The importance of quality in item generation; a prerequisite for Rasch analysis

Stephen McKenna & James Twiss

Galen Research Ltd, Manchester, UK

Email: [email protected]

Quality in item generation; a prerequisite for Rasch analysis

Creating a scale

Let’s create a new scale ……..

We need to write a lot of items and then apply Rasch analysis

Introduction

• But what will it really measure?

• How relevant will the items be to respondents?

The value of Rasch analysis is in perfecting the scale- not in generating its content

Random item generation

Impact of MS Impact of MS

QoL

QoL

Func

Satis

Symp

Targeted item generation based on model of construct

Impact of MS

QoL

Impairment

• Loss or abnormality of psychological, physiological or anatomical structure or function

• Equates to symptoms

• Disturbances at level of organ

• Fatigue, pain, dizziness, depression, sleep problems

Disability (activity limitation)

• Any restriction or lack of ability to perform an activity in the manner or within the range considered normal for a human being

• Equates to functioning or functional status

• E.g. restricted mobility, problems dressing & bathing, social restrictions, problems showing affection

• HRQL measures (such as SF-36) commonly assess functioning in addition to impairment

What is quality of life?

• Not symptoms, activity limitations, utility or clinical

outcome

• Uniquely patient-based

• Holistic – overall impact of disease and its treatment

on the patient

• Unidimensional

“Human life quality is dependent upon the satisfaction of certain basic needs - lack of disease, mobility, adequate nutrition and shelter.”

Sir Thomas More (1478-1535)

Item generation

• Appropriate source of items

• not existing measures

• not the literature

• not test-developers

• not clinicians in most cases

• relevant patients

Item generation

• Qualitative unstructured interviews

• focus groups do not work well

• do not start with a list of issues – only constructs

• let the interviewee guide the interview as much as

possible

• audio-record and produce transcripts

• invest a great deal of time analysing transcripts

Item reduction

Ensure items assess the required construct. Remove items that:

– are poorly worded

– express more than one idea

– appear difficult to translate

– are personal to one interviewee

– will not apply to all respondents; sex, employment etc

– are likely to produce DIF; interest in shopping, attending gyms etc

– are specific to one national culture

– are facts or will not change: ‘I worry that my health will get worse’

PRIMUS

• Patient Reported Indices of Multiple Sclerosis (PRIMUS) • consists of 3 scales:

• Symptoms (impairment)• Activity limitations• Quality of life

Sample items from the PRIMUS QoL scale

• I feel I am a burden to others

• I have little freedom to do what I want to do

• I have to push myself to do things

• I avoid physical intimacy

• I feel as if I have nothing to offer anyone

Rogues gallery – SF-36

• Does your health now limit you in vigorous activities, such as running, lifting

heavy objects, participating in strenuous sports?

– that rules out 95% of the British! Disliked by older respondents

• General Health - How TRUE or FALSE are these statements for you?

– I seem to get sick a little easier than other people

– I am as healthy as anybody I know

– I expect my health to get worse

– My health is excellent

Are these outcomes?

MCS and PCS

Rogues gallery – EQ-5D

5 items – adequate coverage?

Mobility

I have no problems in walking about

I have some problems in walking about

I am confined to bed

There seems to be a gap here!

Anxiety/depression

I am not anxious or depressed

I am moderately anxious or depressed

I am extremely anxious or depressed

Are anxiety and depression the same constructs? What if I am anxious and

depressed? What if I am moderately anxious and severely depressed?

Rogues gallery - WHOQOL

Do you get the kind of support from others that you need?

Are you able to accept your bodily appearance?

How available to you is the information that you need?

How satisfied are you with your access to health services?

How satisfied are you with your transport?

How safe do you feel in your daily life?

Not health outcome as intended but static measure of QoL?

The few relevant items are also poor:

How satisfied are you with your sex life?

How often do you have negative feelings, such as blue mood, despair, anxiety, depression?

Rogues gallery - DLQI

Over the last week, how much has your skin interfered with you going shopping or looking after your home or garden?

Over the last week, how much has your skin created problems with your partner or any of your close friends or relatives?

8/10 items can be answered not relevant (and scored ‘no problem’)

Measure used to decide who should receive biologic treatment for psoriasis

DLQI (not at all, a little, a lot, very much)

DLQI: DIF by gender

Influences the clothes you can wear

Expe

cted

val

ue

Person locations (logits)

FemalesMales

Hypomanic Personality Scale: DIF by gender

‘I find it easy to get people sexually interested in me’

FIS: DIF by age I have to rely more on others to help me or do things for me

CES-D: DIF by country

‘I was happy’

DLQI: DIF by disease

Prevents working or studying

Person locations (logits)

Expe

cted

val

ue PsoriasisAtopic Dermatitis

Conclusions

Quality of patient reported outcome scale depends on:

• Coherent and valid measurement model

• Quality item generation

• Simple response format

Then we can think about fit to Rasch model

Great - now we can make an excellent scale with the help of Rasch!