Factors associated with adherence to dietary prescription among adult patients with chronic kidney disease on hemodialysis in national referral hospitals in Kenya: a mixed-methods survey

Abstract

Introduction: Adherence to dietary prescriptions among patients with chronic kidney disease is known to prevent deterioration of kidney functions and slow down the risk for morbidity and mortality. This study determined factors associated with adherence to dietary prescription among adult patients with chronic kidney disease on hemodialysis.

Methods: A mixed-methods study, using parallel mixed design, was conducted at the renal clinics and dialysis units at the national teaching and referral hospitals in Kenya from September 2018 to January 2019. The study followed a QUAN + qual paradigm, with quantitative survey as the primary method. Adult patients with chronic kidney disease on hemodialysis without kidney transplant were purposively sampled for the quantitative survey. A sub-sample of adult patients and their caregivers were purposively sampled for the qualitative survey. Numeric data were collected using a structured, self-reported questionnaire using Open Data Kit "Collect software" while qualitative data were collected using in-depth interview guides and voice recording. Analysis on STATA software for quantitative and NVIV0 12 for qualitative data was conducted. The dependent variable, "adherence to diet prescription" was analyzed as a binary variable. P values < 0.1 and < 0.05 were considered as statistically significant in univariate and multivariate logistic regression models respectively. Qualitative data were thematically analyzed.

Results: Only 36.3% of the study population adhered to their dietary prescriptions. Factors that were independently associated with adherence to diet prescriptions were "flexibility in the diets" (AOR 2.65, 95% CI 1.11-6.30, P 0.028), "difficulties in following diet recommendations" (AOR 0.24, 95% CI 0.13-0.46, P < 001), and "adherence to limiting fluid intake" (AOR 9.74, 95% CI 4.90-19.38, P < 0.001).

Conclusions: For patients with chronic kidney disease on hemodialysis, diet prescriptions with less restrictions and requiring minimal extra efforts and resources are more likely to be adhered to than the restrictive ones. Patients who adhere to their fluid intake restrictions easily follow their diet prescriptions. Prescribed diets should be based on the individual patient's usual dietary habits and assessed levels of challenges in using such diets. Additionally, diet adherence messages should be integrated with fluid limitation messages. Further research on understanding patients' adherence to fluid restriction is also suggested.

Keywords: Adherence; Diet; Food; Hemodialysis; Kenya; Mixed-methods study; Nutrition; Renal.

Fig. 1

Flexibility of prescribed diets. This was assessed as the “flexibility of the prescribed diet to fit in with usual dietary habits” of the participants. It was assessed on a 9-point Likert scale [31, 33] of 1 = it fits in with my usual way of eating; 2 = it seems to contradict what I thought was healthy; 3 = it is difficult to combine with the rest of the family; 4 = it makes it difficult to eat out; 5 = it combines easily with other dietary advice I have been given; 6 = it is more expensive than my usual way of eating; 7 = I seem to have to eat more than I want; 8 = there are lots of foods I can no longer eat; 9 = I do not need to make any changes. During analysis, the responses for “1, 5, and 9” were combined to represent “flexible and fits with usual way of eating and previous dietary advice received”; “3 and 4” were combined to represent “not flexible, cannot combine with family meals, and cannot eat out.” Responses “6 and 7” were also combined to represent “more expensive and I seem to eat more”

Fig. 2

Adherence to diet prescription and fluid restriction. Adherence was self-reported on a 5-point Likert scale where 1 = adherence to dietary prescriptions, with the recommendations followed all the time; 2 = mild non-adherence, with the recommendations followed most of the time; 3 = moderate non-adherence, with the recommendations followed about half of the time; 4 = severe non-adherence with the recommendations followed very seldom; 5 = very severe non-adherence, with the recommendations not followed at any time in the past 7 days [7, 24]. A binary variable was computed from this 5-point Likert scale during data analysis as “1 = 1” coded as “adherence” for “recommendations followed all the time” and “2–5 = 0” coded as “non-adherence” for “recommendations not followed all the time.”