TREATMENT DETAILS (revised 8/2005) 

Primary care management

Destabilizing factors

More on mood stabilizer use

Mood stabilizing without medications

SPEAK: guiding patients to mood-improving behaviors

Most primary care physicians I have met with do not regard themselves as doing "psychotherapy". Yet the most basic elements in "psychotherapy" are all present in your routine interactions: the presence of a caring individual who listens empathically and wants to help. These are the core ingredients (e.g. as defined by the research of Rogers and Truax in the 50’s) in therapeutic interactions. You are "doing therapy" whether you intend to or not! Psychotherapists use some specific tools as well, but most research indicates that the "technique" used is much less important than these core ingredients which are common to all "psychotherapy" modalities.

As you may know, there are three types of psychotherapy which have been shown in randomized trials to be more effective than placebo. In mild to moderate depression, they have been shown to be as effective as medication, and possibly more likely to maintain benefits over time.Frank and Thase These are cognitive, behavioral (often combined), and interpersonal psychotherapy.

Most of you feel you could do more if time allowed. When you able to allot the time, here are some specific areas you can focus on, from each of these three techniques (organized by John Christensen, Ph.D., who teaches behavioral health skills in the Internal Medicine Residency Program at Oregon Health Sciences University; used with gratitude). The presentation of each skill is from my patient’s site: you can use the same language with your patients:

  • S = Schedule (activities) — Behavioral
  • P = Pleasurable Activities — Behavioral
  • E = Exercise — Behavioral
  • A = Assertiveness — Interpersonal
  • K = Kind Thoughts About You — Cognitive
  • Schedule: make a plan now and have a calendar or log for behaviors you intend to pursue. Here is a sample log you can change to suit you:

    Day Time of Day Activity planned With whom? Where? Planning needed
               
               
               
               

    When you make a log like this, and keep it up to date, you have a good record of exactly what you have done for your own health. If you are trying to increase your activity, this log could show you whether you are making progress! Even if you can see a small increase, this can be a big "boost", to see that you are able to move yourself forward. It may give you the motivation and energy to move further yet.

    Pleasurable: as many people experience loss of pleasure due to their mood state, they can stop arranging for activities that used to provide enjoyment, leading to a vicious downward spiral. Many people can reverse this trend, though it’s slow at first, by deliberately scheduling pleasurable activities. In the beginning you’ll have to force yourself, and it won’t be fun like it used to. But with some repetition, you might notice a hint of improvement, and if you do, take advantage of that by trying to do a little more. That’s how some people turn the spiral around, into a slowly improving one.

    Exercise may have a mood stabilizing effect in bipolar disorder. You are probably aware of the many health benefits of regular physical activity.Phelps I tell patients, "if these benefits could be put in a pill, everyone would be taking it!". Obviously the hard part is getting yourself to do it.   If you have doubts about your ability to get regular exercise, please have a look at Exercise and Mood: not the usual rap on this site.  

    Make a plan and a chart like this:

    Day Time of Day Type of exercise How Long? Intensity 1-5 How I felt
    before (1-10)
    After
    (1-10)
                 
                 
                 
                 

    Start with just a little bit. Walking is usually a good starting place. Rate your mood on a scale of 1-10, where 1 is the worst you get, and 10 is the best, before and after. See your primary care physician if you have been very inactive, before you begin, for an okay and more guidelines.

    Assertiveness means asking for what you want. When two people are in conflict over something, four things can happen:

    1. I win, you lose.
    2. You win, I lose.
    3. We both lose.
    4. We both win.

    Assertiveness means trying to get what you want, where hopefully the other person still gets what he/she wants. There is a lovely book about negotiating skills by Harvard experts on this topic,Fisher and Ury but there are also many "self-help" books about this in your local bookstore. Many people with mood problems have lost, or never had, good assertiveness skills. A good therapist (e.g. MSW, LCSW, Ph.D.) can often help a lot with this.

    Kind thoughts about you: many people with mood problems can be extremely unrealistic in their view of themselves. They minimize their skills while magnifying their weaknesses; inflating small errors into major gaffs; overlooking their successes, and so forth. "Cognitive therapy" tries to help people be realistic in their view of themselves, calling attention to such "cognitive errors". In most cases, this eventually means treating oneself as one would likely treat others: with kind thoughts, rather than harsh negative assessments. At one level, this is "think positive", but at a more sophisticated level, it is "think fair/realistic/evidence-based". A good book about cognitive techniques is Feeling GoodBurns (a friend of mine says "great book, terrible title"). If you’re really motivated, you may be able to work through this book on your own. It costs less than an insurance co-pay ($6.00 on line, last look). This book contains all the ingredients of good cognitive therapy (written by a master). However, most people do this kind of work with a therapist who knows the technique.

    When should I refer a patient?

    Most bipolar patients are complicated enough to warrant referral as soon as you suspect this diagnosis. This site was created in recognition of the fact that many primary care physicians do not have ready access to a psychiatrist for such referrals, and that even when referred some patients may not go -- as many as 50%, in one recent report.Katon In this circumstance, trying valproate and lithium may be warranted. If you are able to establish that these agents do indeed help, but the patient is still struggling, referral is again indicated. Hopefully your patient will be more willing to go now, and the risk of the diagnosis being "missed" by a psychiatrist who has not adopted the broader criteria for bipolar disorder will be lower (if this "diagnosis-advocacy" concerns you, see the "bipolar bandwagon" section).

    How can I make referring a patient easier?

    Many communities will have much more ready access to psychotherapists than psychiatrists. But doctors and patients both dislike the "1-800-here’s-your-therapist" system many insurance companies advocate for referring patients to mental health providers. In our area we are working on a creating a table of providers, their specialities, and the insurance panels on which they serve. The hope is to return to a system where a doctor knows a few particular therapists she likes working with, and makes referrals directly to them (knowing, from the table, which of her favorites can take the insurance a particular patient has).

    Perhaps you could make one for your area? Most therapists seem delighted to hear from primary care physicians wanting to re-establish a person-to-person referral system. Invite some of the therapists about whom you have heard good things to come to your office for lunch, and record their data in a table, e.g.:

    THERAPIST PHONE SPECIALTY AREAS INSURANCE PANELS
    Cindy Aron, M.S.W. 757 5235 Insight oriented group therapy

    Individual adult therapy

    Eating disorders

    HMO-O; PacificSource
    Valerie Dougher, M.S.W 757 5235 Children and adolescents

    Family therapy

    Individual adult therapy

    HMO-O; PacificSource
     

     

         
     

     

     

         

    How do you manage a suicidal patient in a primary care office?

    The simplest answer: believe that the patient is safe for further outpatient followup, or hospitalize. This means: obtain a believable story from the patient that he or she will be safe until seen again. If in doubt ask for a verbal commitment to stay safe, insuring that it is plausible (e.g. eye contact, body language, tone of voice all match during her statement). If still in doubt, discuss your intent to pursue immediate assistance (mental health evaluation), and if in doubt about her ability to wait for this, call 911. For details on (a) determining intent, and (b) risk factors associated with suicide, read on.

    Determining intent
    This is difficult and even trained mental health personnel cannot judge suicide risk well. Davis,Ayd Your patient tells you she is thinking a lot about suicide (ideation). Your first step in handling this is to determine her level of intent: are these just thoughts, or is there a motivation to act on these thoughts as well? One way to characterize level of intent is to ask about plans: has she thought about how she would take her life? If the answers are detailed, a question about means is warranted: does she actually have the tools she describes using? For example, if she says she’ll overdose, what pills is she thinking of using for this? Does she have ready access to these? Is there a gun she can obtain?

    There is no simple "test" to establish level of intent. The above questions will give you some sense of this. Patients who plan to use a gun, antifreeze, carbon monoxide or other lethal means should be regarded as manifesting serious intent unless it can clearly demonstrated otherwise. But since over-the-counter pills can be lethal also, the decision regarding what kind of intervention is required will ultimately depend on your overall sense of the patient’s immediate intent. Subdued energy, unmodulated tone of voice, lack of eye contact, no intent to inform anyone, plans for a remote location — all suggest higher risk.

    If the patient has thoughts of suicide but no plan, or only vague plans and no momentum toward obtaining means; and if he can clearly indicate intent to follow up with you or a mental health professional; and if he can give believable assurance (does this patient really have the capacity to offer assurance?) he will be safe until that follow up date: then you may consider outpatient management. Document your assessment of ideation, intent, plans, and means.

    Such a patient can sometimes be helped by reminders that thoughts are not dangerous by themselves, only when actions are taken based upon them. For a lovely essay on this point, see this site.Metanoia When patients can see their thoughts going into action, they need to contact emergency services. If they can state their ability and intent to so make contact, that is usually a strong indication of safety (not conclusive, but helpful if consistent with the rest of her/his presentationMiller).

    Somewhere along the way in this process you may feel very strongly that the patient is safe. If not, it’s time for an immediate intervention. In my area here are the options:

      a) Call the county mental health crisis worker if the patient can safely wait in your office:

      • Benton 541 757 6844
      • Linn967 3866
      • Lincoln265 4179
      • Marion503 581 5535

      b) If the patient is not safe waiting, call 911 for an immediate intervention by emergency services.

    Risk factors associated with suicide
    For an excellent, thorough review available on line, see Ayd and Palma. Age (geriatric or teenage), alcohol, living alone, unemployment, and a history of recent loss are common but far more common than suicide itself, of course. The relative rarity of suicide makes prediction extremely difficult, as no factors demonstrate specificity.

    Anxiety has been implicated as one of the most predictive risk factors for suicide.Fawcett(b) Bipolar in its "mixed state" includes both severe depression and a high energy state. Patients in this state endorse the term "agitation".

    Agitation has so far been 100% sensitive (though of course with very low specificity) for suicide risk in my 5 years of full-time practice: two of my patients have committed suicide, and both were in a bipolar mixed state. Both had decided against taking medication and could not be persuaded to change their minds, and were not delusional or overtly suicidal at the time of this decision. The level of agitation was very high for both patients.

    Door #1, Door #2: The "Let’s Make a Deal" analogy of medication change risk

    For people with only minor signs of bipolar disorder, or some improvement on a medication, but not enough, I almost always invoke the old TV show "Let’s Make a Deal" (remember Monty Hall? "oh, Monty, Monty!"). In it, participants were offered the option of giving up what they had already won, to take some unknown prize. It could be a Cadillac, or a pile of newspaper. Participants were not able to "go back" to what they had given up for door #1 or door #2.

    Similarly, some people may not be able to go back to the gains a previous medication has offered. They may move to a different illness stage in the process (for more on this, see "Does bipolar disorder get worse over time?"). They should make the choice of a new medication approach with this risk in mind. Obviously, to take this risk, something really needs to be "broke" and require "fixing"!

    Are steroids destabilizing in bipolar disorder? (revised 6/3/00)

    The risk of precipitating mood instability (most commonly mania) with steroid medications has not been quantified, but has been clearly described and repeatedly observed.Brown and Suppes There are reports of a testosterone patchKline and dihydroepiandosterone (DHEA, available in health food stores)Weiss inducing manic episodes.  Obviously I don’t get to see the many patients who have done fine on predisone for their poison oak, but I have seen multiple patients who have had severe manic episodes when given steroids by their primary care physician.

    Avoid giving a bipolar patient steroids if at all possible, and if you must, be in communication with the patient’s managing psychiatrist. You don’t want to be on your own in this situation.

    Whereas in bipolar I the response to steroids can be full manic episodes (e.g. my very proper patient over 65 who went to the brothels in Nevada after oral steroids for arthritis), in bipolar II the net effect is a destabilizing influence which is less overt or destructive but can be severely compromising nevertheless (e.g. my patient who just could never quite get stable until he stopped his steroid eye drops!).

    What about steroid inhalers for asthma, or steroid nasal sprays for allergies?  I have never seen a report in the psychiatric literature about these as a problem in bipolar disorder.  However, just recently (6/00) I have begun to seriously wonder.  At a recent conference on allergies I learned that these types of steroids are well known to enter the bloodstream.  In children they can cause growth retardation, so we know for sure that they can have systemic effects.  In my view that's probably enough to merit worrying about bipolar disorder too, and I'm going to start watching these medications closely in patients whose bipolar disorder does not stabilize with routine mood

    How much alcohol is ok?

    Zero would be nice. However, suggesting total abstinence forever, when the patient is still coming to terms with having a "mental illness", seems more than patients can usually "hear". I have found it very effective to suggest that patients try 1-2 months with no alcohol at all, so that they can then evaluate how destabilizing alcohol is for them personally, and in what amounts. Most seem to find that about 1 drink per week does not affect them, but that more than one at a time, or more than one per week total, does. They must also be aware that the mood stabilizer medications will often make that one drink substantially more powerful; warn them not to drive after alcohol while on medications until they are very familiar with the impact alcohol is having upon them.

    A patient who is not stabilizing as expected must try for abstinence until she/he improves; i.e. zero alcohol is a must for patients with continued mood problems, at least until their mood improves (and cautiously after that).

    What are other destabilizing influences in bipolar disorder?

    You have already encountered the major recognized destabilizing factors: antidepressants, alcohol, and steroids. Now for the more subtle ones.

    Sleep deprivation is pro-manic, even in patients without "bipolar disorder" per se. You know that feeling you get the middle of the day after a night up on call? The giddy, loose, hilarity/on the edge feelings? And how a few hours later it’s degenerating into irritability, disorganization, agitation? Take those feelings and multiply by 1 to 100, and you have most of the spectrum of hypomania.

    For people with bipolar variations, night after night of too little sleep is clearly part of the whole problem. Generally, when people get better they sleep much better. In fact, I use sleep as one of the best markers of whether a patient has enough mood stabilizer: when they do, they will sleep well (6-8 hours, unbroken or able easily to go back to sleep if awakened). If they still need some sleeping medication, that’s a warning sign that we still don’t have enough lithium/valproate type effect. Once a patient is finally sleeping well on these medications, they can use poor sleep as a marker for trouble: when they start to sleep less than 6 hours, it’s time to watch closely for any other symptom of bipolar disorder, and increase the mood stabilizer if such a symptom shows up.

    Travel crossing time zones creates automatic potential for sleep change: deprivation going east, additional time going west. One study reports mania precipitated by eastward travel, and depression by westward travel.Young Major travel within the same time zone (e.g. my Oregon patient who became manic in Mexico) may be destabilizing due to stress alone.

    Stress alters many nervous system chemicals, including corticotropin releasing hormone (CRH) which releases the body’s own steroids.Arborelius Reproductive steroids such as estrogen and progesterone respond to stress; for example, women may stop menstruating when severely stressed. Serotonin, the mood-related neurotransmitter, also clearly changes with stress.Duman All of these brain chemicals influence mood.

    Stress is part of modern life (some would say more so in recent centuries?). Achieving a stress free life is not realistic for anyone (one could wonder what that would look like). People with bipolar disorder can be expected to show mood instability even during positive stresses such as moving into a new house or accepting a new job, let alone after trauma or losses. So while lowering stress levels may help maintain mood stability, people with bipolar disorder need to have enough mood stability from their medications to be able to handle at least some stresses.

    In bipolar I, the first episodes are commonly associated with a severe stress, whereas later episodes often appear "out of the blue", i.e. with no apparent stressor to bring them on. This raised the concern that somehow earlier episodes were making subsequent episodes easier to trigger. For more on this, read the next section.

    Does bipolar disorder get worse over time? (the "kindling theory") (revised March 2000)

    In bipolar I, there are clearly many patients whose susceptibility to subsequent episodes increases with time, as shown in this man’s pattern:

    This observation led to the so-called "kindling theory". The idea of "kindling" is based on the finding that a region of mouse brain repeatedly exposed to small electric shocks will eventually start to have spontaneous seizure-like electrical events. That is, repeated episodes seem to make subsequent episodes more likely to occur spontaneously. This is precisely the pattern observed in bipolar I.

    Three out of five of the currently recognized "mood stabilizers" are anticonvulsants (valproate, carbamazepine, lamotrigine; as opposed to lithium and olanzapine). So it has been tempting to assume that bipolar phenomena might be "seizure-like" in some way. Unfortunately, there still is no clear understanding, nor even dominant guess, as to the basis of bipolar disorder. The accuracy of a seizure-like model is unknown. However, as estrogen has been shown to increase seizures, and progesterone to decrease them; and both seem also to modulate bipolar mood states; there is further support for a seizure/bipolar relationship of some kind.

    Bipolar II seems to get worse with time in many people, especially women. Some researchersLeibenluft have guessed that each menstrual cycle is somehow driving the illness toward greater severity, and that’s why it looks so different in women compared to men. 

    As the illness gets worse, medications that seemed to have helped a person in the past don’t seem to "work" anymore. Antidepressants start to cause mixed state symptoms (can’t sleep, anxious, can’t concentrate, irritable — as well as depressed) and rapid cycling, where at first they were actually very helpful.  

    This potential worsening and increasing complexity of symptoms with time leads to the conclusion that (in the words of Dr. Alan Swann, author of a recent study of this phenomenon Swann et al) if "the illness is progressive, then clearly the sooner you can start treating and preventing episodes the better.... Patients with a very unstable course of illness must understand that maintaining the highest possible level of mood stability is in their best interest.  Eternal vigilance is the price of liberty." Swann

    Also because of this potential worsening with time, patients should not count on being able to "go back" to a previously effective medication. As I read the experts, and watching what happens to my patients, the "name of the game" may be to prevent cycling. That might be how you keep from getting worse. If that’s true, now you have two reasons to get your patient’s symptoms controlled: first, because she’d rather not have symptoms; but also because they could mark a worsening process that you might be able to interrupt with a fully effective treatment.

    Troubleshooting the mood stabilizers

    Problems with lithium can be grouped into two major categories: toxicity and minor side effects. The only common problems with valproate are nausea, weight gain and hair loss.

    LITHIUM

    Lithium toxicity can have major consequences: it can cause renal damage, which when severe can be fatal. It is avoided by scrupulous monitoring of blood levels when the dose goes beyond 600 mg per day, or any time the patient experiences symptoms such as ataxia, slurred speech, or confusion. "Scrupulous" means checking it 4-5 days after each dose increment, using increments of 150mg if uncertain, and checking every six months. If the level is over 1.2 mmol/L, lower the dose; if 1.1 or 1.2 mmol/L, watch very closely and consider lowering as soon as possible.

    Lithium definitely can interfere with thyroid function.Lazarus Unless the patient has a family history of thyroid problems, she/he faces no major risk from this effect, however. About one person in every tenGittoes taking lithium has to take thyroid hormone replacement because lithium lowers thyroid hormone levels. Because this is so common, and because becoming "hypothyroid" can have its own mood effects, it is very important to check thyroid at least once a year, and most experts recommend every six months.

    In almost all cases lithium side effects are less likely with a slow release preparation. Side effects most likely to lead the patient to reject further use are (in roughly descending order of discontinuation):

    • diarrhea
    • tremor
    • nausea
    • urinary frequency

    Diarrhea rarely remits with time on lithium: you must reduce the dose (if already on slow release). Patients will put up with tremor until it interferes with function, such as signing checks in front of people (increased severity in anxious social situations is very common). If dose reduction leads to loss of benefit, you can try low dose immediate release propanolol: most people find this very effective if it doesn’t slow them down too much. Prescribe a 10mg tablet and instruct the patient to use up to 4 per day if necessary, including taking two at once if one is not enough, after they have a feel for the effects of one alone.

    Some patients with nausea will see a decrease in this side effect with time. However, others will have unremitting nausea, even with slow release versions. You must lower the dose until this disappears. If nausea appears after the patient has tolerated it well, suspect toxicity and check a blood level.

    Urinary frequency is almost universal for patients on lithium and increases with blood level. Generally nocturia is the complaint you’ll hear, if you hear anything at all: patients seem to adapt to this one (and the thirst that generally accompanies it). It should diminish somewhat if you lower the level a little.

    Patients should know that if they become dehydrated their body will hang on to sodium, and thus lithium (tubular reabsorbtion). A patient whose routine level is 0.7mmol/L would be unlikely to become toxic just from dehydration, but one with levels of 0.9-1.1 may. Only if they do not respond to lower levels should you maintain a patient this high; periodically try lowering again, as patients can improve enough to tolerate a lower level, after several months of stability.

    Anything which changes glomerular filtration rate, or sodium resorption, will affect lithium level, often substantially. ACE inhibitors and diuretics, and NSAIDs, all can increase lithium levels. These are all relatively contraindicated; if you must use them, you need a medication-oriented psychiatrist to help you manage the patient’s mood stabilizers.

    VALPROATE

    Weight gain is the bane of this medication. Hair loss can occur when at weight gain doses, but the latter is what makes patients really angry if you do not warn them well in advance. Here’s how I manage the weight issue.

    First, tell the patient as soon as you introduce the very idea of this medication that weight gain is a risk, but that if she/he gains any weight at all, we’re going to change the plan. Weight gain is not going to be something they’re just going to have to put up with. Moreover, it doesn’t generally occur unless there has been a marked increase in appetite, to really bizarre levels. One patient described it as "wanting to eat the refrigerator": hungry all the time, and not satiated when eating, even with enormous amounts of food. If she/he starts getting that, it’s time to change the plan.

    The manufacturer says if patients don’t eat more, they won’t gain weight, but even mentioning this has not been practical in my experience: patients don’t believe it, as they see themselves gaining. How about using exercise to control weight? As discussed elsewhere, physical activity is a great health agent generally, a great antidepressant, and probably a mood stabilizer overall: but adherence is a problem even for patients without mood disorders. Advocating increased physical activity will address this weight problem for only a small minority. So, when appetite increases, it’s time to change the plan.

    Lower the dose to the point where this appetite effect disappears. Most patients can tell when it stops, as thought there were a very discrete "threshold" beyond which the problem develops, and under which it ceases. For most patients this "threshold" seems to be around 750mg, which unfortunately is generally the lowest dose sufficient for symptom control with valproate monotherapy. (Some do not get the appetite increase even with as much as 2000mg per day, but this is the minority. While at least some patients who require higher doses will see the appetite effect disappear if they continue, this is confined to men in my experience thus far, and not all men at that. By contrast, women can gain 20-30 pounds in less than a year, and many have said it was harder for them to lose that weight than in previous dieting experiences).

    Of course, when you lower the dose you will lose some efficacy as well. You can see now why I underline the positives of both lithium and valproate in the initial PAR (instead of setting one against the other to decide with which to start): when you must lower one because of side effects, you can make up for it with a low dose of the other. This "rational polypharmacy"Post is now a routine strategy for me, whereas when I initially began using mood stabilizers the idea of two, where one ought to do, was difficult to contemplate.

    What labs, how often, are necessary during follow-up? (revised 8/2005)

    For lithium, one must watch for hypothyroidism by following TSH (remember, the incidence of hypothyroidism is 1/10Gittoes; that figure is likely much higher, closer to 1 in 5 patients, for those who have a family history of thyroid problems.  This may be a sufficient reason to choose another mood stabilizer for that group). Around 0.5-1.0 is ideal. Make sure it's not over 3, don't wait for it to be frankly hypothyroid. See the thyroid and bipolar page for more information on these statements, including two studies showing better response to antidepressants and mood stabilizers, respectively, when TSH was low-normal as opposed to high-normal.

    While on lithium, demonstrate continued normal renal function (creatinine and/or BUN). These should be checked every six months , and patients should know to watch for hypothyroid signs, calling them to your attention. Obviously this includes depression as such a sign! The q 6 month check affords the opportunity to make sure lithium levels are still stable.

    For valproate, there is no clear consensus on mandatory screening or timing of same. With baseline values in place, if SGOT increases, valproate should be suspected. Most experts suggest that elevations up to twice (and I have heard three times) normal are not sufficient to warrant discontinuation and pose no long term risk. Platelet decreases are not uncommon, but if there is no bruising or bleeding, at least one hematologist has authorized me to continue valproate when platelets were as low as 57! The patient is still on it and doing well after 2.5 years (but in this case we tried everything else first!). How often, then, should you do a lab test? There is no clear, practical guideline for this. Unfortunately, anyone who advocates anything other than the manufacturer’s legally driven conservative recommendations incurs major liability. We are left with the manufacturer’s timings as a result. Here is a page describing the literature on lab testing for valproate and carbamazepine. 

    How long should a patient stay on these medications?

    When starting mood stabilizers patients will often ask: "will I have to take this for the rest of my life?" I generally answer this by suggesting that first we should see whether the medication seems to work; later, if it works, we’ll discuss how long to continue it. Patients often seem to grasp intuitively that if they have had symptoms for many years, they will probably require medications to "normalize their brain chemistry" for many years.

    Bipolar I is a chronic illness that requires lifelong prophylaxis, at least after several manic or depressed phases have occurred. Bipolar II is less well defined but our patients’ intuition is generally correct: the longer they have had symptoms, the longer it makes sense to continue medication before a trial of tapering it off. In any case, "taper" is the most important concept: there are several studies in Bipolar I which seem to indicate that rapidly discontinuing lithium can frequently lead to rapid relapse, where tapering off does not present that risk. For lithium at least, stopping should take months, decreasing by 150mg increments all the way to zero; and this probably applies by extension (for the moment, at least, until we have some data to go on) to other mood stabilizers.

    However, patients obviously have ultimate control over how their medication is managed. One must take into account the likelihood that they will try a period off medication, authorized or not, at some point. Education about the risk of stopping suddenly is most important. Clear crisis plans including how to handle suicidal thoughts/impulses (go to your local emergency room or call 911, if afraid of acting on thoughts) are also vital. Generally I have found that enhancing a patient’s sense of autonomy improves outcomes in the long run, unless there is clear risk involved. For example, if a patient is determined to stop the medication, focus on the communications and plan that will take place if things get worse, rather than insisting on "compliance". I try to position myself so that time is on my side.

    Patients must be aware that they cannot expect the medications will "work" again if they stop taking them, then resume. Bipolar disorder in many cases seems to progress, as though each cycle was increasing the likelihood and the severity of yet more cycles. Left uncontrolled for a period of time, it can worsen such that previously effective treatments are no longer adequate. I use the analogy of the old TV show "Let’s Make a Deal" to illustrate this with patients.

    The thoughtful reader may have wondered: "is there any evidence that starting mood stabilizers then stopping them is worse than never having started them at all?" There is no evidence of this. We should worry about it, following the Hippocratic dictum of "primum non nocere" (First, Do No Harm). I have seen about three patients where it looked to me as though this may have occurred (out of more than a thousand), but there are so many variables involved it is impossible to be at all certain. The fact that I have yet to encounter other experts writing about this possibility is somewhat reassuring.

    What about other mood stabilizers? 

    This section got so big it has been moved to its own page, which attempts to list all the current known options, each with a page offering more information. It also includes a discussion of how one chooses between all these options (it's nice to have so many choices, but can become quite confusing evaluating them all!) Here's that Mood Stabilizers page; don't forget the non-medication approaches as well, coming up below. 

    . Is ECT effective in Bipolar II?

    Electro-convulsive therapy (ECT) has efficacy in bipolar disorder at least equal to medications, in the 60-70% range (improved or much improved).Kusumakar Most experts agree that ECT should stand high in treatment algorithms (e.g. UTMB bipolar algorithmTexas). However, these judgements are based on efficacy in bipolar I, as most of the studies of ECT in bipolar disorder preceded the formal recognition of BPII. To my knowledge there has been no published trial of ECT for BPII per se. Several of my BPII patients have had ECT with positive responses. Six patients with very rapid cycling have been presented, the authors concluding that length of illness prior to ECT has tremendous impact on efficacy (longer time ill correlates with lower likelihood of response, and with high likelihood of relapse after treatment).Wolpert

    Unfortunately, in all forms of bipolar disorder one must think about long-term prevention as well as acute treatment, and "maintenance ECT" (repeated single treatments at regular intervals, as prophylaxis) has been much less studied, even for bipolar I. Rapid transcranial magnetic stimulation (rTMS) using a hand-held but very powerful magnet (similar in power to an MRI magnet) has been shown to have antidepressant effectsGeorge, but exactly where to stimulate, how long, with what intensity and frequency, have yet to be worked out. Mark George’s group in South Carolina (the same group which recently demonstrated that retrograde stimulation of the vagus nerve has antidepressant properties!) has a randomized trial of rTMS for bipolar depression underway. This treatment holds some promise: though it requires repetition about every two weeks in most people, there are no significant side effects and almost zero risk.

    Light therapy (or in this case, dark therapy)

    Light therapy clearly works as an antidepressant.  It has much less likelihood of triggering mania, as the regular antidepressants do (as you've probably heard me rant about elsewhere on this website).   But to get a mood stabilizer  effect, you need to consider "dark therapy".  This has much less research to support it, compared to light therapy, but it's worth knowing.  Here's the story on "dark therapy".   

    What is the role of exercise in bipolar disorder treatment?

    Exercise clearly has antidepressant effects,Phelps, Weyerer even being shown recently to be equal to Zoloft -- and perhaps better, 6 months after the end of treatment!Babyak  It would be one of the most widely used antidepressant modalities, but for adherence: it’s hard enough to get people without mood problems to get regular physical activity! Unfortunately this is not a joking matter: though Oregonians are more active than almost all other states in the U.S., still half of the state’s population gets no regular physical activityCD summary. So expecting people with severe mood disturbances to get regular exercise is extremely unrealistic. However, my patients with the worst symptoms seem actually to have greater adherence to exercise regimens: they seem to have learned from experience that it helps, and they are looking desperately for any help they can get.

    One patient in her late 20’s, plagued with profound agitation, described exercise as an anxiolytic. She said that after about 20 minutes of vigorous aerobic effort she would feel something shift. As she "came down" from the workout, her anxiety would seem to "come down" with it. Another patient with an extensive history of self-mutilation (cutting, burning) found that weight work-outs seemed to provide the same kind of benefit: it would keep her from dissociating, or feeling like she was "flying apart".

    As you can imagine, I recommend routine physical activity as a part of an overall mood-stabilizing "lifestyle", but I also warn patients that maintaining such a regimen is extremely difficult and will require great effort on their part. The more severe their symptoms, the more beneficial exercise may be, and the more sustainable therefore (ironically).  Read Exercise and Mood: not the usual rap for specific recommendations. 

    What is "Social Rhythm Therapy"?

    Bipolar disorder is in part a disturbance of circadian rhythmicity: energy "on" in the middle of the night, or "off" in the morning in depressed phases, for example. Almost all BPII patients will experience hypomanic symptoms (remember, this often has the form of severe anxiety/agitation, difficulty concentrating, irritability) after sustained sleep deprivation, sometimes with as little as a single night’s interrupted sleep. These features have led to the hypothesis that "anchoring" a patient’s circadian rhythm might have mood stabilizing effects.

    The Western Psychiatric Clinic has been testing this hypothesis with an adjunctive treatment for bipolar disorder they call Social Rhythm TherapyFrank. Daily routines such as time of awakening, time of rising, time of first meal, time going to bed, time of going to sleep are kept very consistent in this treatment.

    The program is used in conjunction with mood stabilizer medications. Preliminary results show moderate benefit, especially after months of maintaining a very regular schedule of daily activities. For patients whose activities are highly irregular, including especially shift work, there may be value in helping them identify the sources of irregularity and determining how to minimize these influences. I have written employers suggesting that shift work could be contributing to a patient’s instability, suggesting a routine day shift.