Content Analysis

This is about analysing the text of documents such as letters, patient experience comments and complaints, and website comments. 

Pros and Cons

Number of people: this method can be used to consider thousands, or millions of records at a time.

Useful for

• This could be to collect quantitative data - for example 'how many of our patient information publications use the phrase 'suffering with dementia', rather than more positive language? 
• This could be to collect qualitative data - for example looking for themes or major ideas in documents.

Disadvantages

• The computer cannot determine what the word meant to the person who used it.
• Dementia may change the way people use words and language - perhaps noting down a word other than the one they intended, or using a word and attributing a different meaning to it than would generally be expected.
• there is potential for error. For example, if records were searched for particular words to learn whether service users had a positive experience and outcomes, the word 'benefits' might be wrongly understood to relate to this, whereas it might be that a top need for help service users had was to get help about their welfare benefits entitlements.

Cost: this can be very expensive as it may require using particular technology or buying in experts as consultants.  

Time: this may take time to set up

How it can be done

• Key Word in Context ("KWIC") is usually done by computer.
  
  For example a web-based search for the word "rude" in comments patients have made in patient experience surveys might reveal that it occurs significantly more about receptionists, than about doctors (real example - not dementia-specific). 

Real-life example featured in this resource

Mapping the Dementia Diagnosis Journey to improve self-guided support