PHAR 628: Research Methodology

• PHAR 628 Research Methods & Biostatistics Lecture on Research Methodology by Lixian Zhong, Ph.D.
• Lecture date: January 22nd, 2025

Objectives

• The lecture aims to build an understanding of terminologies related to research methodology.
• Common research methodologies used in clinical research will be discussed.

Research Methodology

• Research methodology emphasizes data collection and measurement techniques.
• Data collection is a critical step after finalizing a suitable research design.
• Data collection techniques are classified into primary or secondary methods.

Primary Methods

• Primary methods involve collecting data specifically for the research question in question.
• The benefits are that data is collected that that fits the needs and the study design.
• A disadvantage is that primary methods can be resource intensive.

Secondary Methods

• Secondary data collection methods use data that was originally collected for a different purpose
• A benefit is that the data already exists.
• A disadvantage is that available data may be limited.

Research Methodologies

• Primary methods: Biological Assessments, Observations, Self-reports
• Secondary methods: Medical Records, Medical Claims, Research Data

Biological Measures

• Biological assessments use biophysical, biochemical, or microbiological methods.
• Clinical or laboratory tests require specialized instruments such as electrocardiograms, glucometers, or microscopes.
• Objective markers of a disease or patient's health play a critical role in clinical research.

Biophysical Assessments

• Biophysical assessments measure physical characteristics like bone density, blood pressure, and forced expiratory volume.
• Biophysical devices include blood pressure monitors, x-rays, computed tomography (CT), electrocardiograph (ECG), ultrasound, and magnetic resonance imaging (MRI).

Biochemical Methods

• Biochemical methods measure chemical constituents in body fluids like blood or urine.
• Examples include blood glucose, urine creatinine, and serum drug levels.
• Clinical tests require chemical analysis or assays using instruments like chromatography systems.

Microbiological Methods

• Microbiological methods evaluate microorganisms like bacteria or viruses in body fluids.
• These use growth of bacterial culture with evaluations involving microscopic examinations.

Observations

• Observational techniques are used to gather data on participant activities, characteristics, communication, interaction, and time taken to complete a task.
• In obtrusive observation, the participant is aware of being observed and may alter behavior.
• In unobtrusive observation, the participant is unaware of the observer, which may raise ethical and legal concerns.

Self-Reports

• Self-reports collect data by directly questioning research subjects to gather data regarding patients’ thoughts, perceptions, attitudes, and behaviors. Topics for self-reports include:
  • Psych distress
  • Rx satisfaction
  • Pain/fatigue
  • Overall quality of life

Self-Reports - Surveys

• Surveys use a survey instrument, which includes open-ended or close-ended questions aimed at collecting data relevant to a specific topic.

Close-Ended Questions

• Close-ended questions provide respondents with specific response choices from which they must select the answer.
• Examples include: "Did you experience any side effects from the medication?" with the available answer choices of "Yes" or "No,"
• Benefits: The questions are easier and take less time to answer.
• Cons: They limit option choices.
• Most research surveys involve close-ended questions for ease of coding and analysis.

Open-Ended Questions

• Open-ended questions let patients/participants write responses in their own words without answer choices.
• These questions enable researchers to gather in-depth information on patient experiences and opinions.
• The disadvantage is that they are challenging and time-consuming for patients and unsuitable for statistical approaches to data analysis.

Surveys - Administration

• Surveys are administered through different modes:
  • Mail
  • Internet
  • Telephone
  • Face-to-face interviews are discussed in the section titled 'Interview'
• Mail surveys involve mailing the questionnaire along with a cover letter and postage-paid return envelope to a sample of people.
• Online or internet surveys use web-based survey solution systems such as SurveyMonkey and Qualtrics.
• Surveys are exemplified by the SF-12 Health Survey and the Multiple Sclerosis Impact Scale.

Research Methodologies - Guiding Factors

• A decision to use a particular data collection technique is guided by three factors:
  • Reliability
  • Validity
  • Practicality

Reliability of Survey Instruments

• Reliability is the consistency and reproducibility of results.
• Results from the measuring instrument should be consistent when measured repeatedly over different time periods.
• Test-retest reliability is the extent to which answers to the same instrument correlate when measured in the same sample over different time periods.
• Inter-rater reliability is the extent to which results are consistent when the same measurement instrument is used by multiple investigators (reproducibility).

Validity of Survey Instruments

• Validity refers to the extent to which an instrument measures what it is intended to measure.
• Health care research often quantifies abstract concepts such as disease severity, drug efficacy/safety, burden of illness, patient satisfaction, and quality of life.
• Measuring an abstract concept requires translating the concept into more concrete indicators.

Interviews

• Face-to-face interviews can be structured, semi-structured, or unstructured.
• Structured (or standardized) interviews use the same set of questions presented in the same order to all participants.
• Unstructured interviews are non-standardized, flexible so the question-and-answer categories remain undetermined.
• Semi-structured interviews use structured questions with follow-up and/or clarification questions.
• Telephone interviews rely solely on verbal communication.
• Most research interviews are structured for ease of coding and analyzing.

National Health Interview Survey (NHIS)

• The National Health Interview Survey (NHIS) has monitored the health of the nation since 1957.
• NHIS data on a broad range of health topics are collected through personal household interviews.
• The U.S. Census Bureau has been the data collection agent for the National Health Interview (NHIS) Survey for over 50 years.
• Survey results have been instrumental in tracking health status, health care access, and progress toward achieving national health objectives.

Medical Records

• A confidential patient record is kept by a healthcare professional or organization.
• It typically contains:
  • The patient's demographic information
  • A summary of the patient's medical history
  • Documentation of each event, including symptoms, diagnosis, treatment and outcome
  • Other relevant documents and correspondence
• The main purpose of medical records are:
  • Providing a summary of a person's contact with a healthcare provider and the treatment provided to ensure appropriate healthcare.
  • Providing data for monitoring patient care, conducting clinical audits and assessing pattern of care and service delivery

Paper vs Electronic Medical Records

• Medical records are traditionally paper-based.
• An EMR is a generic name for a digital version of a patient's paper medical records.
• EMRs can include medical history, physical examination, diagnoses, clinical activities, lab results, radiology imaging and results, and electronic prescribing data.

EMR for Clinical Research

• EMRs harbor a wealth of health-related information.
• EMRs provide access to a broader and more diverse patient population.
• EMR systems have integrated data and could speed up data acquisition and searching.
• EMR systems allow mass sampling and computing.

Health Insurance Claims Data

• Health insurance claims data - Claims, Costs, Diagnosis, Discharge, Pharmacy data:
  • Designed for billing purposes not research
  • Have detailed information on clinical encounters, costs, diagnosis, admission/discharge, pharmacy dispensing data
  • These make claims data very useful for clinical research as well

Health Research Databases - Examples

• National Health Interview Survey (NHIS): https://www.cdc.gov/nchs/nhis/index.htm
• Medical Expenditure Panel Survey (MEPS): https://meps.ahrq.gov/mepsweb/
• National Health and Nutrition Examination Survey: https://www.cdc.gov/nchs/nhanes/index.htm
• Surveillance, Epidemiology, and End Results Program (SEER): https://seer.cancer.gov/resources/