Don't Use the Rapid Mood Screener Without Modifications for Greater Sensitivity

The Rapid Mood Screener (RMS) as currently written will fail to detect more than half of all the people with bipolar, leaving them exposed to years of unnecessary suffering due to misdiagnosis and the risks of antidepressants. But two modifications and probabilistic scoring restore sufficient sensitivity for routine screening. The WHO's World Mental Health Survey found lifetime prevalence rates of 1.4% for subthreshold bipolar, and 2.4% for bipolar spectrum; versus 0.6 and 0.4 for Bipolar I and II respectively [1]. While we yet lack data on how best to treat subthreshold bipolarity, the risk of adverse reactions to antidepressants among such patients is an established concern. For example, in youth presenting with depression or anxiety who had a parent with bipolar, 57% of those started on an antidepressant experienced irritability, aggression, impulsivity, or hyperactivity leading to discontinuation [2]. I was a senior at Brown University, age 21, when my first depression hit; and starting medical school at Dartmouth during the next one. Twenty five years of hellish symptoms and functional compromise later, I received a bipolar diagnosis, effective treatment with lamotrigine and now 18 years of remission (with the exception of one year during perimenopause.) I lost many years of my life due to not ever being screened for bipolar. Unfortunately, the RMS would not have identified my bipolar. My score during my first depression would have been zero, as I'd not previously had hypomanic or manic symptoms. Even ten years and three more episodes later for a total of 32 months of crippling depression, I still would only have scored a “1”, for antidepressant-associated irritability, far from the cutoff for a positive test (≥ 4). In fact, I would never have reached that “positive” threshold. My hypomania looks like someone with high energy, except when triggered by an antidepressant. Before finally being diagnosed with bipolar, neither I nor my close friends and family recognized my “ups” as hypomanic. The RMS will miss people like Sara until they have six more episodes of depression; become hyper or irritable on an antidepressant; and only if their first episode was before age 18. These thresholds prioritize specificity, whereas sensitivity is more important in a screening questionnaire [3]. Therefore we recommend three simple changes, shifting to thresholds used in the thrice validated Bipolarity Index and the DSM-5. To warrant a “yes” answer: A. RMS Question 1: Lower required depression episodes from 6 to 3. B. RMS Question 2: Widen age of onset from < 18 to ≤ 25. C. RMS Questions 4–6: Use DSM-5 duration for hypomania—4 days not 7. Positive predictive values will be lower using these thresholds. But negative predictive value, key for screening, will increase. (In this regard, note that sensitivity was only 0.68 in the RMS validation that included Bipolar II [4]). Granted, a reduction in specificity will increase the number of patients for whom more information is needed before offering treatment (in categorical terms, an increase in false positives). Additional questionnaires such as MoodCheck, which combines the Bipolar Spectrum Diagnostic scale with elements of the Bipolarity Index, are an efficient second step. SS: Even with the two simple changes above, I'd still have been “negative” on the RMS at age 21. Indeed I'd still have been negative even after 25 years and many more horrible episodes of depression. But just one more bit of information could have changed the calculus. Someone needed to ask me about my family. My mother had depressions fully alleviated by lithium. Her father had episodes of sitting for hours upon hours on the back porch, missing his high-level academic work. Neither had a diagnosis of “bipolar”. As clinicians know, family histories can be murky. Sorting through patients' accounts takes time. Likewise, seeking collateral information, which can be essential in many cases, requires obtaining consent and perhaps even scheduling another appointment. For clinical efficiency, it is appropriate to gather family history, details of illness course (episode frequency, duration, and postpartum experience), and collateral information only for those whose RMS results suggest further diagnostic investigation. What is the best cut-off to use for this function? Answer: don't use a cut-off at all. Dichotomous positive/negative scoring is appropriate when signs, symptoms and laboratory findings delineate a “zone of rarity” between affected and unaffected populations, but such a zone is not seen between unipolar and bipolar depression [5]. Instead, a “probabilistic” approach to diagnosis was recommended by the ISBD's Diagnosis Task Force and since refined. This approach does not leave clinicians in a fog (a common rejoinder). Rather, as shown in Table 1, there are basically three clinical paths for patients presenting with depression or anxiety, depending on their probability of bipolar. SS: Closing note: my Bipolar II was finally identified, by the fifth psychiatrist, with a simple questionnaire that included the features we are suggesting: age of onset up to age 25, more than 3 episodes of depression, and a consideration of my family history. For the first time I was prescribed a mood stabilizer, lamotrigine. It saved my life. We suggest simple modifications for anyone giving or taking the RMS: use previously validated, sensitive thresholds for episode number, age of onset, and episode duration; and score probabilistically. We recognize that some practitioners prefer a screening tool with direct, simple implications: for example, for primary care, “antidepressant, or refer?”; or for psychiatric specialists (and primary care providers who cannot easily refer), “antidepressant, or mood stabilizer?” But screening tools do not make diagnoses so should not be used in this fashion [3]. Rather, for people with depression or anxiety, a modified RMS can efficiently identify those who do not need additional assessment before choosing a treatment (psychotherapy or medication) and also identify those for whom the probability of bipolar is very high. For people in the middle, additional steps are necessary: (1) obtaining further diagnostic information with additional questionnaires, focused interviewing, and sometimes collateral input; (2) education about the mood spectrum; and (3) balanced information about treatment pros and cons—all leading to (4) shared decision-making. Sara Schley acknowledges that she is making the personal disclosures in this paper of her own accord and takes full responsibility for doing so, having previously shared these and other details of her experience of bipolar in a memoir, Brainstorm: From Broken to Blessed on the Bipolar Spectrum—which inspired the documentary BrainStorm, available now, and showing on national PBS in Spring 2027. The authors have nothing to report.