This is a prospective cohort study examining the effects of prolonged clinic visit on patient's cardiovascular (CV) risk factors control and whether it will increase the risk of adverse clinical events. This study leveraged the clinic visit deferment arrangement implemented during the COVID-19 pandemic to examine the difference in CV risk factors between patients who deferred and a control group who attended their clinic visit as scheduled.

Introduction Hypertension, diabetes and dyslipidaemia are key cardiovascular risk factors (CVRF) that contribute to adverse cardiovascular events and increase the risk of premature deaths from cardiovascular causes(1-3). Effective management of these risk factors is crucial in reducing the incidence of adverse vascular events such as stroke or myocardial infarction. Currently recommendations suggest follow-up intervals of 3 to 12 months for stable patients, but these recommendations are mostly arbitrary (1, 3-5). The association between more frequent follow-ups and improved cardiovascular risk factor control remains unclear, with conflicting data on its effectiveness (6-10). Notably, no prospective studies have examined the feasibility and safety of extended follow-up intervals beyond 6 months in patients with these CV risk factors. The COVID-19 pandemic introduced a unique challenge, as concerns about person-to-person transmission led many patients to voluntarily defer their scheduled clinic visits. This study aims to prospectively assess the impact of extended follow-up intervals on disease management and risk factors control in patients with CVRF.

Method This single-centre, prospective cohort study compared CVRF control between patients who deferred their physical clinic visit and those who attended as scheduled. The study protocol was approved by the Jointed Chinese University of Hong Kong and North Territory East Cluster Clinical Research Ethic Committee. All patients provided written informed consent data collection. This study followed The Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines for cohort studies reporting.

Setting This study was conducted at the Medical Specialist Outpatient Clinics of the Prince of Wales Hospital, a tertiary regional teaching hospital in Hong Kong, from April 2020 to January 2022. This timeframe was selected to coincide with the implementation of the Drug Refilling Clinic Scheme during the COVID-19 pandemic, which allowed patients to voluntarily defer their scheduled clinic visits while receiving medications from their last prescription for their chronic conditions via postal delivery. Data was collected from clinical notes and the electronic health records.

Participants Patients were recruited if they have at least one of three CVRF: diabetes, hypertension, or dyslipidaemia. Patient newly diagnosed with these conditions were excluded. Additionally, unstable patients defined as those with a history of hospitalization or emergency department visit within the preceding three months, those with complex medical problems indicated by a Charlson Comorbidity Index (CCI)(11) of equal or greater than five, or those whose primary medical issues were unrelated to cardiovascular disease - were excluded. Patients who defaulted their subsequent follow-up appointments with no recorded clinic visit by the end of the study period were also excluded. The study flow is summarized in figure 1.

Patients who deferred their appointment were categorized into the 'deferred' group, while those who attended their appointment as scheduled were placed in the 'control' group. Patients in the deferred group received a prescription refill by post along with a new appointment date. They were instructed to seek medical attention if they experienced any health concerns between follow-up intervals. Routine Blood test including lipid profile, HbA1c and other standard assessments, were conducted at the time of their next clinic visit. Additionally, they were instructed to monitor and record their blood pressure and glucose level at home with these records reviewed during the new follow-up appointments. Patients in the control group followed the standard local practice of clinic visits at intervals of 3 to 6 months. These patients were randomly selected from the list of clinic attendees and screened for eligibility at a 2:1 ratio to the 'deferred' group. A 2:1 ratio was used as more patients attended their scheduled visits than deferring. Similar blood test was performed during each clinic visit, as in the deferred group.

Variables Information on medications, blood tests result, and office blood pressure readings were recorded during each clinic visits. The baseline for subject in the control group was established at the time of the scheduled clinic visit (time 0). Since only subjects in the control group attended the clinic, the baseline for subjects in the deferred group was derived from their last recorded clinic visit before recruitment. Data from two subsequent physical clinic visits were collected in both groups (time 1 and time 2). Two follow-up visits were used instead of one for both groups to allow the second clinic visit to serve as an internal control.

Outcome measures The primary outcome was uncontrolled CVRF, defined as the presence of any of the following: uncontrolled DM with HbA1c >7%, uncontrolled hypertension with office systolic blood pressure (SBP)>130mmHg and/or diastolic blood pressure (DBP) >80mmHg, or uncontrolled dyslipidaemia defined as low-density-lipoprotein (LDL) >1.8mmol/L. The primary safety outcomes are the composite endpoints of unscheduled hospitalization, hospitalization due to cardiovascular events, and/or major adverse cardiovascular events (MACE) which included acute coronary syndrome, myocardial infarction, ischemic stroke and death within the study period. The secondary endpoints are changes in HbA1c level, Low density lipid (LDL) level and office blood pressure reading from baseline (time 0) to time 1 and to time 2. In addition, a clinic satisfaction questionnaire (CSQ) was administered to a subset of patients, and satisfaction scores were compared between the two groups.

Sample Size and Statistical Method We based the sample size calculation on establishing non-inferiority of deferred group over control group. The prevalence of uncontrolled HbA1c, LDL and hypertension was estimated to be 60%, 30%, and 60% respectively (12-14). Assuming 80% of power and one-side α = 0.025, a sample size of 432 patients- with 288 patients in the control group and 144 patients in the deferred group - would be required to demonstrate non-inferiority in the deferred group, with inferior margin set at an absolute difference of 5% point.

Continuous variables will be reported using descriptive statistics i.e. mean, median, and standard deviation. Qualitative variable will be summarized in frequency and percentage. T-test or sign rank test will be used for comparison between continuous variables, and chi-square or Fisher's exact test will be used for comparison of categorical variables between groups.

Kaplan-Meier method will be used for primary safety outcomes analysis and estimate time-to-event distributions between the two arms. The log-rank test will be used for comparison. Time 0 was set as the baseline clinic visit in the control group, or last recorded clinic visit in the deferred group. Follow-up end date was set at the occurrence of primary safety outcomes or 1 year from time 2, whichever occurs earlier.

Univariate logistic regression will first be performed to identified individual variable associated with primary outcomes and primary safety outcomes. Those variables with significant association with the primary outcome (defined as a two-tailed p value <0.05) was entered into the Multivariate logistic regression analysis. The final multivariate model was then constructed by backward deletion of the least significant characteristic, until all remaining variables were significantly associated with the endpoint (p < 0.05). This final model will be then assessed in a separate validation cohort which by calculating the area under the receiver-operating-characteristic (ROC) curve (AUC). A perfect model will have an AUC =1, while a worthless model will have an AUC of 0.5. Statistic analysis was performed using STATA and R software.

Inclusion Criteria:

1. Patients who attend outpatient clinics at the Prince of Wales Hospital
2. Patients who are followed up at the clinics for at least 1 CV risk factors, namely diabetes, hypertension, dyslipidaemia, or history of vascular events including cerebrovascular accidence, ischemic heart disease or peripheral vascular disease.

Exclusion Criteria:

1. New referrals to the outpatient clinics
2. Unstable patients such as patients with recent hospitalization within 3 months to the interviewing date
3. Patients with complex medical problems indicated by Charlson Comorbidity Index of equal or more than 5
4. Patients whose active medical problems are not cardiovascular related
5. Patient who are taking vitamin K antagonist
6. Patients who defaulted their follow-up appointment with no new appointment on record.

Study participants will be recruited primarily form Medical Specialist Outpatient Clinics of the Prince of Wales Hospital, a tertiary regional teaching hospital in Hong Kong.