Third Millennium Analytics, Inc. provides research and analytical services for a variety of health-related organizations, including public health departments and agencies, hospitals and health clinics, mental health facilities, HMOs, health insurance providers, and the pharmaceutical and medical device industry.   Assistance can be provided with the design and analysis of clinical trials, as well as evaluations of social marketing efforts, secondary prevention interventions, and randomized trials.  Demographic techniques can be applied for various health administrative purposes, and biostatistical methods applied for the analysis of epidemiological data. 

Health Program Evaluation.  Whether your health program or service is designed for health promotion, risk reduction, or secondary prevention, we can design and implement evaluations for your interventions, randomized trials, or social marketing campaigns.  These programs include those designed to reduce health risks (e.g., smoking cessation, weight-loss, childhood obesity prevention, drug abuse treatment), to improve health-protective attitudes or behaviors (adoption of low-fat diets, exercise regimens, safer sex practices, etc.), as well as treatments to improve mental health statuses.  Secondary data analysis can also be conducted on previously collected data.  Cost-efficiency assessments are available for health services provided to patients whereby the focus is on the measurement of health benefits of treatments relative to the costs of producing them.  For businesses trying to promote preventive health care to control costs, assistance can also be provided to evaluate programs designed to improve or promote employee health or on-the-job safety practices.  For more details on our evaluation research services, please [click here].

Clinical Research.  As a CRO, we can assist with the design and analysis of efficacy and safety assessments for Phase I-IV drug or medical device trials.  We begin by gaining a thorough understanding of the clinical and scientific principles of each project, including all institutional and human subjects protocols, and follow CDISC standards (ODM, SDTM, ADaM formatting, etc.).  Our assistance includes the selection of the most appropriate research designs (multicenter, active control, combination, equivalence, factorial designs with multiple treatments, etc.), statistical designs (crossover, parallel, titration, enrichment, etc.), and guidance regarding when to beneficially apply stratified analyses.  We also provide guidance regarding various issues encountered with efficacy assessments e.g., baseline comparisons, the proper treatment of dropout and missing data (intention to treat per protocol, last observation, multiple imputation), and covariate adjustment (demographic or baseline measurements, or concomitant therapies, data monitoring).   We are familiar with various methods of detecting and controlling for confounding, binomial methods (exact and/or asymptotic), and methods for closed cohort data (e.g., odds ratios, risk ratios, risk differences).  Assistance is also available for monitoring and group sequential procedures for interim analyses, special analyses for multicenter studies, quality of life assessments, and analyses of safety data (e.g., extent of exposure, coding and analysis of adverse events and laboratory data).

Epidemiology.  Biostatistical methods can be applied to analyze epidemiological data and identify risk factors (e.g., demographic, risk exposures, behavioral) related to the occurrence and distribution of infectious or chronic disease.  Regardless of the study design (open or closed cohort, case control, etc.), we will apply the most appropriate method to identify risk factors and interactions while controlling for extraneous or confounding factors.  Appropriate techniques can be applied for closed cohort data (e.g., binomial, odds ratios for stratified and unstratified data, risk difference methods), survival or open cohort data (e.g., proportional hazards models, and Kaplan-Meier, actuarial, and Poisson methods), and case control data (e.g., odds ratio methods for open and closed cohorts, methods for matched data).  We can properly calculate standardized rates, and apply logistic regression (e.g., for a closed cohort study with a continuous exposure variable), Cox proportional hazards models (e.g., for an open cohort study where exposure is dichotomous and separate hazard functions are produced for the exposed and unexposed cohorts), and Mantel-Haenszel tests and estimates.

Health Care Administration.  We can conduct evaluations of health policies and patient satisfaction, secondary analysis of large scale medical databases through the application of data mining, and assist with analyses of clinical protocols, chart reviews, audits of data, and transcriptions.  We can apply demographic techniques to derive population estimates and projections, which are essential for planning future health care needs, including the siting of future health care facilities, potential public health emergencies, etc.  Population size, concentration, distribution, and rates of change are important factors to consider for meeting thresholds and efficient distribution of health services and resources.  When assessing health service needs, population composition is essential to consider, which includes both biosocial factors (e.g., age and sex structure, race, and ethnicity) and sociocultural factors (e.g., marital status, family structure, income and educational levels, labor force characteristics).  Projections of population change are also very helpful for the planning of future health care needs (e.g., for geriatric care, ambulatory obstetric services, financial assistance).  We can highlight major fertility, migration, mortality and morbidity trends and patterns, which have pertinent implications for the demand of medical care, drugs and supplies, health facilities and personnel, and the allocation of resources for preventive medicine.  Assistance can also be provided for studies to identify various correlates of health status, health behavior, and health services utilization for your service area.