To identify the research question a preliminary literature review was done of risks of Type 2 DM among older people living with HIV. This review provided useful insight into some factors that predispose a person to Type 2 DM, such as genetics, which are not age dependent. This insight on factors associated with Type 2 DM provided clarity on risks affecting older people living with HIV aiding in refining the research question to ‘what are the risk factors associated with the development of Type 2 DM among older people living with HIV?’ This refinement aided in the selection of studies.

The second step after the identification of the relevant research question is the selection of relevant studies to include in the search.¹⁹ Published literature was drawn from three databases, PubMed, Mendeley and the Cochrane Library. These databases were selected as they provided literature from different health disciplines. The selection of studies considered the context, the concepts and the population. Keywords for the selection of studies were ‘diabetes’, ‘risks’, ‘people living with HIV’. The literature reviewed was from 2012 to 2022 and was published in English. Studies that were not in English were excluded because of the time and cost involved in the translation and the risk of distortion in translation. The studies identified from the three databases were 996, and duplications were removed using the Mendeley Reference Manager. The literature search was conducted from 15 April 2022 to 13 July 2022, with follow-up performed until the 31 August 2022. Grey literature was searched through targeted websites such as the WHO and the International Diabetes Federation, and no results were found.

The inclusion and exclusion criteria detailed the studies to be considered, and these studies included the key words of ‘people living with HIV’, ‘risks’ and ‘Type 2 DM’. To select these studies, articles were selected based on the following inclusion criteria:

Literature was excluded based on the following criterion:

A team of reviewers reviewed selected sources to ensure the selection was comprehensive.¹⁸ Three steps were involved in selecting the studies: the first was a review of all titles and abstracts identified in the search by the first author; the second author then reviewed a minimum of 10% of the articles selected by the first author until an agreement was reached on the number of selected articles, based on the inclusion and exclusion criteria. The second step involved the review of all articles selected by both authors and the application of the inclusion and exclusion criteria by the first author, which were all articles that included ‘ageing’ as a variable. The second author then reviewed at least 10% of the selected articles after review by the first author. The third step was the first author referring all the articles they were unsure about including to the second author for review and a final check.

The fourth step in conducting a scoping review of literature is charting the data.¹⁸ Peters et al.¹⁹ note that the researcher will go to each paper and extract information that is relevant to the aim of the study. The Souza data extraction tool was used to extract data, and this included the authors, title, year of publication, country of origin, purpose, research objectives, research designs, sampling and population, key concepts and results that relate to the study objective.²⁰ Data were charted using Microsoft Excel (Microsoft Corporation, Redmond, Washington, United States).

The fifth step was collating study findings and discussing the implications for further research.¹⁸ This step involved three main steps: the first step was a basic numeric analysis which highlighted the years of study, the origins, the study designs and the context in which the studies took place. The second step was thematic content analysis. The themes extracted pertained to the risk factors for the development of Type 2 DM among older people living with HIV. The third and last step was describing the implications of the findings for practice, future research, education and policy.¹⁹

Ethical approval to conduct the study was obtained form the Research Ethics Committee of the University of South Africa (ref. no. 14056739_CREC_CHS_2022).

Numerical data for the findings were analysed using Microsoft Excel, which highlighted the years of the studies, the types of studies and the geographical coverage of the studies. A total of 15 studies were selected from the literature. The articles selected were from the years 2012 to 2022. Table 1 shows the years in which the articles retrieved were published.

Three (20%) of the studies reviewed were from 2017.^22,23,24 Another three studies^25,26,27 (20%) were from 2018, and another three^28,29,30 (20%) were from 2019. An additional three (20.00%) studies^31,32,33 were from 2021. One study³⁴ (6.67%) reviewed was from 2012, another (6.67%) study³⁵ was from 2016 and one (6.67%) study³⁶ was from 2022. No studies were retrieved from 2013, 2014, 2015 and 2020.

All studies reviewed were quantitative studies and used different study designs. Table 2 shows the study designs.

The majority of the studies (60.0%) were cross-sectional studies,^{23,25,26,27,28,31,32,34,35} while 13.33% of the studies^24,33 were secondary data analysis. Nonintervention cohort studies,³⁶ regional observational studies,³⁰ cohort studies²² and match control studies²⁹ consisted of 6.67% of each of the total studies reviewed.

With regards to the place of origin of the studies, Table 3 outlines where the studies were conducted.

The majority of the studies (20%) were conducted in Italy,^22,26,34 while 13.3% were conducted in Zimbabwe.^23,32 One study (6.67%) drew participants from Italy and Romania.²⁹ Another study³⁰ (6.67%) was a multicountry study from the Asia and Pacific region where data were taken from China, Cambodia, Hong Kong Special Administrative Region, Indonesia, India, Malaysia, Japan, the Philippines, Singapore, South Korea, Taiwan, Vietnam and Thailand. The remaining studies reviewed were from the United Kingdom (UK)²⁵ (6.67%), Denmark³⁶ (6.67%), Tanzania²⁷ (6.67%), Rwanda²⁸ (6.67%), South Africa³³ (6.67%), Canada²⁴ (6.67%), the US³² (6.67%) and Brazil³¹ (6.67%).

The population of interest in the study was older people living with HIV. From the 15 studies selected, the sample sizes ranged from 82 to 13 397. All studies included people living with HIV, and one match-control study from Italy²⁹ included children who had acquired HIV prenatally; this study also included older people and geriatric patients who were living with HIV aged more than 50 years and were assessed between the ages of 75 and 80 years. Two studies^26,36 used only respondents aged more than 65 years and more than 50 years, respectively.

In terms of the study participants’ dwellings, two studies^27,28 from Africa reported collecting data from rural locations. One study³² noted that the study respondents were from low socio-economic status. One study²⁵ in the UK also noted the diversity of the sample by revealing that respondents were residents of London but were composed of persons from 61 different countries, which represents the diversity of the population in the area. Table 4 summarises the study populations of the selected studies.

The first theme identifying the risk of Type 2 DM among older people living with HIV was the duration of HIV infection. The issue of long duration of HIV infection was noted in the findings of seven studies. Duncan, Goff and Peters,²⁵ in their study findings from a cross-sectional study in the UK, used a 2005 cohort and a 2015 cohort which was much older, with a mean age of 49 years compared to 41 years in the 2005 cohort. In their findings, the authors²⁵ reported a higher prevalence of Type 2 DM (32.3%) among the older 2015 cohort and attributed this to a ‘longer known duration of HIV infection’. The risk posed by a long duration of HIV infection and ageing is also described by a study conducted in Italy by Guaraldi et al.,²⁶ who found a higher prevalence of Type 2 DM among people living with HIV longer than 20 years. Notably, the respondents in this study²⁶ were more than 65 years of age (mean age 71 years among women and 72 years among men) and were compared to a cohort who were HIV-negative. In a later match-control study, Guaraldi et al.²⁹ used a sample from Italy which was older and a sample of people living with HIV who acquired HIV prenatally from Romania and concluded that HIV infection modulates the age at which people living with HIV acquire comorbidities, inclusive of Type 2 DM. The study²⁹ followed three control groups for 25 years: those who acquired HIV prenatally, those who acquired HIV after the age of 25 and those who acquired HIV at 50 years; the study concluded that a longer HIV duration in people who acquired HIV after the age of 50 years predisposes them to comorbidities such as Type 2 DM.²⁹ Notably, the conclusion made by Guaraldi et al.²⁹ that acquisition of HIV infection after the age of 50 increases the likelihood of Type 2 DM is in contrast to findings by Samad et al.,²⁴ who found that there is no association between incident Type 2 DM and HIV acquisition before or after the age of 50 years. In Italy, Galli et al.³⁴ in a cross-sectional study also concluded that Type 2 DM was associated with older age and longer duration of HIV infection. The risk posed by a long duration of HIV infection to the development of Type 2 DM is also highlighted by Høgh et al.,³⁶ in a study in Denmark with respondents drawn from the Copenhagen area with a mean age of 52 years, noted that insulin resistance was associated with duration of HIV infection and increased with every 5 years. This issue of the long duration of HIV infection increasing the likelihood of Type 2 DM is further elaborated by Cassenote et al.,³¹ in a study in Brazil, which found that the mean time to developing Type 2 DM was 10.5 years with a median age of 11.8 years. Samad et al.²⁴ in Canada also concluded that there was an association between the development of Type 2 DM and a long duration of HIV infection among older people living with HIV; the authors²⁴ found that the mean time to development of Type 2 DM among older people living with HIV was 9 years (range 6.2–11.6 years). In their study,²⁴ the authors used a sample of people living with HIV above the age of 50 years.

In the US, Tiozzo et al.,³² in an analytical cross-sectional survey, used a sample with a mean age of 48 years and also reported that a ‘longer duration of HIV; consequentially ageing’ is associated with risk of Type 2 DM. The study³² defined a longer duration of HIV infection as longer than 16 years and noted that the majority (59%) of their respondents, who were older than 50 years and had a longer duration of HIV infection, were most likely to develop Type 2 DM. In Africa, one study³⁵ from Zimbabwe also noted that a long duration of HIV infection predisposes patients to Type 2 DM. In this cross-sectional study conducted by Magodoro et al.,³⁵ a sample from Mpilo Central Hospital of 1225 people living with HIV concluded that the duration of HIV infection longer than 7 years was associated with common comorbidities such as Type 2 DM. The authors³⁵ further note, like Guaraldi et al.,²⁹ that age above 50 years and a long duration of HIV infection are associated with the likelihood of Type 2 DM.

A subtheme to emerge on the duration of HIV infection was that a lower nadir CD4 is associated with Type 2 DM.

The study by Høgh et al.³⁶ also associated the risk of insulin resistance with a long duration of HIV and a lower CD4 cell nadir. The authors³⁶ noted the risk of insulin resistance is increased among people living with HIV who had a lower CD4+ cell nadir of less than 200 cells/mm³. In Italy, Galli et al.,³⁴ who used a sample of people living with HIV from the St Raffaele Infectious Diseases Department, also noted the high incidence of Type 2 DM among people living with HIV who had a lower nadir CD4 and a long duration of HIV infection. Samad et al.,²⁴ concurring with these studies^34,36 conducted from the St Raffaele Infectious Diseases Department and greater Copenhagen area, found that Type 2 DM was related to a history of a lower CD4 cell nadir of less than 80 cells/mm³, a viral load more than 500 copies/mm³ and a long duration of HIV infection.

The second main theme emerging from the studies reviewed is the issue of adverse effects of ART on older people living with HIV predisposed to Type 2 DM. From this theme, subtheme of adverse effects of some ART increases likelihood of Type 2 DM.

The study conducted in Canada by Samad et al.,²⁴ who used a secondary data analysis method from the outpatient HIV clinic at St Paul’s Hospital in Vancouver, reported that Type 2 DM was associated with ART initiation between 1997 and 2004. In their study, the authors²⁴ compared this likelihood of developing Type 2 DM with people living with HIV who initiated ART between 2010 and 2015 who had a lesser incidence of Type 2 DM. The authors²⁴ also noted that Type 2 DM was associated with a longer duration on ART as well as having commenced ART in an ‘earlier calendar year’. Similar to this study²⁴ in Canada, Høgh et al.³⁶ used a 2-year cross-sectional study design with respondents from March 2015 through December 2016 with follow-up from April 2017 to April 2019, and they found that previous use of older-generation ART was also associated with the development of Type 2 DM. The study³⁶ defined older-generation ART as the thymidine analogues of stavudine, zidovudine and didanosine. Both studies by Samad et al.²⁴ and Høgh et al.³⁶ further elaborate that older-generation ART was associated with abdominal obesity, which is also a risk factor for developing Type 2 DM. Also concurring with Samad et al.²⁴ and Høgh et al.,³⁶ Guaraldi et al.²⁶ in Italy in a study with respondents aged more than 65 years revealed an association between Type 2 DM and first-generation ART use. Guaraldi et al.²⁶ also explained the cause of this risk of exposure to older-generation ART by revealing that older-generation ART caused lipodystrophy. Studies by Høgh et al.,³⁶ Guaraldi et al.²⁶ and Samad et al.,²⁴ in addition to concluding that older-generation ART is associated with the development of Type 2 DM, also associate Type 2 DM with long duration of HIV infection.

The study also found that some classes of ART used in older people living with HIV were associated with the development of Type 2 DM. Cassenote et al.,³¹ in a study in Brazil with 6727 respondents drawn from four out of five regions of the country, found that exposure to some nucleoside reverse transcriptase inhibitors (NRTIs) in older people living with HIV is associated with the incidence of Type 2 DM. The authors³¹ highlighted the likelihood of Type 2 DM among people living with HIV between the ages of 40 and 50 years and those aged more than 50 years. The study³¹ also found that exposure to stavudine for longer than 2 years was associated with incident Type DM in older people living with HIV. Concurring with Cassenote et al.,³¹ De Luca et al.,²² in a multicentre study of 42 tertiary healthcare facilities in Italy conducted with an almost similar sample size of 6505, found that diabetes could be predicted among older people who were treated with regimens containing an NRTI (stavudine and lamivudine) and an ‘unboosted’ protease inhibitor. Samad et al.,²⁴ albeit using a smaller sample size of 703, also noted that older people living with HIV had a higher incidence of Type 2 DM with the use of protease inhibitors in comparison to non-nucleoside reverse transcriptase inhibitors (NNRTIs). In their study, the authors²⁴ also explained that exposure to stavudine, didanosine, zidovudine, nelfinavir, indinavir and lopinavir was associated with the development of Type 2 DM among people living with HIV older than 50 years.

The adverse effects of ART use in older people living with HIV were also reiterated by Han et al.,³⁰ in a study in the Asia and Pacific region who, albeit using a sample with a mean age of 35, noted that ART initiation after the between ages 41 and 50 years and more than 50 years was associated with the development of Type 2 DM. Similarly, Duncan, Goff and Peters²⁵ noted that weight gain after ART initiation in older people living with HIV was associated with Type 2 DM.

From the studies reviewed, three studies found an association between a high BMI among older people living with HIV and the development of Type 2 DM.

In the UK, Duncan, Goff and Peters²⁵ found that Type 2 DM prevalence was higher in their second older (mean age 49 years) cohort assessed in 2015 who had a higher BMI of 27.4 kg/m². In their description of the 2015 cohort, Duncan, Goff and Peters²⁵ revealed that the older cohort was heavier and had lower rates of lipodystrophy. Høgh et al.,³⁶ in their study in Denmark, also found a higher incidence of insulin resistance among people living with HIV who had a BMI between 25 kg/m² and 29 kg/m² and who had central obesity. Guaraldi et al.²⁶ also associated the likelihood of multimorbidity inclusive of Type 2 DM with a higher BMI and an age above 75 years. The authors²⁶ described the risk posed by a high BMI as an independent predictor of Type 2 DM. Similarly, the study²³ in Zimbabwe also discussed that a higher incidence of Type 2 DM in people living with HIV older than 50 years was associated with an increase in age and a higher BMI. On the contrary, Samad et al.²⁴ concluded that the likelihood of Type 2 DM was also found among people living with HIV older than 50 years who had normal or low BMI.

From the literature reviewed, the issue of hypertension was also associated with older people living with HIV who developed Type 2 DM.

In South Africa, Chiwandire et al.³³ used a sample size of 978 in 2005, 1023 in 2008 and 2483 in 2017 and consistently reported from all three cohorts that age above 45 years and hypertension were predictors of Type 2 DM. The study³³ used a secondary data analysis method and data from people living with HIV aged more than 25 years from the South African National HIV Prevalence, Incidence, Behaviour and Communication, which is a population-based survey. Duncan, Goff and Peters,²⁵ in their 10-year cross-sectional study from 2005 to 2015 in the UK, also found an association between the development of Type 2 DM and comorbidity with hypertension. Duncan, Goff and Peters²⁵ defined hypertension with three consecutive readings of more than 140/90 mmHg using an electronic sphygmomanometer or current use of antihypertensives. The authors²⁵ from this second older 2015 cohort concluded that the respondents had a higher prevalence of hypertension and Type 2 DM. In Italy, Guaraldi et al.²⁶ in their match-control study also discuss that the acquisition of HIV after the age of 50 years is also associated with the development of both Type 2 DM and hypertension. The authors²⁶ used guidelines from the European Clinical AIDS Society to define hypertension, which defines hypertension Grade 1 as a reading of 140 mmHg – 159 mmHg/90 mmHg – 99 mmHg.

The fifth main theme that emerged from the data analysis was that a lack of knowledge on modifiable risk factors for Type 2 DM predisposes older people living with HIV to Type 2 DM. Biraguma et al.,²⁸ in a study in Rwanda, concluded that people living with HIV older than 50 years had poor knowledge of the modifiable risks of noncommunicable diseases such as Type 2 DM. The authors²⁸ used a sample with a mean age of 38 years and an age range of 18–70 years stratified into four groups: 18–30 years, 31–40 years, 40–50 years and more than 50 years. This sample was drawn from Southern and Northern provinces, which are rural areas, and Kigali City, an urban area. In assessing knowledge levels, Biraguma et al.²⁸ used a data collection tool that enquired if respondents had heard of diabetes and knew of risk factors such as smoking, physical inactivity, alcohol use, stress and an unbalanced diet. In a similar study in Tanzania, Kagaruki et al.²⁷ found similar results to those found by Biraguma et al.²⁸ in Rwanda. The authors²⁷ used a sample of 745 people living with HIV from rural Mbeya and urban Dar-es-Salam and found that 42.4% of people living with HIV older than 40 years had low knowledge of modifiable risk factors for developing Type 2 DM. Notably, the study in²⁷ Tanzania stratified the sample into two categories: those below 40 years and those above 40 years. In collecting data on knowledge of risks of Type 2 DM, the authors,²⁷ like Biraguma et al.,²⁸ also assessed knowledge on healthy diet, alcohol use, family history of diabetes and physical exercise.

While every effort was made to ensure the accuracy of results reported in this study, the study was limited by the lack of consistency in the definition of Type 2 DM from the studies reviewed. Da Luca et al.²² defined Type 2 DM as fasting blood sugar levels above 125 mg/dL or a treating clinician’s diagnosis and use of antidiabetic agents. Han et al.³⁰ defined Type 2 DM using any of four criteria: fasting blood sugar of more than or equal to 126 mg/dL or glycated haemoglobin of more than 6.5% or a 2-h fasting plasma oral glucose tolerance test of more than 200 mg/dL or a random plasma glucose more than 200 mg/dL. In two other reviewed studies, Chimbetete et al.²³ and Samad et al.²⁴ defined Type 2 DM as random blood sugar of more than 11.1 mmol/L, confirmed by a fasting blood sugar above 7 mmol/L and symptomatic people living with HIV who had random blood sugar more than 11 mmol/L. Duncan, Goff and Peters²⁵ defined Type 2 DM using one criterion of a fasting blood sugar of more than 7 mmol/L. Høgh et al.³⁶ defined insulin resistance using homeostasis model assessment – insulin resistance (HOMA-IR).

Based on the study findings, clinicians working with older people living with HIV should screen older people living with HIV for these excess risks for developing Type 2 DM and provide health education to reduce the likelihood of developing Type 2 DM. These findings on risks of Type 2 DM among older people living with HIV should also be used by educators teaching Type 2 DM among different population groups. Furthermore, additional research is recommended on the risks of developing Type 2 DM that older people living with HIV face in specific countries to reduce the incidence of Type 2 DM. The risks of developing Type 2 DM among older people living with HIV should also be used by policymakers in the development of guidelines aimed at reducing Type 2 DM.