Cosmos Study

Most PHQ-9 Responses Are Submitted During the Day, but Those Submitted Overnight Show Higher Depression Severity and Thoughts of Self-Harm Rates

August 18, 2026

Team A:Jeff Trinkl, MDEric Barkley

Team B:Dave Little, MDSam Sahakian

Key Findings

Most Patient Health Questionnaire-9 (PHQ-9) responses, which are primarily patient entered, were submitted during the day; submission volumes peaked mid-morning and were lowest overnight, with only about 2.5% of submissions between 11 p.m. and 5:59 a.m.
The few surveys submitted overnight were frequently more severe: at the 2 a.m. hour, about 14.1% of surveys scored in the moderately severe or severe range; at the 6 a.m. hour, about 5.3% were.
Positive responses to question 9 (thoughts of self-harm or being better off dead) were highest at the 2 a.m. hour (10.1% of responses). The positivity rate was lowest during the 6 a.m. hour (4.2%). Even so, nearly 97% of all positive responses to question 9 and 96% of severe-range responses were submitted during the day.

The Patient Health Questionnaire-9 (PHQ-9) is a nine-item questionnaire widely used to screen for and measure the severity of depression.¹ Patients are increasingly completing the questionnaires themselves through online patient portals like MyChart, which means responses can arrive at all hours rather than only during clinic visits. Previous research on mood and suicide has found that risk is not evenly distributed across the day. After accounting for how many people are awake at each hour, deaths by suicide are disproportionately concentrated overnight.² Being awake at night has been proposed as an independent risk factor for suicidal thoughts and behavior.³ Patterns in PHQ-9 responses have not previously been well described. Understanding when patients disclose the most severe symptoms could help health systems time outreach, monitoring, and crisis-response resources.

We examined more than 600,000 PHQ-9 scores for U.S. patients aged 12 and older. Based on the questionnaire data assessed, we expect that the majority of surveys were completed by patients themselves in the patient portal, while some might have been entered by a clinician on their behalf. Each survey was assigned to the hour of day it was first submitted. Each PHQ-9 question is scored from 0 to 3, for a total of 0 to 27. A score of 15 or higher indicates moderately severe to severe depression, and question 9 asks specifically about thoughts of self-harm or being better off dead.

PHQ-9 surveys were overwhelmingly completed during the daytime. Submission volume climbed through the morning to a peak around the 9 a.m. hour and fell to its lowest overnight, with only about 2.5% of all surveys arriving overnight between 11 p.m. and 5:59 a.m., as seen in Figure 1. The small number of overnight surveys, however, skewed markedly toward higher severity. At the 2 a.m. hour, 14.1% of surveys reached the moderately severe to severe range (a score of 15 or higher), compared with 5.3% of surveys submitted at the 6 a.m. hour. While the data alone cannot account for the reasons for this, two possible explanations are that patients’ symptoms might genuinely be worse at night, or that the small group who complete a questionnaire unprompted in the middle of the night might simply be more distressed than the far-larger daytime population, many of whom are screened routinely during or prior to clinic visits.

Figure 1

Distribution of PHQ-9 Score Category by Hour of Submission

Figure 1. The number and rate of completed PHQ-9 responses by category and hour of submission.

Question 9 of the PHQ-9 asks about thoughts of self-harm or being better off dead. Responses to this question followed the same daily pattern as severity above, with the share of surveys with a positive response more than twice as high at the 2 a.m. hour (10.1%) as at its lowest points around the 6 a.m. hour (about 4.2%), as seen in Figure 2. As with overall severity, the elevated overnight rate reflects only a small slice of the total: because so few surveys are submitted overnight, only about 3% of all positive question 9 responses arrived overnight (11 p.m. to 5:59 a.m.), while the remaining 97% came in during the day.

Figure 2

Surveys Indicating Thoughts of Self-Harm or Being Better Off Dead by Hour

Figure 2. The volume and rate of completed PHQ-9s with a positive response to question 9 (thoughts of self-harm or being better off dead) by hour of submission.

These data come from Cosmos, a dataset created in collaboration with a community of Epic health systems representing more than 322 million patient records from 2,200 hospitals and more than 50,000 clinics from all 50 U.S. states, Canada, Lebanon, and Saudi Arabia. This study was completed by two teams that worked independently, each composed of a clinician and research scientist. The two teams came to similar conclusions. Graphics by Brian Olson.

References

Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606-613. doi:10.1046/j.1525-1497.2001.016009606.x
Perlis ML, Grandner MA, Brown GK, et al. Nocturnal Wakefulness as a Previously Unrecognized Risk Factor for Suicide. J Clin Psychiatry. 2016;77(6):e726-e733. doi:10.4088/JCP.15m10131
Tubbs AS, Perlis ML, Basner M, et al. Relationship of Nocturnal Wakefulness to Suicide Risk Across Months and Methods of Suicide. J Clin Psychiatry. 2020;81(2):19m12964. Published 2020 Feb 25. doi:10.4088/JCP.19m12964

Data Definitions

Study period: 1/1/2023 to 4/30/2026

Study population: inclusion

Patients:

Aged 12 or older at their first survey response
With the associated encounter department located in a U.S. state
With a completed PHQ-9 survey (either answers for all nine questions or answers of “Not at all” to the first two questions and no answers for the following seven)

Exposures: Local hour of survey submission (0–23), derived from the initial survey response in the department’s local time

Outcomes: Volume of PHQ-9 surveys entered by hour
PHQ-9 total score by hour
Rate of positive PHQ-9 question 9 responses by hour

PHQ-9 question 9: A survey response to the item “Thoughts that you would be better off dead or of hurting yourself” indicating “Several days” or more

PHQ-9 total score: Sum of response values (0–3) for questions 1 through 9 on a single survey submission (range 0–27) or responses of 0 for questions 1 and 2 and the rest unanswered

Model specifications: Descriptive statistics
Sensitivity analysis: We excluded surveys where only the first two questions were answered and found similar results. Additionally, those were a small proportion of the overall surveys in this study.

Limitations: Survey submissions were categorized by the associated encounter department’s time zone, which could differ from the patient’s time zone where they took the survey.
The data cannot reliably distinguish surveys entered by patients through a portal from those documented by a clinician during a visit.
PHQ-9 might be documented through other methods by a clinician during an encounter, such as flowsheet templates, within notes, or paper screenings that were not incorporated into this analysis.