Overcoming Barriers to Good Outcome: Non-Compliance...

Patients may comply less well with their prescribed antidepressant regimens than they do with treatments for their physical disorders.¹

What follows is an algorithm for addressing a variety of non-pharmacologic barriers to good response in the acute pharmacologic treatment of depression. Recognition and management of non-compliance is one factor which receives extended consideration. Non-compliance is also an issue in maintenance treatment and this section may be helpful for that, as well. Other causes of frequent relapse are addressed in a separate "Place to Go" entitled "Frequent relapses..."

¹Cramer JA, Rosenheck R. Compliance with medication regimens for mental and physical disorders. Psychiatric Services 1998;49:196-201.

uestion: Is the patient taking the medication as prescribed?

• No: non-compliance is strongly suspected
• Yes: compliance is assured (or, choose this answer if you have addressed compliance problems and are looking for additional barriers that could be contributing)

Help: Suspect non-compliance or partial compliance if any of the following apply:

1. Patient is missing appointments
2. Patient is actively abusing/dependent on alcohol or other substances
3. Patient reports no side effects at all
4. Patient reports huge amount of side effects which clearly outweigh the benefits occurring or expected, yet denies non-compliance
5. Patient is not refilling medication when prescriptions are scheduled to require refill [Ask the patient to bring their medication containers to appointments in order to review dosing instructions. This can be helpful to detect forgetfulness as a cause of non-compliance, but if patient is actively and secretly resisting following instructions, there will be resistance to bringing in the containers! It may become clear that the patient never even had the prescription filled.]
6. Patient indicates inability to afford the cost of the medication
7. Patient does not accept/lacks insight into the diagnosis of depression, denies need for treatment, or has not expressed clear desire to improve
8. Patient has secondary gain from staying depressed (e.g. - compensation issue)
9. Patient resisted idea of medication when first proposed, and sees use of medication as stigmatizing for some reason

uestion: Has sufficient time been spent with the patient to assess subjective response to the medication, and address concerns about side effects?

• Yes
• No

Comment: The education process must be ongoing and ideally should involve the family.¹ Use of a multidisciplinary team approach results in the best compliance, with oral and written information given by nurses, pharmacists, and psychotherapists supplementing and reinforcing the information provided by the physician.² When the prescriber is a primary care physician working in a general medical practice, the above is much less likely to occur.³ Physicians may spend even less time with patients who are in pre-paid insurance plans vs. fee-for-service plans.³

¹Frank E. Enhancing patient outcomes: treatment adherence. J Clin Psychiatry 1997;58[suppl 1]:11-14.

²Katon W et al. A multifaceted intervention to improve treatment of depression in primary care. Arch Gen Psychiatry 1996;53:924-932.

³Meredith LS et al. Counseling typically provided for depression: role of clinician specialty and payment system. Arch Gen Psychiatry 1996;53:905-912.

Recommendations:

Consider some of the following factors which may contribute to poor outcome

1. Review diagnosis. Is it DSM-4 criteria based? Remember that it must be, in order to use this algorithm.
   1. Have medical conditions been ruled out as cause of the depression?
   2. Is there a severe, acute precipitant (so that this is an Adjustment Disorder?)
   3. Is there an undetected, occult psychotic disorder?
   4. Consider reviewing the algorithm for Treatment Resistant Depression in the main flowchart. This will consider Axis II personality disorders and Axis I comorbidity. Normally, the main depression algorithm calls for at least two "adequate" trials of antidepressaants before it routinely reviews Axis II and comorbidity, but you may want to look at the decision tree before that.
2. Be sure that there is no ongoing active alcohol abuse or dependence. This is associated with incomplete recovery in antidepressant-treated patients.¹ The amount of impact on outcome is proportional to the quantity of alcohol use.²
3. Is the patient currently receiving adequate concomitant psychotherapy³ to address psychosocial and stress factors contributing to the depression? Or, is the patient receiving brief "medication visits" only?
4. If the patient is receiving psychotherapy, does the patient have a positive alliance with the therapist? Could there be a "negative therapeutic reaction" wherein the patient is angry or upset with the psychotherapist and this is undermining medication treatment outcome? Although the quality of the alliance predicts treatment outcome with medication, placebo, or psychotherapy, it is unclear if good alliance can be actively fostered or whether it is a pre-existing patient characteristic in the relationship with the clinician. (4) Therapist (as opposed to patient) perception of the quality of the alliance seems to predict outcome with pharmacotherapy. (5)
5. "Split therapy:" Is it a problem in this case? Or -- is it needed?

¹Akiskal HS. Factors associated with incomplete recovery in primary depressive illness. J Clin Psychiatry 1982;43:266-271.

²Worthington J, Fava M, Agustin C, et al. Consumption of alcohol, nicotine, and caffeine among depressed outpatients: relationship with response to treatment. Psychosomatics 1996;37:518-522.

³Hollon SD, Fawcett J. Combined medication and psychotherapy. In Gabbard G, ed. Treatment of Psychiatric Disorders. Washington, DC: American Psychiatric Press, 1995:1221-1236.

(4)Krupnick JL et al. The role of therapeutic alliance in psychotherapy and pharmacotherapy outcome: findings in the NIMH Treatment of Depression Collaborative Research Program. J Consult Clin Psychol 1996;64:532-9.

(5)Weiss M, Gaston L, Propst A, et al. The role of the alliance in the pharmacologic treatment of depression. J Clin Psychiatry 1997;58:186-204

uestion: What is the patient's understanding of the causes of their depressive illness?

• Patient attributes depressive syndrome to a simple explanation like "not getting enough sleep," or to stress, e.g. - "working too hard." This irrational or misinformed explanation might support non-compliance with the prescribed medication regimen.
• Patient has good understanding of depressive illness and accepts diagnosis.

Comment: Lack of insight is very common and has been established as a factor in treatment non-compliance in patients with schizophrenia.¹ Although it has received less attention in depression, the patient's "explanatory model" of the illness is probably a significant factor in non-compliance with treatment.² The clinician should elicit the patient's self-formulation and look for irrational beliefs.

¹Dickerson PF et al. Lack of insight among outpatients with schizophrenia. Psychiatric Services 1997;48:195-199.

²Sperry L. Psychopharmacology and Psychotherapy: Strategies for Maximizing Treatment Outcomes. New York, Brunner/Mazel, 1995:50-51.

Recommendation:

Sufficient time should be spent to define the illness, describe target symptoms for medication, suggest expected time course of response, review side effects of medication, sequence of doses and what to do if problems arise, and encourage the patient to ask questions. Patients should be given a patient information sheet about the medication to take with them. Review all the side effects listed on the sheet during the session and refer back to it in subsequent meetings if necessary. An appointment in one week is recommended after starting antidepressant medication, to check for possible deterioration, early side effects, or other reasons for early non-compliance.¹

Sexual side effects seem to be underemphasized by clinicians. They occur in as high as 75% of patients taking SSRI antidepressants² such as fluoxetine (Prozac), sertraline (Zoloft), and paroxetine (Paxil), and have been described as the "Achilles Heel" of these medications. Patients are often not clearly informed about them, and even if they are informed, they may be reluctant to bring them up with their physician. They must contribute to non-compliance more than has been generally appreciated.

See under Places to Go: Dealing With Side Effects.

¹Fawcett J. Compliance: definitions and key issues. J Clin Psychiatry 1995;56[suppl 1]:4-10.

²Patterson WM. Fluoxetine-induced sexual dysfunction. J Clin Psychiatry 1993;54:71

Recommendation:

The patient may need more counseling and negotiation to resolve the differing formulations of the problem. The clinician should offer a more comprehensive model of the factors (biological and psychosocial) that are contributing to the depression and how different treatments address the different factors. This should be tailored to the patients experience by incorporating some aspect of the patient's explanation if possible. For example, if the patient believes that insomnia is the cause of everything, the patient should be told that inability to sleep is, in fact, an important symptom of the depression syndrome, and improvement in sleep would be a good indicator of when the depression is starting to improve. Then, sleep would be monitored closely.¹

¹Ward N. Psychosocial approaches to pharmacotherapy. In Beitman B, Klerman G, eds. Integrating pharmacotherapy and psychotherapy. Washington, DC: American Psychiatric Press, 1991:69-104.

uestion: Does the patient have a complicated medication treatment regimen that could be simplified?

• Yes
• No

Comment: Even if the patient is only on one, relatively uncomplicated pharmacotherapeutic agent, he/she may be on other non-psychiatric medications which make the overall task of keeping them organized rather difficult. This is especially common in the elderly, where non-compliance rates have been reported to be as high as 75%.¹ Patients with organicity, attention-deficit disorder, or long and demanding work schedules often forget or confuse their medication schedules. Actually, it is probably the exceptional patient who is compulsively reliable with oral medication compliance.

¹Salzman C. Medication compliance in the elderly. J Clin Psychiatry 1995;56[suppl 1]:18-22.

Recommendation:

To help patients with complicated medication regimens, the following might be tried:

1. Give written instructions.
2. Use pill boxes organized weekly or, better, monthly, with clear instructions about what to do if a dose is missed.
3. Reduce the number of pills/capsules per day by giving larger size pills. Keep in mind that certain medications have a maximum dose that may be given at any one time: eg - bupropion (Wellbutrin) 150 mg, clozapine (Clozaril) 450 mg. Consult the Physician's Desk Reference or package insert.
4. Reduce the number of times per day that the medication must be taken. Some medications are often given several times a day when this is unnecessary because the half-life of the drug is 24 hours or more. Examples of this are tricyclic antidepressants and neuroleptics. Note that bupropion (Wellbutrin) is now available in a sustained-release preparation and may be given twice a day, as opposed to three times a day for the standard form.
5. Organize medication ingestion around daily routines, such as meals.
6. Some complicated, but suboptimally effective pharmacotherapy regimens evolve from well-motivated efforts, but simpler approaches might have been passed over which may work better. A review of the accompanying Algorithm for the Pharmacotherapy of Depression from the beginning may be helpful.

uestion: Does the patient have significant others who are not supportive of the pharmacotherapy? (e.g. - spouse, parents, AA sponsor)

• Yes
• No

Comment: "Not supportive" could mean overt opposition ("There's nothing wrong with you, you don't need that crutch.") However, it could also mean excessive (albeit well-motivated) reminders ("nagging") that stimulates oppositional behavior. Sometimes pill-pushing by significant others is motivated in part by their own resistance to accepting any responsibility for contributing to the stress in the patient's life that is fueling the depression. The patient may feel that taking the medication is equivalent to admitting that the problem is all theirs and that the significant other does not need to change in any way.

Recommendation:

Consider having a family compliance counseling session.¹ Explore what the significant other's understanding of the illness is, and what they think the patient's needs are. Psychoeducational work may be helpful.² Sometimes, more complicated and multidetermined undermining of the treatment plan may be going on, which may require a series of meetings to address.

¹Sperry L. Psychopharmacology and psychotherapy: Strategies for meximizing treatment outcomes. New York: Brunner/Mazel, 1995:56-58.

²Frank E. Enhancing patient outcomes: treatment adherence. J Clin Psychiatry 1997;58[suppl 1]:11-14.

uestion: Does the patient have certain personality traits that might call for a particular individualized approach to improving compliance? [This is the last question in the compliance algorithm]

• Dependent
• Compulsive/paranoid
• Passive-aggressive/hostile-dependent

Comment: A paper by Ward was the source of the suggestions found in the links to the three trait categories listed.¹

¹Ward P. Psychosocial approaches to pharmacotherapy. In Beitman BD, Klerman GL, eds. Integrating pharmacotherapy and psychotherapy. Washington, DC: American Psychiatric Press, 1991:69-104.

Suggestion:

For dependent patients: These patients are generally compliant but if they are not, they may hide this fact so as to not displease the clinician, family members, or significant others. Supportively explore possible reasons for non-compliance: explanatory model, cost, side effects, etc.

Suggestion:

For the compulsive/paranoid patient: These patients require more extensive discussion and reassurance than others. They particularly appreciate being offered different treatment choices or alternatives so they can actively make choices and feel in control. Allow more options and go over contingencies in advance when presenting the titration schedule.

Suggestion:

Passive-aggressive/hostile-dependent patients: These patients are particularly prone to poor compliance. They require detailed explanations of the medication, indications, target symptoms, side effects, etc., and yet they may still refuse to try it on a minor point. The best approach is to be very clear about the limitations of drug therapy. Avoid excessive claims for benefits and undue reassurance about side effects. These patients report more side effects than other patients, and this should be responded to with sober professionalism. Sometimes a "counterprojective approach" of slightly exaggerating the problems with medication and anticipating their negative appraisal of the medication may allow them to feel more comfortable with going ahead with an adequate trial.