3M has developed a composite measure, a3M Value Index Score, which provides a st

Question

3M has developed a composite measure, a3M Value Index Score, which provides a standard metric to assess accountable value in health care. This index care domains derived from 16 measures of key processes and outcomes that effectvalue in health care. According to 3M, this Value Index Score can increase the understanding f provider and system performance thereby accelerating and prioritizing the areas where improvement is needed. This measure can be used by both providers and payers to help improve patient outcomes and to control costs.Suppose you were asked to develop a sampling plan for a study to determine the value index, and general attitudes toward the index, what sampling plan would you use?What is the target population(or are there multiple target populations)?Which of the four types of random sampling discussed in the chapter would you use and why? If you were to use stratified random sampling, what would be your strata?

Explanation / Answer

The best way to approach this problem is to go with either stratified or cluster sampling. Among the two cluster samping should be the slightly preferred option

Population in healthcare is very diverse in nature, and all types of categories in terms of age group, body types etc. need healthcare services, therefore making a lot of sub-groups. To counter this problem, stratified clustering can be done, where in various sub-groups cab be clubbed together, based upon similarities between them, to make a single sub-group. Such an activity would create stratas and elements for sampling can then be selected randomly selected from various stratas. This results in stratified sampling.

My stratas in this case would be based on age groups, i.e., 20-30, 30-40....70-80. The target population would be people in the age group of 20-80 years.

In case of random sampling, I would go for cluster sampling. Cluster sampling is very close to stratified sampling except that it selects few sub-groups randomly (instead of using all the sub-groups), and then survey all the entities in those sub-groups. Such an activity would ensure that there's no error at the lower level as I am surveying all the entities in a sub-group.

Related Questions

Navigate

• Browse (All)
• Browse 3
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.