Apr 06, 2026

Risk factors for decline in estimated glomerular filtration rate amongst Malawian adults living in rural Karonga: protocol for a prospective cohort study using cystatin C- and creatinine-based eGFR

DOI
https://dx.doi.org/10.17504/protocols.io.rm7vz4rnxlx1/v1
  • Charlotte M Snead1,2,3,
  • Chimwemwe Mkandawire2,
  • Paul Kambiya2,
  • Shekinah Munthali-Mkandawire2,
  • Fredrick Kalobekamo2,
  • Albert Dube2,
  • Thandile Nkosi-Gondwe2,4,
  • Baltazar Bananga Mtenga2,5,
  • Dominic Nzundah2,
  • Lenford Kwamkwanya2,
  • Desire Bellings2,
  • Wisdom Nakanga2,6,
  • June Fabian7,
  • Robert Kalyesubula8,
  • Chimota Phiri9,
  • Felix Limbani1,3,
  • Dominic M Taylor10,11,
  • Amelia C Crampin2,4,5,
  • Henry C Mwandumba1,3,
  • Alison J Price2,4
  • 1Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, L3 5QA, United Kingdom;
  • 2Malawi Epidemiology and Intervention Research Unit, PO Box 46, Chilumba, Malawi;
  • 3Malawi Liverpool Wellcome Research Programme, Queen Elizabeth Central Hospital Campus, Chipatala Avenue, PO Box 30096 Chichiri, Blantyre 3, Malawi;
  • 4London School of Hygiene and Tropical Medicine, Keppel Street, London, WC1E 7HT, United Kingdom;
  • 5University of Glasgow, University Avenue, Glasgow, G12 8QQ, United Kingdom;
  • 6Deanery of Clinical Sciences, College of Medicine and Veterinary Sciences, University of Edinburgh, 49 Little France Crescent, The Chancellor’s Building, Edinburgh, EH16 4SB, United Kingdom;
  • 7Wits Donald Gordon Medical Research Institute, Faculty of Health Sciences, University of Witwatersrand, 7 York Rd, Parktown, Johannesburg, South Africa;
  • 8Makerere University, 7062 University Road, Kampala, Uganda;
  • 9Kamuzu University of Health Sciences, Private Bag 360, Chichiri, Blantyre 3, Malawi;
  • 10Population Health Sciences, Bristol Medical School, University of Bristol, Canynge Hall, 39 Whatley Road, Bristol, BS8 2PS, United Kingdom;
  • 11Richard Bright Renal Service, North Bristol NHS Trust, Southmead Road, Bristol, BS10 5NB, United Kingdom
charlotte.snead
Icon indicating open access to content
QR code linking to this content
DOI: https://dx.doi.org/10.17504/protocols.io.rm7vz4rnxlx1/v1
Protocol Citation: Charlotte M Snead, Chimwemwe Mkandawire, Paul Kambiya, Shekinah Munthali-Mkandawire, Fredrick Kalobekamo, Albert Dube, Thandile Nkosi-Gondwe, Baltazar Bananga Mtenga, Dominic Nzundah, Lenford Kwamkwanya, Desire Bellings, Wisdom Nakanga, June Fabian, Robert Kalyesubula, Chimota Phiri, Felix Limbani, Dominic M Taylor, Amelia C Crampin, Henry C Mwandumba, Alison J Price 2026. Risk factors for decline in estimated glomerular filtration rate amongst Malawian adults living in rural Karonga: protocol for a prospective cohort study using cystatin C- and creatinine-based eGFR. protocols.io https://dx.doi.org/10.17504/protocols.io.rm7vz4rnxlx1/v1
License: This is an open access  protocol  distributed under the terms of the  Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: April 06, 2026
Last Modified: April 06, 2026
Protocol  Integer ID: 314526
Keywords: glomerular filtration rate amongst malawian adult, ckd progression amongst malawian adult, risk factors for progressive kidney function decline, estimated glomerular filtration rate, ckd prevalence, kidney function decline, glomerular filtration rate, chronic kidney disease, progressive kidney function decline, ckd risk factor, gfr than creatinine, targeting ckd screening, risk factors for ckd development, ckd screening, use of serum creatinine, serum creatinine, estimated gfr, treatment of ckd, proteinuria, egfr decline, malawian adult, prospective cohort study, egfr background the global burden, ckd progression, health policies in malawi, african population, ckd development, demographic surveillance site in rural karonga, progression in africa, creatinine, serum sample, gfr, ckd, northern malawi, urine sample, general population health, malawi, effect size estimates of key risk factor
Funders Acknowledgements:
Wellcome Trust Clinical PhD Fellowship
Grant ID: 223502/Z/21/Z
Wellcome Trust
Grant ID: 098610/Z/12/Z
Wellcome Trust
Grant ID: 098610/B/12/A
Wellcome Trust
Grant ID: 217073/Z/9/Z
GSK Africa Non-Communicable Disease Open Lab
Grant ID: Project number:8111
Wellcome Trust
Grant ID: 220740/Z/20/Z
UKRI/MRC
Grant ID: MR/Z504853/1
Abstract
Background
The global burden of chronic kidney disease (CKD) is rising, disproportionately impacting on low- and middle-income countries. In African populations, use of serum creatinine to estimate glomerular filtration rate (GFR) significantly underestimates CKD prevalence, contributing to its under-recognition as a health problem. Serum cystatin C provides more accurate estimates of Iohexol measured GFR than creatinine. Early diagnosis and treatment of CKD is essential to reduce premature morbidity and mortality. However, little is known about risk factors for CKD development and progression in Africa owing to limited longitudinal data. This study aims to determine risk factors for progressive kidney function decline among adults living in rural, northern Malawi.
Methods
This protocol describes a prospective study being conducted in a general population Health and Demographic Surveillance Site in rural Karonga, Malawi (2024-2025). Eligible participants are adults aged 18 years and over who participated in two population-based surveys of long-term health conditions (2013-2016 and 2021-2025), with availability of baseline measures. New household-level data is being prospectively collected on CKD risk factors, alongside blood and urine samples. Cystatin C and creatinine will be tested on individual-level paired, stored serum samples collected at three longitudinal time points. Urine will undergo dipstick urinalysis, microscopy and testing for albumin and creatinine to quantify proteinuria. The primary outcome will be sustained 25% reduction in estimated GFR (eGFR) from baseline and change in eGFR category, determined using serum cystatin C. Multivariable logistic regression will be used to determine effect size estimates of key risk factors for kidney function decline.
Discussion
This study will provide important data on risk factors for eGFR decline and CKD progression amongst Malawian adults. The findings will inform future research into important context-specific risk factors, and could directly inform health policies in Malawi for targeting CKD screening, prevention and treatment strategies to high-risk patient groups.
Aims and Objectives
The aim of this prospective cohort study is to determine key risk factors for progressive decline in eGFR amongst adults participating in a general population cohort in rural northern Malawi. The primary objective is to determine predictors for progressive decline in eGFRcysC. A secondary objective is to determine risk factors for decline in eGFRcreat and development of CKD.
Study design and setting
The “Impso” (“kidney” in Chichewa, a local language) study is a prospective cohort study set within Karonga Health and Demographic Surveillance System (HDSS), an open, population-based cohort established in 2002 and covering an area of 135km2 of rural subsistence farming and fishing communities in northern Malawi(1) managed by the Malawi Epidemiology and Intervention Research Unit (MEIRU).
The HDSS includes a population of over 53,000 (data to early 2025) and is served by four health facilities (rural hospitals and health centres)(1). The nearest secondary referral hospital is located 80km to the north in Karonga boma. The HDSS undergoes continual population surveillance of births and deaths, with annual census of inward- and outward migration and collection of detailed sociodemographic data. All deaths undergo verbal autopsy (VA) with physician-assigned cause of death following a medical interview and case review using the standardised WHO VA tool.(1) Use of unique identifiers within the HDSS creates capacity for linkage across longitudinal studies. The Karonga HDSS, alongside a more recently established peri-urban HDSS in the outskirts of Lilongwe, is the site of two comprehensive population-based surveys of long-term health conditions (LTCs): a non-communicable disease (NCD) survey of cardiometabolic conditions conducted 2013-16(2, 3); and the currently ongoing Healthy Lives Malawi (HLM) LTC survey; 2021-25(4), which aims to re-survey prevalence and assess progression of cardiometabolic conditions, in addition to newly determining prevalence of additional LTC and more detailed risk factor data. Both studies contribute to a biobank.

The currently described study capitalises on these two larger studies and additionally leverages data from a large multinational study conducted in 2018-19 by the African Research in Kidney Disease (ARK) network to investigate methods for measuring kidney function in African populations(5), which included 510 adults residing in the Karonga HDSS. Available data from all three linked studies includes household data (socioeconomic status; geolocators), interview data (demographics; lifestyle factors; clinical history, including prior diagnosis, screening and medications for chronic conditions), examination and physical measures (anthropometry; blood pressure; hand grip strength; peripheral arterial measures) and biological samples (serum, plasma and whole blood samples stored at -80 Celsius and other biological material). Baseline cystatin C and creatinine have already been tested on stored serum samples collected between 2013 and 2019(5, 6) for 2792 and 4637 adults respectively.
Sample size calculation
The target sample size is 1000 participants. A sample of n=1000 participants with eGFRcysc available at baseline will give ≥80% power to detect odds ratio (OR) of between 1.66 and 4.38 for the effect of risk factors on eGFR decline. This is assuming that; (1) the risk factors of interest increase probability and rate of eGFR decline; (2) risk factor prevalences range between 5% and 30% (2, 6); and (3) eGFR decline outcome measures occur in 5 to 15% of individuals unexposed to the risk factors of interest.
Participants
This study is prospectively recruiting adults living within the Karonga HDSS, who meet the following eligibility criteria:
  • Alive, self-defined as resident within the Karonga HDSS[1], and provided consent to ongoing participation in the HDSS and sampling for further studies.
  • Participated in both the NCD (2013-16) and HLM-LTC (2021-25) population surveys.
  • Aged ≥ 18 years at the time of participation in the NCD (2013-16) survey.
  • Availability of baseline cystatin C tested on a stored historical serum sample (2013-2019).
  • Baseline eGFRcysC <90 ml/min/1.73m2.
Exclusion criteria include:
  • Unable to provide informed written consent, or assent with proxy informed written consent from a nominated guardian.
  • Acute presentation of physical or mental illness or rapidly declining poor health at the time of attempted recruitment, which might be exacerbated by participation in the study.
  • Currently a hospital inpatient at time of follow-up (at time of screening for participation in Impso study)
  • Pregnant at the time of baseline sample collection for cystatin C testing, and/or when approached for recruitment into the current study, with pregnancy to continue beyond the end of the study recruitment period.
[1] In the event of internal migration within the HDSS, resident status is maintained.
Recruitment and consent
Individuals potentially meeting the inclusion criteria are identified by screening the MEIRU database. For this purpose, the MEIRU-authorised study team members were granted access to linked datasets (secured in encrypted servers, only accessible by authorised personnel through provided credentials) on 08/08/2023 for the purposes of screening, study visits and results dissemination. All participants in the datasets have given consent for use of their data in future linked studies. Final datasets are stripped of personal identifying data. No study team members have access to identifying after data collection and results dissemination are complete. All study team members undergo training in good clinical practice in research.
Potentially eligible participants are approached at their household to invite participation in the study, a minimum of three months after their participation in the HLM LTC survey. Up to three attempts are made to visit each potentially eligible participant, including at different times of day, to facilitate recruitment of adults of working age who may be less likely to be at home and thereby minimise selection bias. After checking for exclusion criteria, individuals who accept receive information about the study in the form of a participant information sheet, provided in their preferred language (Chitumbuka, Chichewa or English). The information sheet is read aloud to the potential participant by a member of the study team and the individual is given the opportunity to ask questions. Individuals who wish to take part proceed to provide informed, written consent. Individuals who are unable to read or write provide a thumbprint instead of a signature. In this event, an impartial witness is present throughout the information-giving and consent process, and also signs the informed consent form. For individuals who lack capacity to provide informed consent, but are able to provide assent, proxy consent is sought from the nominated guardian. Informed written consent is obtained for all planned study activities, including for long-term storage of study data and biological samples.
Household-level data collection
Data collection takes place over the course of two to three household visits.
Interviewer-led questionnaires
The first household visit involves delivery of several questionnaires, covering; (1) context-specific risk factors for CKD, including detailed questions on medical history, symptoms and medication use, guided by a priori hypotheses on exposures of importance in this setting(6, 7), and supported by review of the participant’s health passport if available; (2) health-related quality of life (HrQOL), using the RAND 36-Item Short Form Survey Instrument (SF-36), and the EQ-5D-5L, both translated into Chitumbuka and Chichewa, with the SF-36 adapted where appropriate for relevance to the Malawi setting(8), and use of the EQ-5D-5L translations (already validated in Chichewa)(9) approved by EuroQol; and (3) screening questions to determine optimal timing for early morning biological sample collection. Sample collection is delayed in the event of urinary tract infection symptoms, current or recent (within past 24 hours) menstruation, and hospital admission >24 hours duration within the preceding 90 days.

HIV testing and counselling (HTC) using point-of-care DetermineTM rapid test kits is offered to participants who have never been tested, are unsure about their HIV status, or whose most recent test with a non-reactive result was conducted over six months previously (ascertained from both self-report and health passport documentation). Study staff performing HTC have undergone formal training and certification though Malawi Ministry of Health HTC training providers. Any participants with a reactive result on the DetermineTM rapid HIV test are referred to the HIV clinic at Chilumba Rural Hospital for repeat counselling and full testing using the three-test diagnostic algorithm, following Ministry of Health guidelines.(10)

All questionnaire data is captured electronically on Android tablets using the Open Data Kit (ODK) platform(11, 12). Before completing the first household visit, the study staff ensure that clear instructions have been given to participants in preparation for mid-stream urine (MSU) sample collection using a standardised information sheet provided in Chitumbuka, Chichewa or English.

Biological sample collection
Collection of biological samples (MSU and single venepuncture serum sample) takes place at a second, early morning household visit, the day after the interview date, or else on the first alternative convenient date for the participant. Immediately prior to sample collection, study staff repeat screening questions for menstruation and UTI symptoms, and check that the MSU collection instructions have been followed. MSU samples are collected in sterile containers, contained with sealed sterile packaging until the time of collection, and are inspected by the study team prior to dipstick urinalysis testing. Samples collected incorrectly, that are clearly contaminated, leaking, or that are inadequate (< 20 ml) in volume are discarded without testing and arrangements are made for re-collection.
Dipstick urinalysis is being conducted on acceptable MSU samples at the household using Siemens Multistix‱ 10 SG urine dipsticks following a strict standard operating procedure, including use of a sterile Pasteur pipette to apply urine to dipstick test pads, and use of an electronic timer to ensure dipstick results for each reagent are read at the times specified by manufacturer. Urine dipstick results are captured electronically at the household using ODK and are photographed at two minutes for quality control purposes. Any abnormalities on dipstick urinalysis triggers a third household visit in ≥ three weeks’ time for collection and testing of a repeat (second) MSU sample. In the event of microscopic haematuria newly present on a second sample, a third and final MSU collection visit takes place. Venepuncture serum samples will be collected in 10ml plain serum tubes using vacutainer needles. Samples are placed upright in cool boxes maintained at 2 – 8 ˚C immediately after collection and urinalysis, and are transported back to the MEIRU laboratory within four hours of sample collection. Within this four-hour window, urine samples spend a maximum of two hours at ambient temperature prior to the cold storage.
Laboratory methods
The condition of all biological specimens is inspected on arrival at the laboratory; samples that are leaking, contaminated or inadequate are rejected and arrangements are made for recollection. Urine microscopy involves both wet mount examination and Gram stain of urine sediment. After mixing by inversion, 10 ml from each mid-stream urine sample is centrifuged at 720 g for five minutes immediately after reaching the laboratory.  After separating from the supernatant, urine sediment is emulsified and the sediment examined by light microscopy at both low power field (LPF, x10 objective) and high power field (HPF, x40 objective), the latter for identification and quantification of cells (erythrocytes, leucocytes, epithelial cells), casts, crystals and Schistosoma ova. Cell counts are reported per HPF, and Schistosoma ova per 10 ml of urine. Prior to Gram stain, urine sediment is mounted on the slide and gently heat fixed for five minutes.  Gram stain is then performed by sequentially flooding slides, first with 0.5% crystal violet (left for 60 seconds), rinsed off with water, then with Gram’s iodine (left for 60 seconds), washed off with acetone decolouriser before rinsing with water, and finally with safranin (left for 30 seconds), again rinsed off with water.  After blotting dry, slides are examined using x100 objective lens with emersion oil. Participants with organisms identified on Gram stain of the baseline urine sample receive a follow-up visit for repeat sample collection, with reiteration of MSU collection instructions. Urine culture is not being performed owing to budget constraints of the study. The remaining uncentrifuged portion of the urine sample for each participant is mixed by inversion and used to extract three 1ml aliquots using 2ml cryotubes, which are stored at -20°C temporarily, then transferred to the -80°C  freezer within 12 hours of sample collection for short-term storage until the time of urine chemistry analysis, which will take place after participant recruitment is complete.

Blood samples collected in plain tubes are centrifuged within two hours of reaching the laboratory by spinning for ten minutes at 1620g in a swing bucket centrifuge. The separated serum is used to make five aliquots of 0.5ml using 2ml cryotubes, stored at -20°C temporarily, then transferred to -80°C freezer within 12 hours of sample collection for long-term storage. Prior to chemistry testing, stored serum and urine samples will be removed from the -80°C freezer and allowed to thaw for 30 minutes to 1 hour. After completing thawing, samples will be vortexed and put in sample cups ready for testing. All chemistry testing will use Beckman Coulter AU480 chemistry analysers.
Serum samples from this study, and individual level paired serum samples from the HLM-LTC survey, will be tested for serum cystatin C (immunoturbidimetric method, standardised using ERM- DA471/IFCC reference material) and creatinine (kinetic compensated Jaffe method, method A, standardised to an isotope-dilution mass spectroscopy [IDMS] assay). For cystatin C testing, serum will be mixed with Gentian Cystatin C reagent containing cystatin C immunoparticles. Cystatin C from the serum and anti cystatin C from the immunoparticles form aggregates. The complex particles created absorb light, and by turbidimetry the absorption is related to cystatin C concentration via interpolation on an established standard calibration curve. The AU480 platforms automatically calculate the results. The machine will be calibrated using Gentian Cystatin C Calibrator which is standardised against the international calibrator standard ERM-DA471/IFCC. A six-point standard curve will be established using standards one to six with assigned values given on the analytical value sheet provided with the calibrator. In the kinetic compensated Jaffe method, creatinine will form a coloured compound with picric acid in alkaline medium. The rate of change in absorbance at 520/800nm is proportional to the creatinine concentration in the sample. Interference from protein will be mathematically corrected by subtracting 18 μmol/L from each test result. The Beckman Coulter AU480 analyser will automatically compute the creatinine concentration of each sample. Serum calibrator creatinine value for method A is traceable to the Isotope Dilution Mass Spectroscopy (IDMS) method via National Institute of Standards and Technology (NIST) Standard Reference Material (SRM) 967.
Urine will be tested for albumin (turbidimetric method) and creatinine (kinetic compensated Jaffe method). In the turbidimetric method for urine albumin testing, anti-human serum albumin antibodies combine with albumin from the sample to form immune complexes that scatter light in proportion to their size, shape and concentration. The absorbance of these aggregates is proportional to the albumin concentration in the sample. Change in absorbance is measured at 380nm with subtraction of a reference wavelength at 800nm. The urine calibrator creatinine value is traceable to the IDMS method. Urine albumin calibrator values are traceable to the International Federation of Clinical Chemistry Certified Reference Material CRM470.
Urine samples will be discarded after validation of results; but serum samples will be stored long-term to allow additional testing in future studies.
Laboratory quality control processes
Prior to microscopic examination of urine (for both Gram staining and wet mount examination of urine sediment) the performance of the objective lenses is verified using standardised positive and negative control slides. Quality control of the microscopy procedures is maintained by independent verification of findings by two qualified technologists. During serum cystatin C and creatinine testing, low and high concentration controls will be tested each day before any samples are measured to validate the calibration curve. The controls have an assigned value range that will be met before measuring samples; these are given in the Analytical Value sheet included with the Gentian Cystatin C and creatinine control kits. During urine albumin and creatinine testing, low and high concentration controls of Biorad Liquichek assayed quality control material will be tested a minimum of once a day on each day of testing.
Pilot phase
The data collection tools and field activities strategy were piloted both internally and externally to understand and address in a timely manner any issues associated with the workflow, logistics and time required, for the information giving, consent, medical interview, and biological sample collection processes. During this pilot phase, the questionnaire and field procedures were reviewed and revised based on feedback raised. This included improvements to some questionnaire questions, answer options and translations to improve understandability and cultural acceptability, in addition to revision of working procedures for more efficient collection of interview data and biological samples.
Timelines
Recruitment for the pilot phase took place between 14 February and 1 March 2024. Recruitment for the main study commenced on 5 March 2024, anticipated to complete by 30 September 2025. Data collection is anticipated to complete by 30 November 2025, and results are anticipated by March 2026.
Analysis Plan
eGFR equations
By the end of the study, each participant will have up to three stored serum samples tested for cystatin C and creatinine, allowing calculation of eGFRcysC and eGFRcreat at different points in time: (i) baseline (blood collected between 2013 and 2019); (ii) HLM LTC survey (blood collected between 2021 and 2025); and (iii) Impso study (blood collected between 2024 and 2025). HLM LTC and Impso study samples are collected a minimum of 90 days apart. All eGFRcysC values will be calculated using the CKD-EPI 2012 equation.(13) eGFRcreat will be calculated using the CKD-EPI (2009) equation(14) without adjustment for African American ethnicity, as this has been demonstrated to have slightly higher accuracy and lower relative bias than the newer CKD-EPI (2021) equation when compared to Iohexol mGFR in African populations, including in adults from rural Karonga.(5) However sensitivity analyses for will also be conducted using the CKD-EPI (2021) creatinine equation (which has no ethnicity coefficient), the CKD-EPI (2021) combined creatinine-cystatin C equation and the European Kidney Function Consortium (EFKC) equations for both creatinine and cystatin C.(15-18)
Outcome measures
The study outcome measures are guided by the Kidney Disease Improving Global Outcomes (KDIGO) as well as the National Kidney Foundation and Food and Drug Administration guidance on surrogate end points for CKD progression.(19, 20)  In each of the below definitions, for the purposes of this study, the term “sustained” means that at an individual participant level, the outcome is present on testing both the sample collected in the 2021-25 HLM LTC survey, and on the sample collected in the currently described 2024-25 Impso study, when compared to testing of the individual-level paired baseline sample (2013-19). “Non-sustained” means that the outcome has been reached for the first time at the point of the most recent (Impso study) sample collection.

The primary outcome measure will be a sustained 25% reduction in eGFRcysC from baseline and change in eGFRcysC category.
The secondary outcome measures will be:
  1. Non-sustained 25% reduction in eGFRcysC from baseline and change in eGFRcysC category
  2. (a) Sustained and (b) non-sustained 25% reduction in eGFRcreat from baseline and change in eGFRcreat category
  3. (a) Sustained and (b) non-sustained decline in (a) eGFRcysC and (b) eGFRcreat by:
-- 57%(equates to doubling of serum creatinine)(20)
-- 40%
-- 30%
4. Estimated annual decline* in (a) eGFRcysC and (b) eGFRcreat of:
-- -5ml/min/1.73m2/year
-- -3ml/min/1.73m2/year
*(Data from other settings suggests that age-related GFR decline in healthy adults without hypertension is no more than −1.07 mL/min/1.73m2/year)(21-23)
5. Incident CKD, defined using eGFR criteria as; (a) Sustained and (b) non-sustained new eGFR <60ml/min/1.73m2 (not present at baseline) using both eGFRcysC and eGFRcreat
6. Kidney failure, defined as:
-- (a) Sustained and (b) non-sustained eGFRcysC <15ml/min/1.73m2; or
-- (a) Sustained and (b) non-sustained eGFRcreat <15ml/min/1.73m2; or
-- Commencement of maintenance KRT (≥4 weeks duration)[2]
Statistical analysis
Logistic regression (univariate and multivariate models) will be used to determine effect size estimates (odds ratios) of key covariates (risk factors) for kidney function decline (primary and secondary outcomes) to interrogate how much of kidney disease progression is explained by potential novel risk factors of interest, such as self-reported history of severe malaria or sepsis, HIV infection, previous TB or schistosomiasis infection, use of non-steroidal anti-inflammatory medications, use of traditional medicines, use of antibiotics, pregnancy-related complications and occupational exposures. Adjustment will be made for traditional risk factors (hypertension, diabetes, smoking and cardiovascular disease), established confounders (age, sex, body mass index, socioeconomic status) and for baseline eGFR. Potential interaction between sex and the effect of risk factors on outcome measures will be examined, as there is evidence from other settings that sex may modulate the relationship with renal function decline, in particular cardiometabolic risk factors.(24) Heterogeneity by age in the associations with outcome measures will also be investigated. In addition linear mixed effects models will be used to determine eGFR trajectories over time and estimate differences in eGFR slopes according to baseline risk factors.(25)
[2] Commencement of maintenance KRT forms part of the KDIGO definition of kidney failure, however we do not expect to find any individuals meeting this outcome criterion in our cohort.
Ethics, Governance And Regulations
Ethical approval and informed consent
This study was approved by the National Health Sciences Research Committee (NHSRC) in Malawi (protocol #23/04/4049) and by the Liverpool School of Tropical Medicine, UK (protocol #23-019). All participants provided written, informed consent to participate.
Feedback of individual participant test results and clinical referral
Individual urine dipstick results will be communicated to study participants immediately after testing and will be recorded in health passports. Participants with abnormal baseline urine results requiring follow-up urine collection (see ‘Biological Sample Collection’), will receive results of the laboratory urine microscopy at the follow-up visit.

Individual baseline eGFR results are communicated to participants at the point of recruitment, accompanied by provision of general CKD lifestyle advice using a standardised information sheet. Repeat kidney function results (eGFR, uACR) obtained by testing of serum and urine samples collected in this study, and on paired LTC samples, as well as remaining normal urine test results, will be communicated once recruitment and testing is complete. Clinical assessment will take place at an NCD clinic at Chilumba Rural Hospital, which was established by MEIRU to support the wider HLM LTC survey. Any onward referrals required to secondary or tertiary facilities will take place via existing referral pathways.
Study governance and quality control
The study field staff all underwent Good Clinical Practice training in addition to formal training in all study activities as outlined in the Standard Operating Procedures prior to study initiation. This included formal training in study information, informed consent, survey administration, electronic data capture, biological sample collection and transportation. Intermittent spot checks will be carried out of field data collection activities, and regular team meetings will be held to facilitate discussion of study progress, challenges in the field, and for performance-based feedback based on review of incoming study data. Photographs of a random 10% sample of urine dipsticks will be reviewed by the investigator team. In the laboratory, calibrators and controls will be used to ensure reliability and consistency of assay results; in addition to use of standardized operating procedures for sample processing. Spot checks will be carried out to ensure the temperature of the cool boxes used for sample transportation are maintained within the required range.
Patient and public involvement
Prior to initiation of this study, approval for the study was granted by the Karonga District Health Office, and meetings were held with community leaders through the Area Development Committee. HDSS residents had been engaged at an individual- and community-levels on all aspects of the wider MEIRU programme of census activities, HLM LTC survey and its nested studies, including this study, through a series of open community meetings to raise awareness and provide opportunities for questions and discussion. Additional meetings with group village heads and community representatives of each area of the HDSS due to be specifically visited by this study took place one to two weeks before initiation of the study in that particular area, to discuss finer details of this study, enabling wider dissemination of information by community representatives and providing a more focused opportunity for questions and concerns to be raised and addressed.

On completion of the study, community meetings will be held to feedback the results of the study and wider HLM-LTC survey, including with the more recently formed community advisory group. This will include opportunity for questions, discussion, and priorities to be raised with regards to future research directions.
Protocol references
1.         Crampin AC, Dube A, Mboma S, Price A, Chihana M, Jahn A, et al. Profile: the Karonga Health and Demographic Surveillance System. Int J Epidemiol. 2012;41(3):676-85.
2.         Crampin AC, Kayuni N, Amberbir A, Musicha C, Koole O, Tafatatha T, et al. Hypertension and diabetes in Africa: design and implementation of a large population-based study of burden and risk factors in rural and urban Malawi. Emerg Themes Epidemiol. 2016;13:3.
3.         Price AJ, Crampin AC, Amberbir A, Kayuni-Chihana N, Musicha C, Tafatatha T, et al. Prevalence of obesity, hypertension, and diabetes, and cascade of care in sub-Saharan Africa: a cross-sectional, population-based study in rural and urban Malawi. Lancet Diabetes Endocrinol. 2018;6(3):208-22.
4.         Malawi Epidemiology and Intervention Research Unit. Healthy Lives Malawi  [Available from: https://www.meiru.info/healthy-lives-malawi/.
5.         Fabian J, Kalyesubula R, Mkandawire J, Hansen CH, Nitsch D, Musenge E, et al. Measurement of kidney function in Malawi, South Africa, and Uganda: a multicentre cohort study. Lancet Glob Health. 2022;10(8):e1159-e69.
6.         Nakanga WP, Prynn JE, Banda L, Kalyesubula R, Tomlinson LA, Nyirenda M, et al. Prevalence of impaired renal function among rural and urban populations: findings of a cross-sectional study in Malawi. Wellcome open research. 2019;4(92):92.
7.         Stanifer JW, Muiru A, Jafar TH, Patel UD. Chronic kidney disease in low- and middle-income countries. Nephrol Dial Transplant. 2016;31(6):868-74.
8.         Masina T, Chimera B, Kamponda M, Dreyer G. Health related quality of life in patients with end stage kidney disease treated with haemodialysis in Malawi: a cross sectional study. BMC Nephrol. 2016;17(1):61.
9.         Chokotho L, Mkandawire N, Conway D, Wu HH, Shearer DD, Hallan G, et al. Validation and reliability of the Chichewa translation of the EQ-5D quality of life questionnaire in adults with orthopaedic injuries in Malawi. Malawi Med J. 2017;29(2):84-8.
10.       Malawi Ministry of Health. HIV, Syphilis and Hepatitis B Integrated Rapid Testing and Counselling Guidelines and Standard Operating Procedures (1st edition). 2023.
11.       Hartung C, Lerer A, Anokwa Y, Tseng C, Brunette W, Borriello G, editors. Open data kit: tools to build information services for developing regions. Proceedings of the 4th ACM/IEEE international conference on information and communication technologies and development; 2010.
12.       Open Data Kit software.
13.       Inker LA, Schmid CH, Tighiouart H, Eckfeldt JH, Feldman HI, Greene T, et al. Estimating glomerular filtration rate from serum creatinine and cystatin C. N Engl J Med. 2012;367(1):20-9.
14.       Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604-12.
15.       Inker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, et al. New Creatinine- and Cystatin C-Based Equations to Estimate GFR without Race. N Engl J Med. 2021;385(19):1737-49.
16.       Pottel H, Hoste L, Dubourg L, Ebert N, Schaeffner E, Eriksen BO, et al. An estimated glomerular filtration rate equation for the full age spectrum. Nephrology Dialysis Transplantation. 2016;31(5):798-806.
17.       Pottel H, Bjork J, Courbebaisse M, Couzi L, Ebert N, Eriksen BO, et al. Development and Validation of a Modified Full Age Spectrum Creatinine-Based Equation to Estimate Glomerular Filtration Rate: A Cross-sectional Analysis of Pooled Data. Ann Intern Med. 2021;174(2):183-91.
18.       Pottel H, Bjork J, Rule AD, Ebert N, Eriksen BO, Dubourg L, et al. Cystatin C-Based Equation to Estimate GFR without the Inclusion of Race and Sex. N Engl J Med. 2023;388(4):333-43.
19.       Levin A, Agarwal R, Herrington WG, Heerspink HL, Mann JFE, Shahinfar S, et al. International consensus definitions of clinical trial outcomes for kidney failure: 2020. Kidney Int. 2020;98(4):849-59.
20.       Badve SV, Palmer SC, Hawley CM, Pascoe EM, Strippoli GF, Johnson DW. Glomerular filtration rate decline as a surrogate end point in kidney disease progression trials. Nephrol Dial Transplant. 2016;31(9):1425-36.
21.       Guppy M, Thomas ET, Glasziou P, Clark J, Jones M, O'Hara DV, et al. Rate of decline in kidney function with age: a systematic review. BMJ Open. 2024;14(11):e089783.
22.       Correction: Sex Differences in Age-Related Loss of Kidney Function. J Am Soc Nephrol. 2024;35(8):1137.
23.       Melsom T, Norvik JV, Enoksen IT, Stefansson V, Mathisen UD, Fuskevag OM, et al. Sex Differences in Age-Related Loss of Kidney Function. J Am Soc Nephrol. 2022;33(10):1891-902.
24.       Sullivan MK, Lees JS, Rosales BM, Cutting R, Wyld ML, Woodward M, et al. Sex and the Relationship Between Cardiometabolic Risk Factors and Estimated GFR Decline: A Population-Based Cohort Study. Am J Kidney Dis. 2024;84(6):731-41 e1.
25.       Shou H, Hsu JY, Xie D, Yang W, Roy J, Anderson AH, et al. Analytic considerations for repeated measures of eGFR in cohort studies of CKD. Clinical Journal of the American Society of Nephrology. 2017;12(8):1357-65.
Acknowledgements
Acknowledgements
CS is also affiliated with Malawi Epidemiology and Intervention Research Unit (MEIRU) and with the Malawi Liverpool Wellcome Research Programme, Blantyre, Malawi. We thank participants of all the named studies in addition to the wider MEIRU research, management and support teams for their contribution. We thank Professor Segun Fatumo (Queen Mary, University of London) for his support in provision of serum creatinine and cystatin C assays. We acknowledge the wider ARK consortium for their work pre-dating this study and ongoing collaboration. CS also thanks the Liverpool Clinical PhD Programme for Health Priorities in the Global South (Professor Neil French, Elly Wallis, and steering group) for their support and the Liverpool School of Tropical Medicine for sponsorship of this study.

Author contributions
The current study was conceived and designed by CMS, AJP, ACC, HCM, DMT, FL, JF and RK. The study builds directly upon previous linked research studies which were conceived and designed by AJP, ACC, WN, JF and RK. CMS, AJP, FL, and ACC applied for and acquired funding for the study. Surveys were designed by CMS, AJP and DMT; with programming of data collection tools implemented and managed by PK, BBM and DN with support from the wider MEIRU data team. Study standard operating procedures were written by CMS and CM with input and revisions from SMM, PK, AJP, AD, CP and DMT. Community sensitisation work was planned by CMS, AD, FK and conducted by CMS, AD, FK, LK and DB. Staff training was conducted by CMS, AD, TNG, SMM and CM. Pilot data was collected by LK and DB, with supervision by CMS, AD and TNG. Data management systems for the pilot data were managed by PK and BBM. Laboratory procedures were supervised by CM. The manuscript was drafted by CMS. All authors contributed to critical review and revisions of the manuscript prior to publication apart from LK†. This work was conducted as part of CMS’ PhD fellowship; CMS is supervised by AJP, FL, DMT and HCM with collaboration from ACC, JF and RK.

†Deceased, 22/07/2024.