Health-related quality of life among pregnant women living with HIV in Kenya, result of comparison of a patient-generated index and level 3 of Euroqol dimension 5 | BMC Women’s Health

The study was nested in a randomized intervention, the “WelTel PMTCT”, a trial investigating the effect of weekly text messaging on improving care retention over 2 years in HIV transmission prevention programs from mother to child (PMTCT). [25].

Study setting and population

The study was conducted in western Kenya at four government health facilities. Western Kenya bears much of Kenya’s HIV burden with HIV prevalence as high as 20% in some counties [26]. However, a great improvement in ART coverage has been noted since access to antiretroviral therapy (ART) has been expanded over the past 5 years [27]. Public primary health care clinics in Kenya routinely offer PMTCT services, integrated with other antenatal and postnatal services.

Eligibility criteria

Pregnant women (regardless of gestational age) over the age of 18, living with HIV, regardless of time since diagnosis, and whether or not they have started ART, who have given their consent informed written to participate, were eligible for the study.

Study participants

Among the 600 participants enrolled in the Weltel PMTCT trial [24] convenience sampling was used to enroll 100 participants. The sample size of 100 participants was based on other published studies with the PGI [28, 29]. Pregnant women living with HIV who fulfilled the eligibility criteria were consecutively recruited and participated in a structured interview. Mentor mothers and nurses providing routine care at clinics recruiting patients into the WelTel trial described the study to all women attending the clinic, most often those queuing for health staff for a routine check-up or ART refill. A participant who had agreed to participate in the PGI study interview was escorted to a private space within the same clinic for the interview by the research assistants (RAs) who had obtained informed consent from the woman . This was to provide privacy during the interview as well as to minimize disruptions to service delivery at the clinic.

Data gathering

Between July 2015 and April 2016, 100 study participants were interviewed using both the PGI and the EQ-5D-3L. They participated in a structured interview where the PGI and EQ-5D-3L questionnaires were administered by trained RAs familiar with the local health system and socio-cultural environment. The principal investigator (JM) trained the RAs on the PGI and EQ-5D-3L tools before the study. Piloting of the study tools was also done by the principal investigator prior to the trial with 10 pregnant women living with HIV attending a clinic in Kibera, an informal urban settlement with high HIV prevalence in Nairobi, Kenya .

The RAs administered the questionnaires primarily in Kiswahili and English and then filled in the study participants’ responses in the questionnaires. The ERP tool was completed in three stages through the interviewer-administered questionnaire. In the first step, each participant was asked to list the five most important areas or activities of their life affected by their HIV infection. In the second stage, each participant was asked to rate how much their life had been affected in each of the chosen areas, on a numerical scale from 0 to 6, with zero (0) being the worst they could imagine, at six (6) be as good as they could be. In the third stage, each participant was asked to imagine that they had the opportunity to improve the affected areas of their life. They were given a total of 10 imaginary points to “spend” on improving the areas they had identified as being affected by their condition in the first stage. The total points awarded in all areas could not exceed ten points. If no points were awarded, it meant that a patient wanted the area to remain exactly as it was. The points they assigned indicated the relative importance of potential improvements in this area.

After the PGI interview, the EQ-5D-3L was also administered to the study participants by the investigator, during the session. The EQ-5D-3L questionnaire includes five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression, each dimension at three levels: no problems, some problems and extreme problems. For each of these dimensions, the participants indicated their state of health corresponding to one of the 3 levels, indicated by a number. For all 5 dimensions, the resulting combined 5-digit number corresponded to the patients’ health status which was converted into an index score using validated weights. The resulting EQ-5D-3L index score ranged from 0 to 1; 0 being the worst health imaginable and 1 being full health.

Data analysis

Data was analyzed using R© (Version 3.2.5 of 2016).

The sociodemographic characteristics of the study participants were described using means, proportions and percentages.

For the PGI, the index score for each participant was generated by taking each area of ​​importance listed in Step 1, multiplying the scores assigned in Step 2 by the proportion of points assigned in Step 3, generating an index score between 0 and 6. An illustration of the scoring is provided in Table 1.

Table 1 Calculation of the PGI score for a study participant

The scores for each domain were then added together to generate a PGI score for each participant. The resulting score was therefore intended to represent the extent to which reality falls short of patients’ hopes and expectations in areas of life in which they would most appreciate improvement. [17].

Areas of importance listed by participants were also grouped and summarized into themes by consensus of the authors.

The EQ-5D-3L results were summarized for each domain and an index score was derived using the weights from the validated EQ-5D-3L scores from Zimbabwe [30] since no validated weight for the EQ-5D-3L was available from a Kenyan population.

The PGI summary scores were correlated to the EQ-5D-3L index score using the Pearson correlation.


Our study focused specifically on pregnant women living with HIV and aimed to gain insight into issues affecting pregnant women living with HIV. Informed consent to participate in the study was requested from each participant by the recruitment team. The recruitment team consisted of the women’s caregivers and research assistants. Entry and exit from the study were also clearly articulated to be voluntary for participants. All methods were performed in accordance with relevant ethical guidelines and regulations, and the study was approved by the Moi University Institutional Research and Ethics Committee (IREC 1292).

Trial registration: ISRCTN98818734; recorded on December 9, 2014.