Before Prescribing an App for Negative Symptoms

A 46-year-old man has lived with schizophrenia for 15 years. His antipsychotic dose has not changed in a year, and his hallucinations are rare and quiet. He spends most days at home, has stopped calling his brother, and says he does not see the point of the day program. His sister has read about a new FDA-cleared app for "the symptoms medication doesn't fix" and asks whether he should have it.

The app is Motivista (CT-155), from Click Therapeutics. FDA cleared it on July 30, 2026, and the company announced the clearance on October 5. Click calls it "the first FDA-authorized treatment indicated specifically for the treatment of negative symptoms," and the FDA device records we checked show no earlier product with that indication. In our view, because negative symptoms are so hard to treat, the claim is easy to read more broadly than the evidence allows.

The clinical question is narrow: for a stable outpatient with prominent negative symptoms, what can this app be expected to change, and how would anyone know?

Motivista is a 16-week smartphone program. It is a prescription device, not a drug. FDA cleared it through the 510(k) pathway as a Class II "computerized behavioral therapy device for psychiatric disorders," the regulation that also covers prescription apps for substance use, insomnia, depression and anxiety. In its clearance letter, FDA "determined the device is substantially equivalent" to a legally marketed predicate device. The predicate is Rejoyn, a prescription app for major depressive disorder. Devices of this type must also meet special controls, and the 510(k) summary reports clinical trial data.

That is the standard for clearance. It is not an approval, the term FDA uses for drugs and for devices that require a premarket approval application. Healio, reporting a company executive's remarks, nonetheless described Motivista as the first prescription digital therapeutic "approved" by FDA for schizophrenia, although its own lead correctly says the app was cleared.

The cleared indication is specific: "the treatment of negative symptoms of schizophrenia as an adjunct to clinician-managed outpatient care, including antipsychotic therapy, in patients 18 years of age and older with clinically stable positive symptoms." The regulation requires labeling stating that devices of this type are "not intended for use as a standalone therapy" and do "not represent a substitution for the patient's medication."

The pivotal trial, CONVOKE, was published in JAMA Network Open on September 25, 2026. It randomized 457 adults at 66 US sites to Motivista or a control app for 16 weeks. Participants saw the two only as "digital therapeutic A" and "digital therapeutic B." The control app offered a few minutes a day of educational lessons about the illness, matched in frequency to Motivista and designed to be engaging "without therapeutic benefit." Investigators and raters were blinded, and centralized blinded raters scored the primary outcome.

That outcome was change at week 16 in the motivation and pleasure scale of the Clinical Assessment Interview for Negative Symptoms (CAINS-MAP), on which lower scores mean fewer symptoms. Participants started at a mean of 26.3. Scores fell by 6.8 points with Motivista and 4.2 points with the control app (least-squares means from a mixed model; the observed changes were 6.7 and 4.3). The between-group difference was 2.6 points (95% CI 1.2 to 4.0; P < .001), a standardized effect size (Cohen's d) of 0.36, conventionally described as small to moderate. In the authors' sensitivity analyses, the result remained statistically significant until missing outcomes in the Motivista group were assumed about 5 points worse than imputed.

Click describes the result as "a 62% relative improvement." That is the 2.6-point difference between the groups expressed as a share of the control group's 4.2-point improvement; Click also reported the standardized effect alongside it. It is not a 62% reduction in symptoms, and it does not mean 62% of patients improved. Because the denominator is the control group's change, the same 2.6-point advantage would look larger against a control that improved less. The clinically interpretable figures are the 2.6-point difference, on a scale where the average participant began at 26.3, and the effect size of 0.36. Neither the paper nor the trial's protocol and analysis plan define a minimal clinically important difference for CAINS-MAP, and no CAINS-MAP responder analysis was planned or reported. The authors write that the change "can therefore be reasonably interpreted as meaningful change"; that is their interpretation, not a demonstrated threshold.

The control group's 4.2-point improvement cuts two ways. Attention, expectation, repeated assessment and regression toward the mean may all contribute to change on a rated scale, and the trial estimates Motivista's advantage over this particular control under trial conditions. Conversely, if the control app itself helped a little, the 2.6-point difference would understate Motivista's effect compared with usual care alone.

Expressive negative symptoms, such as blunted affect and reduced speech, did not differ significantly between groups at week 16. Within the motivation and pleasure scale, the vocational subdomain did not differ significantly; the social and recreational subdomains did, in post hoc analyses. Patient-rated global improvement separated at week 8 but not at week 16 (P = .07). Secondary outcomes were not corrected for multiple comparisons.

The protocol prespecified social functioning, on the Personal and Social Performance scale (PSP), and defeatist beliefs as secondary endpoints; the statistical analysis plan added patient-rated severity and total PANSS score, and FDA's summary lists all four. The main report and results supplement report no week-8 or week-16 outcomes for these four measures. The analysis plan cites a 7-point change on the PSP as clinically meaningful, but the main report and results supplement do not report PSP outcome results. For now, the published trial evidence concerns a rated symptom domain, not demonstrated gains in work, relationships or daily functioning.

Positive symptoms did not differ between groups on average (difference 0.1 points).

Participants had moderate to severe negative symptoms, were at least six months past diagnosis, and had been on a stable antipsychotic dose for 12 weeks. They were in a stable phase of illness, with stable housing and no inpatient treatment for schizophrenia in the prior 12 weeks. People with prominent positive symptoms, a current mood episode, moderate or severe substance use disorder (other than tobacco) in the past year, active suicidal ideation with method or intent in the past 3 months or suicidal behavior in the past 6 months, more than two antipsychotics, or psychotherapy in the prior 13 weeks were excluded. Enrolled participants had been diagnosed a median of nearly 15 years earlier; 80% were unemployed and about 63% reported incomes under $25,000. Just over half were Black.

The authors state the limits plainly. Newly diagnosed people, those with milder negative symptoms and those recently in psychotherapy or cognitive remediation were not studied, and "smartphone ownership, internet access, and stable housing may have introduced selection bias." Only four participants were aged 18 to 21; FDA's summary relies on pediatric extrapolation data for that range. The trial reports no effectiveness data after the 16-week treatment period; safety was followed for four more weeks. Psychiatric Times reports that Click is working on an open-label extension on engagement and durability of effect.

Early discontinuation was similar, at about 23% and 22%. Engagement is harder to summarize. Control participants used their app on more days (72% vs 61% of days for Motivista, per FDA's summary), and fewer Motivista participants met the trial's per-protocol standard: completing treatment, an activity on at least 67 of 112 days (about 60%), and no major protocol deviation affecting the primary endpoint (119 vs 152). Motivista users, however, had far more days with sessions lasting at least a minute (median 64 vs 21), which fits a control app built for brief daily use. The practical point is that the program asks for sustained effort, and some patients will not sustain it.

Treatment-emergent adverse events were reported in 7.9% of the Motivista group and 11.7% of the control group. FDA's summary records suicidal ideation or behavior during weeks 4 to 16 in seven Motivista participants and four controls, and two psychotic exacerbations in one Motivista participant that the investigator judged possibly related and the sponsor judged unrelated. These numbers are too small to establish a difference in either direction.

Click states that "Motivista cannot send alerts or warnings to the prescriber," and a company executive told Healio that it "is not designed for real-time patient monitoring." Click's safety information tells patients to stop using the app and contact their clinician if symptoms worsen or they have thoughts of harming themselves or others, and to call 988 or 911 or go to an emergency department when needed. It also says the app is not designed for people at serious suicide risk or with unmanaged positive symptoms. Changes in mood, suicidal thinking or positive symptoms during use need the same clinical follow-up as before, with an urgent route agreed in advance.

CONVOKE was funded by Boehringer Ingelheim, which codeveloped CT-155 with Click. The two companies jointly designed the study, protocol and statistical analysis plan; independent contract research organizations ran the trial, and an external vendor held the randomization codes. Seven of the paper's 15 authors were employees of Boehringer Ingelheim or Click. Six more disclosed fees or payments from one of the companies: four personally (one for work outside this study) and two through named clinical research or practice organizations. Sponsored trials can be rigorous, and this one used blinded central raters and an active control. We found no independent replication. The trade coverage we reviewed in Psychiatric Times and Healio quoted only company executives.

The man in the opening resembles the trial population: long illness, stable positive symptoms, stable medication, prominent withdrawal and loss of drive. The app may be reasonable to discuss, starting with what he wants. Before prescribing, a few questions are useful.

A truthful answer to his sister is that the app is, in the FDA records we checked, the first product cleared for these symptoms; it was tested in a blinded trial in people like her brother; and it produced a modest average improvement in motivation and pleasure over a control app, without published trial evidence yet that it changes daily functioning. If he is interested, that is reason to consider it alongside the treatments that remain the foundation of his care, and to measure whether it helps him.

Source note: Regulatory details are from the FDA 510(k) record and clearance package for K254038 and the current eCFR text, read October 21, 2026. The search for earlier devices covered FDA 510(k) and classification records for this regulation and was not exhaustive. Trial results are from the CONVOKE publication and its supplements; signs are reported here as improvements, so the paper's −2.6 (95% CI −4.0 to −1.2) appears as 2.6 (1.2 to 4.0). Per-protocol, days-of-use and suicidality counts are from FDA's 510(k) summary; endpoint lists are from the protocol and statistical analysis plan in the paper's Supplement 1. Click's safety instructions are from its October 5 release; the full instructions for use were not reviewed. The 62% calculation is the editor's arithmetic from the published least-squares means. Billing information is from CMS's CY2026 summary; payer coverage of Motivista was not established. The opening case is a fictional composite for illustration.

Educational Disclaimer: The Psychiatric Record provides general educational information for psychiatric and mental-health professionals. Content does not constitute medical, legal, regulatory, compliance, billing, or other professional advice; does not establish a standard of care; and is not a substitute for independent professional judgment. Appropriate assessment and treatment depend on the individual, setting, and current authoritative guidance. Device use should follow the complete current labeling. The opening case is a fictional composite and does not describe a real patient.