A for a description and copy of each).

|Assessment Instruments: Month: |0 |2 |4 |8 |12 |

|Sociodemographic Characteristics |( | | | | |

|Symptom Measure | | | | | |

|Hamilton Rating Scale for Anxiety (HRS-A) |( |( |( |( |( |

|Functional Status Measures | | | | | |

|Subjective Symptoms Scale |( | |( | |( |

| MOS SF-12 |( | |( | |( |

| Usual Activities Questionnaire |( |( |( |( |( |

| Global Assessment Scale |( |( |( |( |( |

| Panic Disorder Severity Scale |( |( |( |( |( |

| Generalized Anxiety Disorder Scale |( |( |( |( |( |

|Belief Measures | | | | | |

|Anxiety Sensitivity Index |( | |( | |( |

| Satisfaction with Anxiety Care | |( | |( |( |

| Subjects’ Attitudes and Expectations |( | | | | |

| Participatory Decision Making Style |( |( | | |( |

|Recovery Modifiers | | | | | |

|Major Depression | | | | | |

|PRIME-MD PQ/Mood Module |( |( |( |( |( |

| Hamilton Rating Scale for Depression |(‡ |(‡ |(‡ |(‡ |(‡ |

| MOS Social Support |( | | |( | |

|Adherence with Medical Recommendations |( | |( | |( |

| Traumatic Life Events |( | | | | |

|Medical Co-Morbidity |( | | | |( |

|Chronic Disease Score | | | | | |

|Health Care Utilization |( |( |( |( |( |

|Utilization of Health Care Questionnaire | | | | | |

|Chart review | | | | | |

‡ Administered only if patient screens positive for a mood disorder on the PRIME-MD

Table 2: Subject Assessment Instruments

We anticipate each baseline assessment will last 25-40 minutes and subsequent assessments will be 10-30 minutes in duration. Indirect assessments and validation of patients' medical comorbidity and health services utilization will be performed using chart reviews. We will ask subjects to sign a separate release form indicating their consent for us to obtain non-practice medical records over the course of the subject’s enrollment in our study.

We will collect self-reported information on PCPs' sociodemographic and practice characteristics at baseline. We will also ask PCPs to complete a brief questionnaire to determine their baseline knowledge and attitudes towards anxiety disorders and the EMR. Three years later, at the conclusion of our intervention, we will re-survey PCPs on these items. Those physicians exposed to the EMR reminders and the self-management program will be surveyed about their experiences with them. We estimate both assessment batteries will take PCPs approximately 20-30 minutes to complete.

3.4 Data Collection and Statistical Considerations

The primary dependent measure of recovery for the sample size calculations is a 50% decline from the baseline Hamilton Rating Scale for Anxiety (HRS-A) score four months following study recruitment. Outcomes among study patients will be analyzed by the group to which the patient was randomized (Figure 1). The null hypothesis is that there will be no difference in recovery among the groups. If we enroll 20 PCPs, compensate for the dependence of observations nested under each PCP, and experience a 15% lost-to-follow-up rate, we will need 248 patients to detect a: (1) 0.25 difference in proportions between the 2 groups ((=0.05, and (=0.20); or (2) ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download