Deploying predictive analytics to aid with population health management sits atop the wish list of many hospital chief information officers and top health system executives.

But hospitals are more wishing to use predictive analytics than actually doing so, says a new study from Health Catalyst.

Predictive analytics uses data mining, statistics, modeling, machine learning, and artificial intelligence to analyze current data to make predictions. In healthcare predictive analytics is a chief tool for rolling out population health management.

Population health management is the integration of patient data taken from multiple health information systems technology resources and merged into a single and accessible patient record. Population health management tools are used to to improve clinical and financial outcomes for patients.

A recent survey of 136 CIOs, chief executive officers, chief financial officers, chief medical information officers and others reveals that 80% of executives believe predictive analytics can help population health management programs to solve recurring and expensive problems such as preventable readmissions and contracts with health insurance companies that pay less than the cost of providing care.

But only 31% of hospital executives say they currently use predictive analytics. Whats more only 38% of executives surveyed say they will deploy predictive analytics within three years including 12% within one year.

Overall, the survey findings point to a growing need within the provider community for solutions that help to identify long-term rising-risk patients who are on their way to becoming high-cost consumers of heath care, says Health Catalyst director of data science Levi Thatcher. With an ever greater light being cast on system-wide inefficiencies, providers are hungry for analytics that will help them identify and treat these patients before their health deteriorates, both improving their lives and reducing needless spending across the system.

32% of healthcare executives say lack of appropriate tools, data and a technology infrastructure is the top reason for not immediately rolling out a predictive analytics initiative followed by 26% and 20%, respectively, which cite a lack of employees with the right skills and a lack of executive and financial support as the chief reasons for holding off.

Even among those who plan to adopt predictive analytics, few said they have the budget to allocate significant resources to the effort, Thatcher says. Most (37%) are tip-toeing into the space with commitments or planned commitments of one to three people devoted to the task of leveraging analytics for predictions.

The main reasons hospital executives give for wanting to deploy predictive analytics would do so would be to alert caregivers to interventions that may prevent health declines among high-risk patients at 58%, followed by predicting financial outcomes such as patient cost or the likelihood of patients to pay their bills (52%), improving the ability of hospitals to negotiate contracts with insurers (42%), better project patient health outcomes and satisfaction (38%), improving the quality of diagnoses (33%) and forecasting staffing and supply chain needs (27%). Respondents were allowed to choose multiple answers.

To build up a population health management database and extract data using predictive analytics software, 80% of respondents said they would first use clinical data accessed from their electronic health records system, followed by claims and patient outcomes data at 53%.

Health Catalyst develops data warehousing, analytics and outcomes tools and software for healthcare organizations.