South Africa established chronic disease management programmes (CDMPs) called ‘clubs’ to ensure quality diabetes care. However, the effectiveness of these clubs remains unclear in terms of disease risk factor monitoring and complication prevention.
We assessed risk factor monitoring, prevalence and determinants of diabetes related complications amongst type-2 diabetes (T2D) and hypertension (HTN) patients attending two CDMPs.
Urban Township in Cape Town, South Africa.
Cross-sectional survey combined with a 10-year retrospective medical records analysis of adult T2D/HTN patients attending two CDMPs, using a structured survey questionnaire and an audit tool. Statistical Software for Social Sciences (SPSS) version 25 was used to analyse risk factor monitoring and calculate prevalence of complications. Potential determinants of complications were explored through logistic regression.
There were 379 patients in the survey, 372 (97.9%) had HTN whilst 159 (41.9%) had T2D and HTN; 361 medical records were reviewed. Blood pressure (87.7%) and weight (86.6%) were the best monitored risk factors. Foot care (0.0% – 3.9%) and eye screening (0.0% – 1.1%) were least monitored. Nearly 22.0% of patients reported one complication, whilst 9.2% reported ≥ 3 complications. Medically recorded complications ranged from 11.1% (1 complication) to 4.2% with ≥ 3 complications. The most common self-reported and medically recorded complications were eye problems (33%) and peripheral neuropathy (16.4%), respectively. Complication occurrence was positively associated with age and female gender and negatively associated with perceived illness control.
Type-2 diabetes and hypertension patients experienced diabetes related complications and inadequate risk factor monitoring despite attending CDMPs. Increased self-management support is recommended to reduce complication occurrence.
In the 1990–2019 global burden of disease study, high fasting plasma glucose and high body mass index (BMI) were recorded as the most increased risk exposures to health, whilst hypertension (HTN) was the risk factor that showed the largest increase accounting for 10.8 million deaths.
South Africa has implemented several policies to guide self-management programmes for quality diabetes care. One such policy was the establishment of chronic disease management programmes (CDMPs) called ‘clubs’, aimed at equipping patients with the necessary knowledge and self-management skills in order to reduce disease progression and complications.
This was a cross-sectional survey combined with a 10-year retrospective patient records analysis (2009–2018), conducted in March 2018.
The study took place in two community health centres in Khayelitsha, a predominantly black African (90.5%) township in Cape Town with a recent official population for 2019/2020 estimated at 442 721 by the Western Cape government.
The study population consisted of 379 patients with T2D and/or HTN (T2D/HTN) selected from the daily clinic attendance register in both facilities using systematic random sampling. Eligibility criteria included the following: aged between 18 and 74 years; diagnosed with T2D/HTN for at least 6 months; living in Khayelitsha and attending the CDMP at the selected facility for at least 6 months; not experiencing acute illness; and able to give informed consent. Eligible patients were asked to participate whilst waiting for medical consultation. After receiving study information, consenting patients were enrolled.
The prevalence of diabetic retinopathy was used for sample size calculation because it was well documented by a previous study. A minimum sample size of 380 was calculated using Charan and Biswas prevalence estimation formula.
A structured survey questionnaire, based on the World Health Orginazation - STEPwise Approach to NCD Risk Factor Surveillance (STEPS) instrument
Self-management variables of diabetes knowledge and illness perception were also measured. Knowledge was measured using the Diabetes Knowledge Scale consisting of 13 items from the validated Diabetes Knowledge Questionnaire (DKQ-24), selected based on their contextual relevance.
To assess risk factor monitoring for disease control and clinician-recorded complications, a retrospective medical audit of patient files was conducted using an audit tool designed by combining variables from relevant international, national and local guidelines,
Data captured in the audit tools and survey questionnaires, was transferred into Redcap software hosted at the University of the Western Cape (UWC). Two field workers entered the same participant data independently and then compared their entries, enabling data quality checks during the data collection phase. Field worker supervisor and the research team undertook additional data checks (missing data, data consistency, and deviant entries) before exporting data into a Microsoft Excel spreadsheet. After cleaning, data was analysed using Statistical Package for Social Sciences (SPSS) software version 25.
The proportion of eligible patients who received risk factor monitoring annually was calculated to assess disease monitoring. Descriptive statistics were produced and the proportion of patients with specific complications was calculated to estimate prevalence rates. Multivariable logistic regression using R software was performed to explore associations between the occurrence of complications (binary dependent variable) and socio-demographic characteristics and self-management variables (independent variables). In a first model, the socio-demographic characteristics included were age, gender, education level, marital status (married or cohabiting), employment status, and income level. In a second model, self-management variables were added, including diabetes knowledge and perceived illness control, whilst socio-demographic variables and reported intake of oral anti-diabetics were included as control variables. The occurrence of complications corresponded to ‘having any self-reported complication’ or ‘having any clinician recorded complication’.
Approval was granted by the University Biomedical Research Ethics Committee and Western Cape Provincial health authorities (reference: WC_2017RP50_730). Health facilities participating in the study also granted permission to access the facilities, patients and their medical records.
Most patients (
Background characteristics of the study population (
Characteristic | % | |
---|---|---|
HTN | 372 | 98.2 |
T2D/HTN | 159 | 42.0 |
Male | 71 | 18.7 |
Female | 308 | 81.3 |
20–49 | 100 | 26.4 |
50–64 | 208 | 54.9 |
65+ | 71 | 18.7 |
Never married/cohabited | 72 | 19.0 |
Married/cohabiting | 173 | 45.6 |
Separated/divorced/widowed | 134 | 35.4 |
None | 24 | 6.3 |
Primary | 136 | 35.9 |
Secondary | 142 | 37.5 |
Tertiary | 77 | 20.3 |
Employed | 91 | 24.0 |
Unemployed | 210 | 55.4 |
Retired | 76 | 20.1 |
< R1500.00 | 59 | 15.6 |
R1500.00–R3000.00 | 212 | 55.9 |
> R3000.00 | 108 | 28.5 |
1–4 members | 184 | 48.5 |
5 or more members | 195 | 51.5 |
18.5–24.9 (normal) | 42 | 11.4 |
25–29.9 (overweight) | 74 | 19.5 |
30–34.9 (class I obesity) | 84 | 22.2 |
35–39.9 (class II obesity) | 79 | 20.8 |
≥ 40 (class III obesity) | 99 | 26.1 |
48 | 12.7 | |
73 | 19.3 | |
1–3 days | 225 | 84.3 |
4–7 days | 42 | 5.7 |
≤ 50% | 53 | 19.0 |
51% – 75% | 153 | 40.4 |
> 75% | 173 | 45.6 |
No/limited control | 85 | 22.4 |
Good/complete control | 294 | 77.6 |
HTN, hypertension; T2D, type-2 diabetes.
, Two petty traders were excluded from analysis as their unemployment status was unclear;
, South African rand (ZAR).
Of the 372 available medical records, 361 (97.0%) were eligible for review. Most patients (
Annual risk factor monitoring (medical records audit 2009–2018).
Overall, complications were reported by 172 (45.4%) patients. Nearly one quarter (
Proportion of self-reported/medically recorded complications.
Number of complications | Self-reported ( |
Medically recorded ( |
||
---|---|---|---|---|
% | % | |||
0 | 207 | 54.6 | 288 | 79.7 |
1 | 84 | 22.2 | 40 | 11.1 |
2 | 53 | 13.9 | 18 | 5.0 |
≥ 3 | 35 | 9.2 | 15 | 4.2 |
The most common complications reported by participants were eye problems (
Proportion of self-reported and medically recorded complications.
Occurrence of self-reported complications was positively associated with age and female gender (
Determinants of self-reported and medically recorded disease complications.
Variable | Self-reported complications ( |
Medically recorded complications ( |
||||||
---|---|---|---|---|---|---|---|---|
Model 1 | Model 2 | Model 1 | Model 2 | |||||
Est. | s.e. | Est. | s.e. | Est. | s.e. | Est. | s.e. | |
Female gender | 0.62 |
0.30 | 0.52 |
0.31 | 0.62 | 0.38 | 0.81 |
0.41 |
Age | 0.04 |
0.01 | 0.05 |
0.01 | 0.03 |
0.01 | 0.02 | 0.02 |
Primary education | 0.46 | 0.50 | 0.54 | 0.51 | 0.18 | 0.67 | 0.28 | 0.69 |
Secondary education | 0.64 | 0.50 | 0.70 | 0.40 | 0.58 | 0.66 | 0.78 | 0.67 |
Tertiary education | 0.99$ | 0.53 | 1.09 |
0.53 | 0.76 | 0.69 | 0.81 | 0.70 |
Married or cohabiting | −0.12 | 0.24 | −0.19 | 0.24 | 0.59 |
0.29 | 0.52 |
0.30 |
Unemployed | −0.43 | 0.28 | −0.42 | 0.28 | −0.07 | 0.35 | −0.12 | 0.36 |
Income | −0.00008 | 0.0002 | −0.00004 | 0.0002 | −0.0005 | 0.0003 | −0.0005 | 0.0003 |
Oral T2D medication | - | - | −0.60 |
0.23 | - | - | −0.15 | 0.28 |
Perceived illness control | - | - | −0.02 | 0.06 | - | - | −0.35 |
0.08 |
Diabetes knowledge |
- | - | 0.13 | 0.10 | - | - | 0.23 |
0.13 |
Note: Logit estimates of socio-demographic characteristics (Model 1); and socio-demographic characteristics and self-management variables (Model 2).
T2D - type-2 diabetes; s.e., standard error; Est., estimate.
,
,
,
,
, Diabetes knowledge was only estimated for people with type-2 diabetes.
This study assessed the monitoring for disease risk factors and estimated the prevalence and factors associated with disease complications amongst T2D and HTN patients attending two CDMPs in an urban township in South Africa. A significant proportion (45%) of patients reported at least one disease complication whilst medically recorded complications were at one fifth (20%). Age, gender and perception of control over disease were associated with disease complications in this population. There was overall, poor yearly monitoring for disease risk factors in the CDMPs, but showing a gradual increase between 2009 and 2018. In this study, the best monitored risk factors were weight and blood pressure, similar to a study evaluating lifestyle care process amongst T2D patients in South Africa.
Other studies have recorded a higher prevalence of complications than ours, including a similar study performed amongst patients attending clinics in Cape Town in 1997, which recorded a prevalence of any grade of retinopathy of 55.4%, peripheral neuropathy (27.6%), and amputations (1.4%), but a lower proportion of circulation problems as indicated by absent foot pulses (8.2%).
The positive association between female gender and having a complication remaining even after controlling for BMI was unexpected and warrants further investigation. The positive association between tertiary education and occurrence of self-reported complications was unexpected, but could be explained by more adequate reporting by this subgroup. In terms of self-management, perceived control over illness was negatively associated with medically recorded disease complications, underlying the importance of adequate self-management support. Similar findings were reported in other studies
The limitations associated with a cross-sectional survey include the lack of identification of causal relationships. Data collection was based on self-reported information and medical records. Both methods are prone to several types of measurement error (e.g. recall bias, social desirability bias, incomplete registration, etc.). Despite complications, patients may also refrain from seeking care in other facilities which may have resulted in an underestimation of the medically registered complications, especially because screening of complications was poor.
However, including information on complications from two different sources provides the option of triangulation as both sources have an added value, despite being prone to a substantial measurement error. As such, we encourage readers to look at these figures through different lenses whilst on the one hand taking into account measurement error, but including the patient and health worker’s perspective on the other hand. The association between diabetes knowledge and outcomes of interest could also have been influenced by other factors which are not measured in this study, such as adherence to treatment regimens.
We hypothesise that using innovative strategies towards capacitating patients to control their disease during care provision could assist with reduction of disease complications in our setting. Suggested strategies for enhancing patient self-management and control of T2D/HTN include healthcare worker capacitation to provide optimal self-management support to patients and identifying patient success stories of self-management in real life settings to encourage other patients.
We recommend a re-orientation of CDMPs to meet actual patient needs whereby the clubs focus on providing individualised self-management support to patients
In this study, patients with comorbid T2D/HTN experienced disease complications despite attending CDMPs. Important risk factors related to feet and eyes were poorly monitored. Weight control was well monitored but poorly managed. Illness perception in terms of having control over the disease was negatively associated with prevalence of disease complications. To provide optimal care, CDMPs should make patient self-management a core aspect of their strategies, strengthen the risk factor monitoring and effectively address the identified risk factors. An evaluation component such as facility based clinical audits should be added to the CDMPs to monitor implementation of interventions.
The authors wish to acknowledge the field workers for their contribution towards successful implementation of this study and assisting with data collection. They also thank the healthcare workers and staff at the two health centres which were settings for this study for their assistance and support.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
T.M., P.D., and T.P. conceived, planned the project and supervised data collection. T.M, S.O., and J.D carried out the analysis. T.M., S.O., J.D., P.D., and T.P. contributed to the interpretation of the results. T.M. took the lead in writing the manuscript. All authors provided critical feedback and helped shape the research, analysis and manuscript.
European Commission Horizon 2020 Health Coordination Activities (Grant Number 643692) under call ‘HCO-05-2014: Global Alliance for Chronic Diseases: prevention and treatment of type 2 diabetes’. The SMART2D consortium includes a collection of universities in Europe and Africa. The contents of this article are solely the responsibility of the authors and do not reflect the views of the EU.
The data that support the findings of this study are available from the corresponding author, T.M., upon reasonable request.
The views expressed in this article are those of the authors and are not a position of the author institutions