The food environment in three South African public tertiary hospitals and legal doctrinal analysis of policy levers for a healthier hospital food environment

Abstract

Methods Study design, population and setting We conducted a cross–sectional mixed methods study to describe the food environment in three South African public tertiary hospitals (hospitals A, B & C) in Johannesburg, Gauteng province, South Africa, between 1st July and 30th September 2024. Gauteng province has a population of 16 million people and contains two of South Africa’s largest cities – the City of Johannesburg and the City of Tshwane.(11) The study population comprised hospital staff, visitors and ambulatory patients in the three hospitals. South African public tertiary hospitals provide specialist and subspecialist clinical services, training for healthcare service providers and have between 400–800 inpatient beds.(12) Hospital A had 6,760 staff, hospital B had 4,000 staff and hospital C had 1,886 staff. We did not evaluate food prepared by hospitals specifically for inpatients in this study. Data collection Data collection comprised of three components: an intercept survey, a food inventory audit, and semi–structured interview questionnaires. All activities were conducted by field workers trained in research ethics, utilising pilot–tested research tools. Data were captured electronically using tablets. Intercept survey We conducted intercept surveys to investigate food purchasing patterns and dietary choices among hospital staff, visitors and ambulatory patients across the three hospitals. Intercept surveys aimed to capture study participants’ actual behaviour and experiences. We recruited individuals aged 18 years old and above using convenience sampling from hospital locations with high traffic such as cafeterias, corridors and patient waiting areas. Food inventory audit A food inventory audit was conducted to assess type of food available at food outlets in each hospital by food outlet type (i.e., cafeterias, kiosks, shops, vending machines and leased retail premises that sell food and/or beverages. See S1 and S2 Files for a list of definitions). We documented all food and beverages being sold, these included both packaged food and food prepared on– or off–site. Interviews We purposively sampled participants recruited from the intercept survey to ensure diversity in age, race and sex. We conducted semi-structured interviews in–person on hospital premises and explored: reasons for participant’s choice of food and/or beverage, participant’s perception about food options, and their recommendations on food options. All interviews were audio–recorded and transcribed and translated into English, as appropriate. Legal and policy data collection We conducted a review of national laws, regulations, and policy documents relevant to food provision, procurement, and retail in hospital settings. The aim was to identify the legal and policy frameworks that shape hospital food environments and to understand where opportunities exist to strengthen them. Documents were collected from official government websites, gazettes, and legal repositories. Data analysis Descriptive statistics were used to describe the demographic characteristics of study participants and data from the food inventory audit. Frequencies and percentages were calculated for categorical variables that included sex (male, female), type of respondent (ambulatory patient, staff, visitor), age group (18–30, 31–49, ≥50 years), type of food outlets, and food and beverages available at food outlets. Data were presented both overall (all hospitals combined) and separately for each hospital. Quantitative analysis was conducted using STATA® statistical software version 19.5 SE.(13) Verbatim transcripts were uploaded to MAXQDA qualitative data analysis software.(14) We did a thematic analysis using a reflexive deductive and inductive approach following the steps presented by Braun & Clarke (2006). We used legal doctrinal analysis for national laws, regulations, and policy documents relevant to food provision, procurement and retail in hospital settings. The focus was on interpreting the scope and intent of existing provisions, identifying areas that enable or limit regulation of hospital food environments, and assessing alignment with international best practice. Ethical considerations This study was approved by the University of the Witwatersrand Human Research Ethics Committee (HREC REF M230748 MED23–07–023) and senior administrators (CEOs) of each hospital. All participants provided written informed consent, including for audio recording. Participants were provided with information about the study, their rights, and voluntarily opted to participate in the study. They were informed of their ability to withdraw from the survey without penalty and were informed that their data would be kept confidential. A reimbursement of 50 South African Rands (~ 2.8 US Dollars) was provided to study participants, after the interview, to compensate for time spent participating in the interview.