Interventional Studies: Hypertension Control

• Tutorial 4 is about understanding interventional studies

Interventional Studies

• Interventional studies actively attempt to change a disease determinant in a preventative trial.
• Alternatively, they change the progress of the disease through treatment in a therapeutic trial.
• An intervention group receives the intervention.
• A control group receives 'usual care'.
• Outcomes are measured and compared between groups.

Study Design Overview: Hypertension Control

• An article published in the Journal of the American Medical Association (JAMA) in 2008 examines the effectiveness of home blood pressure monitoring, web communication, and pharmacist care on hypertension control.
• The study is a randomized controlled trial.
• Treating hypertension can potentially decrease mortality and disability from cardiovascular disease, but the bulk of hypertension is currently inadequately controlled.
• Objective is to determine if patient web services, home blood pressure (BP) monitoring, and pharmacist-assisted care improves BP control.
• Three-group randomized controlled trial, the Electronic Communications and Home Blood Pressure Monitoring study based on the Chronic Care Model.
• Conducted at an integrated group practice in Washington state.
• 778 participants aged 25 to 75 years with uncontrolled essential hypertension and Internet access.
• Care was delivered over a secure patient Web site from June 2005 to December 2007.
• Participants were randomly assigned to usual care, home BP monitoring and secure patient Web site training only, or home BP monitoring and secure patient Web site training plus pharmacist care management through Web communications.
• Percentage of patients with controlled BP (<140/90 mm Hg) and changes in systolic and diastolic BP at 12 months.

PICO of Hypertension Trial

• Population (P): Uncontrolled hypertensives on medication
• Intervention (I): Home BP monitoring with web services training
• Comparison (C): Usual Care
• Outcome (O): Change in BP (systolic/diastolic) and successful control percentage

Patient Screening and Randomization

• 2937 patients visited the clinic for the screening visit.
• Many of these patients were excluded because they did not fulfil all the criteria for participation.
• Only 778 patients randomized after the screening visit.
• Some reasons patients were excluded include blood pressure was controlled, they refused to participate, they had blood pressure that was too high, their arm was too large for the cuff, medical exclusion, etc.
• The 778 patients were randomly selected from the 2937 hypertensives who came from the screening visit.
• Random assignment into three groups randomly via chance/probability alone.

Random Assignment Rationale

• Investigators chose to do so in the groups so that the characteristics are like eachother in characteristics that could affect blood pressure control.
• Would allow a more valid comparison between the groups at the end of the study
• Ensures any differences in blood pressure control observed after twelve months could be attributed to the treatment they received
• The reasons that the opaque envelopes were used during patient and randomization was to mask the results of who is in each group.

Randomization Defined

• Randomization means each individual has an equal chance of being in the treatment or control group and that assignment is based on probability / 'chance' alone.
• This strategy equalizes baseline differences between groups and helps ensure any outcome difference is due to the intervention alone.

Hypertension Study Treatments

• Group A (Control): Advised to work with physician to improve BP control (usual care)
• Group B (Intervention): Home BP monitor and web training on resources to support BP control
• Group C (Intervention): Home BP monitor, web training, and pharmacist-assisted care (specific, tailored recommendations, lifestyle goals, follow-up plan via phone/web)
• All Participants: Pamphlet "High Blood Pressure and You'" and ‘The No-Waiting Room’

Blinding

• At the 12-month follow-up visit, trained research assistants blinded to the patient's study group measured BP.
• Research assistants did not know which group the patients were assigned to.
• It helps to avoid bias between the researcher or interviewer when preforming tests.

Placebo Effect

• A reported change in status is usually an improvement, stemming from the expectation that the treatment is effective.
• It is the measureable, observable, or felt improvement in health not due to treatment, but to the perception that it works.
• Placebo effects may occur if the outcome is subjective.
• Placebo may occur if there is an expectation among those receiving the intervention.
• Bias refers to a systematic error in the way the study is conducted.
• Outcomes are measured that make the intervention look better or worse than it really is.

Clinical Trial Design Principles

• Uses test intervention and control groups.
• Compared with respect to outcome
• Ensures baseline comparability through Randomization
• Ensures adequate baseline comparability.
• Follow-up and measurement of outcome should be unbiased.
• Use standardized protocols, masking (blinding), and placebos, where applicable.

Measuring Intervention Effect

• Measure effectiveness through quit rates in new cessation measures.
• Control against outcome of current measure.
• You meausre effectiveness through calculating the difference in the outcome in quit rates between the groups.
• Difference in “quit rate” is measured between two groups
• You work this out by calculating the absolute difference which is dividing (60 - 20) / 100,
• Group 1's had 30% higher quit rate.
• Group 1s quit rate were 3x that of group 2's

Example Infection Rate Calculation

• Measure through infection rates after a procedure Vs current standard procedure.
• Group 1 had a 2% lower risk of infection, compared with Group 2 at 5%

Terms

• Absolute risk reduction (ARR) = (Incidence in intervention group – Incidence in control group)
• Relative risk reduction (RRR) = Incidence in intervention group − Incidence in control group /Incidence in control group X 100%

Hypertension Intervention Effectiveness

• At the end of 12 months with the proportion of patients who achieved good control of blood pressure in three assigned groups the data was.
• A - Usual Care (n=247), the proportion with BP controlled was 0.31 (or 0.25-0.37 was the confidence)
• B - BP monitoring and Web services training (n=246), he proportion with BP controlled was 0.36 (or 0.30-0.42 was the confidence)
• C - BP monitoring and Web services training plus pharmacist care (N=237), the proportion with BP controlled was 0.56 (or 0.49-0.62 was the confidence)
• Relative Difference or 16% Increase, with the Group B relative to A
• Relative Difference of 81% Increase, with the Group C relative to A

Interventional Study Design Principles

• The intervention group should be evaluated against a control group, and both are compared with respect to outcome.
• Both groups should be comparable at baseline, so any outcome difference is due to the intervention, not inherent differences.
• Achieved through Randomization
• Follow-up and measurement of outcome in both groups are the same or very similar.
• Reduce biases/errors
• Blinding of participants is one way to reduce bias due to the placebo effect.

Measuring the Effect of Intervention

• Compare the outcomes between the groups.
• Expressed in absolute or in relative terms.