Editor's note: Dahl Gardner is vice president of quality operations, Paria Group, an Orem, Utah research firm.
The growing popularity of managed care coverage and increasing competition and governmental pressures are forcing health maintenance organizations (HMOs) to become more sensitive to customer satisfaction. To find out how to improve service and satisfaction, many HMOs are turning to market research.
With research potentially playing a larger role in HMO decision making, Paria Group, an Orem, Utah research firm, conducted a study to determine the current status of market research in HMOs.
The data was collected from May 16, 1994 through September 15, 1994 at Paria Group's facility using computer-assisted telephone interviewing software. Respondents were identified as the person most knowledgeable about market research in the organization. Respondents who participated in the survey were sent a copy of the executive summary. The study was conducted with representatives of 225 HMOs which were randomly selected from all of the HMOs within the United States (546).
The main objectives of the study were to determine:
- the structure and role of market research within the HMO;
- which research methodologies were used and their level of usage; and
- how the HMO's annual research budget or annual gross revenue affect the research function.
Of the 225 HMOs surveyed, the study found that 99 percent (223) conduct market research on a regular basis. Nearly half (49 percent) have a central market research department. Over one-third (35 percent) have established written policies or guidelines regarding market research.
Gross annual revenue, annual research budget, and plan size (number of covered lives) were found to be important factors affecting market research. The distributions from our sample are shown in Graphs I, II, and III.
As you can see in Graph I, the gross annual revenue is highly concentrated in the over-$10 million range (68 percent). Graph II shows that the annual research budget is concentrated at the under-$20,000 range (23 percent) and the over-$80,000 range (27 percent). The mean of plan sizes is about 40,000 covered lives, with most plans at 50,000 or less, as shown in Graph III.
When researchers at HMOs were asked how important market research was for various purposes, most (84 percent) indicated that market research was important for "determining customer satisfaction, which contributes to continuing quality improvement." Other responses are summarized in Table I.
Research organization
There are a variety of approaches for initiating, organizing and conducting research. They include: a central market research department; a marketing or service department; joint research conducted by several departments; separate research conducted by each division or department; individuals who have the responsibility for research; an outside supplier; and/or combinations of several approaches.
The central market research department is the dominant approach and is used by approximately 49 percent of respondents. Many other HMOs indicated that they plan to establish central market research departments in the future.
Another related trend in research is for HMOs to use contract services in conjunction with in-house personnel rather than using solely in-house personnel. For example, 48 percent of HMOs use a combination of contract services and in-house personnel for project design, compared to 40 percent who use in-house personnel only. Graph IV clearly illustrates this. Similar trends for data collection and data analysis are shown in Graphs V and VI.
Graph V shows the distribution of HMOs in data collection: 36 percent use in-house personnel only, 34 percent use mixed sources, 28 percent use outside vendors.
Graph VI shows the distribution of HMOs in data analysis: 46 percent use in-house personnel only, 39 percent use mixed sources, 11 percent use vendors and 3 percent use outside consultants.
Several trends become apparent when the responses are broken down by annual market research budget. For instance, at HMOs with annual research budgets of $20,000 or less, there is a greater tendency for the marketing department or the vice president of sales/marketing to initiate research projects. They were more likely to use in-house sources for design/planning, data collection and data analysis. They had the highest usage of mail surveys and the lowest usage of CATI surveys. They were more likely to rate market research as being less important in conducting test marketing before new products and services are introduced. They were also less likely to have made changes in the market research they conducted during the past year.
HMOs with research budgets under $40,000 also rated the following tasks as not very important for market research: test marketing new products or services, tracking studies, and business and consumer image assessments.
On the other hand, HMOs with budgets between $40,000 and $60,000 were more likely to turn to a market research department to initiate market research. They were also more likely to indicate plans to increase or intensify their use of market research in the future.
HMOs with budgets from $60,000 to $80,000 tend to be more likely to use outside sources for data collection. However, they were also more likely to use mixed sources (in-house and contract services) for data analysis.
HMOs with research budgets greater than $80,000 were more likely to have written market research policies or guidelines. They tended to use contract services and in-house sources equally for data collection while they tended to use mixed sources for data analysis. They were more likely to use CATI surveys and less likely to use mail surveys for data collection. They were also more likely to contract market research through an advertising agency. They tended to place a high degree of importance on market research in determining member satisfaction. Finally, they were more likely to consider market research important in conducting test marketing of new products or services.
HMOs with annual revenues of $10 million or more tended to have annual market research budgets of $80,000 or more. They were more likely to have department heads making decisions regarding market research.
The research found significant positive correlations between the gross annual revenue and the importance the HMO attaches to using market research to determine member satisfaction. This correlation was also found between gross annual revenue and test-marketing new products and services and evaluating advertising effectiveness. Of little surprise was the correlation between the gross annual revenue and the size of the annual research budget and how well the research function is defined. A negative correlation was found between the gross annual revenue and the use of mail surveys.
HMOs with higher gross revenues were more likely to have established research policies or guidelines (see Graph VII). However, 65 percent of HMOs do not have written research policies or guidelines.
Table II shows the results of a Chi-square test using 95 percent significance level indicating that:
- HMOs without policies/guidelines or individuals responsible for research are more likely to use outside sources for research design (61 percent).
- HMOs with individuals responsible for research are more likely to use outside sources for data collection (28 percent).
- HMOs with neither individuals responsible for research nor policies and guidelines are more likely to use mixed sources for data collection.
- HMOs with policies/guidelines on market research are more likely to use in-house personnel for data analysis; other HMOs are more likely to use outside sources for data analysis.
Table III shows the results following a Chi-square test, which found that:
- HMOs with a central market research department are more likely to use outside sources (61 percent) and less likely to use in-house personnel (39 percent) for data collection;
- Research done by each department is less likely to use outside sources for data collection (9 percent).
- If individuals are responsible for research, they are more likely to use in-house sources for data collection (20 percent).
Mail most popular
Mail surveys are the most frequently used method of data collection: nearly half (48 percent) of all HMOs said they used mail surveys for 41 percent or more of their research. The second-most frequently used method of data collection is computer-assisted telephone interviewing: 28 percent of HMOs said they used CATI surveys for 41 percent or more of their research. Thirteen percent of HMOs use focus groups for 41 percent or more of their research. Panel studies are used by three percent of the HMOs for 41 percent or more of their research (see Table IV).
Research budget
As discussed previously, there is a strong positive correlation between gross annual revenue and annual research budget. Table V provides a summary.
Table VI shows the correlation between the annual research budget and how well the research function is defined.
Table VII breaks down the budget effects on research design, data collection and analysis. Strong correlations are revealed: a lower budget draws each phase inward (the HMOs tend to use more in-house personnel) while a higher budget pushes research outward (they turn to outside sources or mixed sources).
More attention needed
It appears that the market research function needs more attention by top management to ensure that the function is properly organized, objectives are well defined and that the HMO's research is appropriate given those objectives. Second, to ensure quality and improvement of research, a corporate policy should be developed and implemented for those HMOs (65 percent) who do not have one yet. Creating a central market research department appears to aid development and implementation of research policies and helps management cope with an increasing number of research projects. Finally, it would be extremely beneficial to establish some models and/or quantitative indexes to measure research efficiency, effectiveness and progress.