The study participants were hypertensive patients who participated in the National Basic Public Health Service - Management of Hypertension in Community program from June 2018 to September 2023 in Shanghai, China. For inclusion, they had to be 50 to 90 years of age and measured their blood pressure with an automated office blood pressure platform. Until the end of September 2023, 343,400 patients from 142 community health centers across all 16 districts of the city of Shanghai participated in the program, and a total of 1,654,072 measurements were obtained. For the primary analysis, we selected per participant blood pressure measurement on a single occasion. To enrich the number of patients who had their blood pressure ever measured between 7AM and 8AM, we selected the first blood pressure measurement in this time period. Otherwise, we selected the first ever blood pressure measurement in the rest hours. For a secondary analysis in a subgroup of patients with uncontrolled blood pressure between 7AM and 8AM, we further selected the nearest blood pressure measurement between 9AM and 4PM within the same season.

Blood pressure was measured with an automated office blood pressure platform, developed by the Shanghai Municipal Center for Disease Control and Prevention in collaboration with the Shanghai Institute of Hypertension at Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China. At each participating community health center, the platform was established in a dedicated area with a comfortable indoor temperature. Validated automatic electronic blood pressure monitors were placed on tables with chairs properly adjusted to body height and with multiple-sized arm cuffs. Whenever a patient needed any help or encountered any problem, a trained volunteer was present to provide help for proper blood pressure measurement, including but not restricting to the choice of properly sized cuffs, adjustment of the height of chairs, and initiation of the measuring system.

Blood pressure was measured with a standardized procedure, after having rested for 5–10 min in a seated position. Blood pressure was measured twice consecutively with a one-minute interval. If the difference between the first two blood pressure readings, calculated automatically by the software, was ≥5 mmHg, a third blood pressure measurement was obtained. These two or three readings were automatically transmitted to a digital platform and averaged for clinical decisions and for the present analysis.

A quality control program was applied for the hardware and software system, measurement procedures, and training of the medical staff for the management of the system. Briefly, an expert panel evaluated the appropriateness of the hardware, such as the space and area for the installation of the devices, tables and chairs, device for the identification of patients, computers and blood pressure monitors including the cuffs. The software controls the whole process, including blood pressure measurement and data transmission, storage and analysis. Blood pressure monitors were annually calibrated. The staff were trained for blood pressure measurement and ready for handling any possible problems during measurement.

Blood pressure control was defined as a systolic blood pressure below 140 mmHg and a diastolic blood pressure below 90 mmHg. Regions of the community health centers were classified as urban (non-agricultural registration population percentage greater than 70% and a migrant agricultural population percentage no more than 35%), suburban (non-agricultural registration population percentage greater than 70% and a migrant agricultural population percentage over 35%), and rural (non-agricultural registration population percentage less than 70%). Current smoking was self-reported current tobacco consumption by the patients. Diabetes mellitus was also a self-reported diagnosis. Body mass index was classified as underweight and normal (<24.0 kg/m2) and overweight and obesity (≥24.0 kg/m2).

Seasons were defined as follows: March to May as spring, June to August as summer, September to November as autumn, and December to February of the following year as winter. Morning blood pressure was defined as that measured from 7AM to 8AM. Noon blood pressure was defined as that measured from 9AM to 4PM. These morning and noon blood pressure measurements were obtained on different days but within the same season. In case of multiple measurements, the two closest ones were selected. Masked uncontrolled morning hypertension was defined as a systolic/diastolic blood pressure uncontrolled in the morning (≥140/90 mmHg) but controlled around noon (<140/90 mmHg).

The SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA) was used for data management and statistical analysis. The R software version 4.2.2 was used for graphics. Continuous variables were described using means and standard deviations, and categorical variables were described using frequencies and percentages. Chi-square test was used to compare discrepancies between groups. Analysis of variance (ANOVA) was used for continuous variables. Unconditional logistic regression analyses were performed for the computation of adjusted control rate of hypertension, the distributions of control rate of hypertension across months and hours, and 95% confidence interval (95%CI). The adjusted variables included participants’ characteristics, such as gender, age, region, education level, hypertension course, BMI, and diabetes mellitus. Patients with both morning and non-morning blood pressure readings were selected from each quarter to analyze the prevalence of masked hypertension among those with uncontrolled morning blood pressure. Unconditional stepwise logistic regression was used to analyze the factors influencing masked morning hypertension. A significance level of P < 0.05 (two-tailed) indicated statistical significance.