3.3.13.1 Developing categories

A category is defined as a brief description of a key concept arising from the aggregation of two or more like findings, based on the similarity of meaning and representation of the phenomenon of interest. The aim of this aggregation is to integrate the meaning, characteristics and attributes of the grouped findings. The objective of categorisation is to facilitate a meaningful synthesis. All categories are to be accompanied by a description of each category.

A category and its accompanying description convey the whole, inclusive meaning of a group of similar findings. The description avoids new interpretation of the author’s original explanation. Commonly, a single finding is assigned to only one category. However, qualitative findings can be multi-dimensional and may reasonably relate to more than one category. Restricting the finding to a single category in such cases could oversimplify complex phenomena or obscure important connections. Therefore, it may be reasonable to present a finding in more than one category. Caution should be exerted to avoid unintended duplication or overemphasis, and the process should be clearly documented to ensure transparency and consistency in the decision-making process. 

Developing a category description requires consensus among the members of the review team to accurately describe the category capturing the range of findings. The goal is to develop a comprehensive and explanatory statement that informs the key concepts within each category. The process of developing and describing categories should be clearly articulated, with any deviations from the protocol explicitly explained.

Responding to divergent data

Findings that cannot be categorised within the developing synthesis are referred to as divergent data. Divergent data may be a result of insufficient extraction and analysis, or they may represent a nuance of experience related to a specific population or context. If divergent findings are identified, the review team should consider the following:

• Review the extraction of the findings and whether the findings relate to the systematic review question.

• Review the categorisation process.

• Consult an expert qualitative researcher to review the categorisation process.

• Do not try and ‘force’ a finding to ‘fit’ into the analytical scheme.

• Consider features of the population or context (related to the findings in question) that may explain why the findings appear to be different. The narrative text should then explore and explain the differences where possible.

• Carry the single finding into a single category, with the intention of integrating it at the level of synthesised finding. This option should be preceded by the above recommendations.

Examples of categorisation

The examples below explain how a category can be described.

Example 1

Category: A confirming relationship matters (Norberg et al. 2019).

Accompanying description: The patient–provider relationship was a strong incentive to either remain or drop out of care. What patients did not want is clear—a care team that acts in a judging, demeaning and dismissive way. Patients knew what they wanted from this relationship—to be a whole person, to receive respect and genuine empathic caring no matter the circumstances, to be listened to and be heard and to have a trusting and accessible connection with the care team. This humanistic response from providers helped patients deal with their diagnosis, troubleshoot disease management problems and achieve acceptance in adherence. Care could be disrupted with paternalistic, judgemental approaches.

Example 2

Category: Dependence on the system (Allen & Vottero 2020).

Accompanying description: Homeless women believe that their treatment by healthcare providers is based on their dependence on a system that is not equipped to improve the health of those who do not have health insurance or the means to afford healthcare. Many homeless women are without the necessary identification required to obtain health insurance, which further impedes their ability to access care. Economically driven healthcare looks negatively at patients who cannot pay for services; therefore, access is denied or, perhaps even more problematic, improper or lacklustre care is provided. Unfortunately, this poor care is accepted and expected by homeless women, which perpetuates their deficient usage of healthcare services.

Example 3

Category: Experiences and needs of patients and parents for the development of self-management (Min et al. 2022).

Accompanying description: Six findings are grouped into this category, which describes the challenging experiences of patients and parents and the obstacles they encounter in the development of their self-management of juvenile idiopathic arthritis. These experiences and obstacles include lack of information, difficulties in medication management, attending hospital visits and surveillance of symptoms. Knowledge of pain management and drug effects and access to appropriate vocational assistance can facilitate self-management.

Example 4

Category: Personal growth (McCloskey et al. 2023).

Accompanying description: Post-licensure practical nurses (PLPNs) who return to school to become registered nurses must overcome many social, familial, financial and work-related challenges in addition to meeting the high academic demands associated with bridging programs. Being able to maintain their role as student amidst other competing demands helps students to become more confident, satisfied and independent individuals. As expected, completing a bridging program provides PLPNs with knowledge and experience, but students report learning about their own strengths and abilities as individuals.