Six Methods of Quality Assessment

Suppose an investigator (e.g., researcher or quality improvement manager) has identified a group of patients who were recently treated for urinary tract infections. He has access to data about their care, e.g., medical records or summaries of their care. According to Table 5.4, he could assess the quality of their drug therapy by six basic methods. Table 5.4 is based on a list by Brook et al. of five specific approaches for assessing health care quality (methods 1 through 5).14 Table 5.4 classifies these approaches according to their use of process or outcome data, and implicit or explicit criteria or review procedures. The table is completed by a sixth approach (explicit process and outcome criteria). The table also includes some examples to illustrate how each approach can be applied to assessments of the quality of medications use.


Approaches to Quality Assessment


Approaches to Quality Assessment

Drug Therapy

Drug Therapy

Implicit Criteria


Explicit Criteria



1. Was the

Was the drug

4. How well did

Did the drug

process of care


the process of



received by this

care meet

received by this


defined quality

patient meet

appropriate for



his needs at the





2. Could better

Does this

5. Were the

Is the adjusted

care have


outcomes in a

failure rate

improved the

failure or


higher in this




population than

occur when care

consistent with

in a population

was adequate?

outcomes obtained from scientifically validated processes of care?

of patients who received high-quality care?


3. Were both the

Would better

6. Were both the

See examples

process and the

prescribing and

process and the



drug therapy


MacKinnon and

consistent with


consistent with

Faris in


have avoided the problem in this patient?

defined criteria?

Chapters 2 and 3.

Source: Brook, McGlynn, and Cleary, N. Engl. J. Med., 335, 966, 1996.

Source: Brook, McGlynn, and Cleary, N. Engl. J. Med., 335, 966, 1996.

The investigator could assess process quality with either method 1 or method 4. The question with implicit criteria (method 1) is general: Was the drug therapy received by this patient, as documented in the record, appropriate for his needs at the time? The question asked under method 4 is more focused and refers to guidelines.

Drug use evaluation is a process of assessing the quality of prescribing according to explicit criteria for appropriate and inappropriate prescribing. (Drug use evaluation is discussed in Chapter 6.) From the perspective of quality-of-care assessments, most DUE activities would fit into method 4. Often, drug use evaluation employs drug choice criteria exclusively, although DUE criteria can, in principle, encompass dosage, appropriateness for a patient, etc. Still, prescribing is only a part of the medications use process.

From a quality assessment perspective, DUE is usually somewhat limited in scope. For example, process evaluations as described by Brook et al.14 require patient information. A common form of DUE is done, however, without diagnosis or any other patient information. For example, DUE may compare actual prescribing to a list of approved drugs, while the indication for the drug is assumed. This procedure would misclassify, as acceptable prescribing, all instances where a drug of choice had been prescribed, even if the drug had been unnecessary or in an inappropriate (nonindicated) therapeutic class. This method would also miss completely all instances where an indicated drug had not been prescribed.

Despite its name, DUE does not assess the quality of drug use, just prescribing. Furthermore, most ambulatory care DUE is done from prescriptions dispensed and submitted for payment. Occasions when the prescription was not needed, when a prescription should have been written but was not, or when the order was never carried out (e.g., where the patient should have filled the prescription but did not) would often be ignored. DUE ordinarily does not consider drug administration, consumption, or effects, and sometimes does not consider dose. DUE is a useful, but very limited assessment tool, even for prescribing. It should be supplemented by one or more methods useful for evaluating medications use, e.g., as listed in Table 5.4. DUE data may be useful for identifying potential problem areas in prescribing, but decisions based only on DUE data should be appropriately limited.

Methods 2 and 5 use outcome evaluations. With method 2, the review would ascertain whether the outcome was consistent with the reviewer's opinions, beliefs, and judgments about quality of care. In method 5, explicit outcome criteria would be used to review either individual records or population data. Outcomes are difficult to assess independently of process. Method 5 might be best for (a) identifying cases that should be followed up with another method, e.g., method 3; or (b) outcomes that are exceptionally good or bad. A sentinel event is defined as an outcome or other important occurrence that does not happen in the presence of adequate care. The question under method 2 asks, in effect, if an outcome was a sentinel event. In method 5, this approach is more explicit. It asks whether the prevalence of an outcome in the population being reviewed is consistent with the corresponding prevalence in a reference population.

With methods 3 and 6, the assessor attempts to assess process and outcome together. Method 3 requires a case-by-case review. The reviewer might, for example, identify patients with undesired outcomes, such as treatment failures or symptom recurrences, and then ask if the process of care caused the outcome, or if the outcome could have been better if the process of care had been more appropriate, according to the reviewer's opinions. This approach was the method used by most of the studies summarized in Chapters 2 and 3.

Method 6 uses explicit criteria to evaluate both process and outcome. This approach was not included by Brook et al., but it is a logical consequence of the arrangement in Table 5.4. This is the approach of the PDRM indicators described in Chapter 3, which includes many examples. (This will be discussed in more detail in Chapters 7, 8, and 11.) Review of both process and outcome, using explicit criteria, allows assessments of populations and could be used to identify patients for follow-up by another method. It could also be used to produce explicit process criteria with a wider scope than DUE criteria.

Was this article helpful?

0 0
Peripheral Neuropathy Natural Treatment Options

Peripheral Neuropathy Natural Treatment Options

This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.

Get My Free Ebook

Post a comment