Plan for data collection
Plan for data collection
Gathering data can cost a bundle
By Patrice Spath, ART
Brown-Spath Associates
Forest Grove, OR
It’s a common occurrence — a medical staff committee, performance improvement team, or hospital department asks the quality management staff to gather information for a special study. To ensure that the study yields worthwhile information, it is important that several questions be answered by the people who are requesting the data. Without a clear understanding of the purpose and scope of the study, the quality management department may gather data that do not answer the questions being posed by the users of the study results.
The questions that must be considered before beginning data collection are listed below. The group that will be analyzing the study results comprises people who must answer these questions. The more specific they can be in answering each question, the easier it will be for the quality management staff to design a worthwhile performance measurement project. Sample answers for an outcome study involving patients who underwent hip replacement are illustrated in italics.
• How do you plan to use the data?
The study oversight group should articulate its purpose for performing the study, such as what it hopes to discover about an aspect of health care performance that it doesn’t already know. An objective, narrowly focused purpose statement is a vital first step for the data collection plan.
Our purpose is to determine the short-term (six weeks post-op) functional status of patients undergoing hip replacement during 1998 and determine if there are differences in outcomes related to patient age and discharge disposition.
• What are the primary questions you expect to answer with this study?
To prevent the collection of "wouldn’t it be nice to know" data, the study group should focus its attention on no more than three important questions to be answered by the evaluation. Without a clear direction, the group could find itself with an abundance of information that doesn’t tell it what it wishes to know.
Three questions to be answered by this outcome study:
— What is the short-term functional status of patients undergoing hip replacement in 1998?
— Does patient age affect short-term functional status?
— Does patient discharge disposition affect short-term functional status?
• What patients will be included in the assessment?
What inclusion and exclusion criteria will be used to determine the eligibility of study participants? Determine the patient population to be included in the study. Consider age category, gender, principal and secondary diagnoses, principal and secondary procedures, level of care setting — inpatient or outpatient — providers, and other characteristics.
For the hip replacement outcome study, all patients undergoing hip replacement in 1998, excluding those who expired post-op and those admitted from and discharged to a nursing home, will be included in the evaluation.
• How many patients are expected to be included in the assessment process?
Does this population represent an adequate sample size? In general, if the study population is small — fewer than 30 cases — the entire population should be studied. As populations grow larger, sample at least 10% of the affected population or 50 patients, whichever provides the larger sample. Remember, even when all patients from a particular time period are being studied, this population is still considered a sample from all possible cases that could be studied.
The group conducting the hip replacement study determines that, based on historical data, 15 patients per month will meet the inclusion criteria. It should be expected that in 12 months, 180 patients will be included in the study.
• To answer the questions posed by the study group, what processes are of primary interest?
If the study group wishes to determine the effect of certain health care processes on patient outcomes, it is important that the group delineates the activities of interest. If the answer to this question is not already clearly answered by the group’s original study questions, find out what process-related data need to be gathered.
The hip replacement study group is primarily interested in the relationship between patients’ discharge disposition and patients’ functional outcomes. To ensure all relevant data are gathered, the group selects the following disposition categories they want information about: discharged to subacute care/rehab facility, discharged to home with home care, discharged to home with rehabilitation provided in outpatient clinic, discharged to home without home care or outpatient rehabilitation.
• How will data about the processes of primary interest be gathered?
Will you be using an existing data source, or will a new data source need to be developed? Although it may seem as if every possible data element is now being gathered by hospital information systems, there may be times when the data needed for a particular study are not readily available. First determine if the necessary data in an existing information system are accessible. Check to be sure the data definitions are consistent with what you need for the study. For example, if you are conducting a study of nosocomial infections, most infection-related ICD-9 codes (for example, 599.9, urinary tract infection) don’t specify whether the infection originated in the hospital.
The hip replacement study group wants to know very detailed information about patients’ discharge disposition. The disposition codes in the hospital’s financial systems are too general to allow them to be used in the study. A new instrument will need to be designed to capture disposition details. To obtain the needed data, it may be necessary to review records of discharged patients.
• To answer the questions being posed by the study group, what outcomes are of primary interest?
The outcomes of interest may be well defined by the group’s original study questions. If not, now is the time to obtain further details.
As evidenced by the study questions, the hip replacement study group is interested in patients’ short-term (six-week post-op) functional status.
• How will these outcomes be measured?
This is the time for the study group to be sure it has supplied an objective definition of the outcomes it is interested in learning more about. Terms such as "quality of life" and "functional status" may have very different meanings for each member of the study group. That’s why it is important not to make assumptions about the data definitions.
Outcomes are the results of the medical care process. Outcome assessment emphasizes the effect of process on patients’ well-being and quality of life to determine how the patient was affected by the process of care variables.
The study group should objectively define what it means by the term "outcome." The outcomes to be measured can include the clinical gains made during a specified time period, satisfaction that patients experienced with treatment processes, and the long-term benefit of a given treatment course.
There are many types of outcomes that can be measured:
• physical outcomes — biological indicators such as visual acuity post-cataract extraction;
• clinician-reported patient outcomes — the caregiver’s judgment of how well the patient is doing clinically;
• patient-reported outcomes — how satisfied patients are with their quality of life restored and services rendered;
• financial outcomes — the cost of the health care intervention and the cost-benefit of the treatment course as compared to the outcome.
In the hip replacement outcome study, the focus is on patients’ short-term functional status. The study group determines that the following elements of functional status will be measured six weeks postoperatively: patient-reported limitations in physical activities, limitations in social activities, and limitations in usual role activities.
• Will you be able to use an existing data source, or will there be a need to develop a new data source to gather the needed outcome data?
When measuring immediate outcomes of hospital care, it is likely that many of the data elements can be found in existing data sources. Mortality, length of stay, cost of care, and other outcome data are data elements in most clinical-financial databases. Patient-reported outcome data, both immediate and post-discharge, are harder to obtain because such information is not routinely among the gathered data. In these instances, new data-gathering instruments must be developed.
Use existing instruments when possible
Many facilities are using long-term patient outcome data collection instruments that are already developed and tested by health service research. By using existing instruments, the facility can save the time, labor, and expertise involved in developing its own valid and reliable data collection instruments. Also, using a standardized tool allows the facility to compare its results with those of other providers. Don’t spend time designing your own outcome measurement tools when a number of already-validated survey instruments are available.
At the hospital where the hip replacement outcome study is being performed, post-hospital caregivers are not currently using any instrument for gathering short-term functional status data for patients undergoing hip replacement. Therefore, the study group agreed to evaluate already-developed patient-reported functional status measurement tools and select the instrument that would be best for gathering the data elements needed for their study.
• Who will assume responsibility for collecting the necessary data?
In many instances, the quality management department can assume full responsibility for gathering needed data elements. This may include developing new data sources, reviewing close records, or transferring information from one database to another. However, post-discharge outcome data may need to be captured by people outside the hospital. For example, a study of postoperative wound infections may require involvement of physician office staff. The cooperation of providers throughout an episode of care is an important success factor in long-term patient outcome studies.
For the hip replacement study, orthopedic physicians agreed to administer a functional status measurement survey when patients returned to the clinic for the six-week postoperative checkup.
Setting the performance measurement study wheels into motion can be a labor-intensive process. Though it may take some time to get the study group to lay the groundwork for data collection, this planning process is time well spent. Use the checklist on p. 209 as a reminder of all the issues that need to be addressed prior to embarking on the performance measurement project. Gathering data costs money in terms of staff time and computer resources. To ensure effective use of your quality management resources, be sure that the people asking for performance data first answer these few critical questions.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.