Body Mass Index Changes and Predictors among Adults Living with HIV/AIDS Who Are on Anti-Retroviral Therapy at Chiro General Hospital, Eastern Ethiopia: A facility-based retrospective cohort study

doi:10.21203/rs.3.rs-4930429/v1

Download PDF

Research Article

Body Mass Index Changes and Predictors among Adults Living with HIV/AIDS Who Are on Anti-Retroviral Therapy at Chiro General Hospital, Eastern Ethiopia: A facility-based retrospective cohort study

https://doi.org/10.21203/rs.3.rs-4930429/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: Human immunodeficiency virus continues to be a major global public health issue. Body mass index is a general indicator of nutritional status and has emerged as a powerful predictor of morbidity and mortality among adult PLHIV initiating antiretroviral therapy in resource-limited settings. However, there is a dearth of information regarding longitudinal changes in body mass index and its predictors among adult PLHIV in Ethiopia, particularly in the study area. This study aimed to assess body mass index changes and their predictors among adults living with HIV/AIDS who were receiving on antiretroviral therapy at Chiro General Hospital, Eastern Ethiopia from August 15, 2023 to September 30, 2023.

Methods: A Facility-based retrospective cohort study was implemented among 1049 randomly selected charts of adults living with HIV/AIDS. The data were collected by reviewing charts of clients and antiretroviral therapy registers. The data were entered into Epi data statistical software version 4.6 and exported to SPSS version 25 for analysis. Descriptive statistics were used to describe the characteristics of the patients. A linear mixed effect model was used to identify the predictors of body mass index change. A P value of less than 0.05 was considered statistically significant.

Results: Generally, in this study patients presented a linear increase in the mean BMI from 19 kg/m² baseline to 21.2 kg/m² at the 5^th year of follow up. Moreover, the following variables were identified as independent predictors of BMI change: age (β = 0.58, 95% CI; 0.043, 0.072), marital status (β = -0.275, 95% CI: -0.457,-0.093 ), advanced WHO stage (β = -0.496, 95% CI: -0.548, -0.443 ), CD4 count (β = 0.001, 95% CI: 0.001, 0.001), duration of antiretroviral therapy (β = 0.005, 95% CI: 0.001,0.009), time of follow up (β = 0.205, 95% CI: 0.198,0.212), no ART shift (β = -0.844, 95% CI: -1.135, -0.552), no CPT (β =0.591, 95% CI: 0.365,0.817), urban residence (β=0.767, 95% CI:0.401,1.132) and good adherence to ART (β= 0.975, CI:0.302, 1.649).

Conclusion: There was a significant improvement in the mean BMI over time and a reduction in the rate of undernutrition during the follow-up period.

BMI change

HIV/AIDS

ART

Weight

Human immunodeficiency virus (HIV) continues to be a major global public health issue. According to the joint United Nations Program on HIV/AIDS (UNAID) report in 2021, an estimated 37.7 million people were living with HIV with a global prevalence of 0.7% among adults (1). According to 2021 global USAID updates, the vast majority of HIV cases are located in low and middle income countries (2). In Ethiopia approximately 603,537 people live with HIV/AIDS, and 81% are accessing ART. The number of deaths due to HIV/AIDS has significantly decreased due to the introduction of antiretroviral therapy (ART) (3).

The body mass index (BMI) of patients with HIV/AIDS is an important predictor of ART outcome, including the prediction of CD4 + T-cells changes and viral load levels. A negative change in the BMI of patients was also found to be an independent predictor of dropout from HIV care (4). HIV associated weight loss and wasting syndrome are the most frequently occurring AIDS-defining conditions and are associated with increased risks of mortality and morbidity. Conversely, weight gain after antiretroviral therapy (ART) initiation is a good prognostic sign and is associated with a lower risk of mortality in underweight and normal-weight patients (5). The BMI of people living with HIV/AIDS can be affected by clinical factors (WHO staging, functional status), CD4 count, viral load level and other sociodemographic predictors.

Body mass index (BMI) was calculated as weight divided by height in meters squared. It is a critical indicator of nutritional status in patients with human immunodeficiency virus (HIV) infection. The World Health Organization (WHO) has recommended the following categories on the basis of BMI values: underweight (< 18.5 kg/m² ); normal (18.5–24.9 kg/m² ); overweight (25–29.9 kg/m² ); obese (30–39.9 kg/m² ) and severely obese (≥ 40 kg/m²) (6). A body mass index of less than 18.5 kg/m² is a general indicator of poor nutritional status and has emerged as a powerful predictor of morbidity and mortality among adult PLHIV initiating ART in resource-limited settings (7).

World Health Organization (WHO) clinical staging, immunological (CD4-cell count), and routine viral load tests are methods used to monitor the treatment outcomes of people living with HIV/AIDS. Compared with viral load test monitoring, immunological and clinical monitoring have lower positive predictive value and poor sensitivity for identifying treatment failure (8).

There is no doubt that viral load testing is effective in monitoring patient treatment outcomes and HIV resistance, as per WHO guidelines. However, in resource constrained areas that may not always be able to perform viral load testing in a timely manner, BMI measurements are readily available and routinely assess objective measures that may predict early viral nonsuppression (high viral load). BMI is an inexpensive, noninvasive and simple anthropometric measurement that helps identify early clinical failure among PLHIV. Therefore, the results of this study can be used by District health care providers working with patients with HIV/AIDS to better serve their patients on ART. Stake holders including NGOs working in areas with people with HIV/AIDS, can also use the results of this study to monitor and evaluate the nutritional status of patients on ART. Finally, the findings have important implications for policy regarding nutritional status and monitoring people living with HIV/AIDS on antiretroviral therapy. The objective of this research was to assess BMI change and its predictors among adults living with HIV/AIDS who are on ART at Chiro General Hospital, Eastern Ethiopia from August 15, 2023 to September 30, 2023.

Study Setting, Period and Design

The study was conducted from August 15, 2023 to September 30, 2023 at Chiro General Hospital. Chiro General Hospital provides comprehensive HIV prevention and control services (case detection, care and treatment of HIV positive individuals) by implementing new initiatives in collaboration with different partners/donors such us the CDC and ICAP. The ART unit is staffed with two BSc nurses, one general practitioner doctor, one pharmacist, two data clerks, two case managers and two adherence supporters. According to the DHIS2/Smart Care report, 1177 people were alive and on ART at the end of June 2022. Among these, 42 were children under 15 years old and 1135 (96%) were adults ≥ 15 years old. When the data were disaggregated by sex, 768 were females and 409 were males (source: Smart Care electronic database, ART registers, DHIS2). A Facility- based retrospective cohort study was performed.

Study population, inclusion criteria and exclusion criteria

Study population: All adult clients (≥ 15 years) living with HIV/AIDS who were on ART at Chiro General Hospital from July1, 2017, to June 30, 2022, whose cards were available during the data extraction period. Adult clients (≥ 15 years) living with HIV/AIDS who were receiving ART between July1, 2017, and June 30, 2022, at Chiro General Hospital were included and pregnant women, severely ill patients and those clients who were transferred without baseline data were excluded from the study.

Sample size determination and sampling technique

G*Power version 3.1.9.7 statistical software was used to estimate the required sample size for both of the objectives. For the first objective (BMI change), the following parameters were fed to the software on the basis of mean differences among groups (i.e., ANOVA for repeated measurements): effect size = 0.25 (recommended/default), corresponding Z-value for 95% confidence level = 1.96, power = 80%, expected maximum number of groups = 5 and correlation among repeated measures = 0.5 (default/recommended value). Finally, the software produced a sample size of 110.

For the second objective (predictors of BMI change), the following parameters were entered into the software and linear mixed effect model (LMM) analysis was applied to identify the predictors of BMI change: effect size (f²) = 0.02 (small effect size), corresponding Z-value for 95% confidence level = 1.96, power = 80%, number of total planned variables = 15 and number of planned variables to be tested = 15 (all the planned independent variables were assumed to be tested). The software generated a sample size of 954.

A comparison of the sample sizes estimated for the first and second objectives revealed that the second objective provided a greater sample size (i.e., 954). When 10% of the charts were lost, the final sample size was set to1049.

A computer-generated simple random sampling technique was used to draw 1049 sample charts from a total of 1135 adult clients who were receiving ART from July 2017 to June 2022. Initially, a list containing the medical records of adult clients was obtained from the electronic database (smart care software) of the ART clinic. The medical registration number was subsequently exported to Microsoft Excel. A random number was generated from Excel. Then, 1049 charts were randomly selected.

Data collection instrument: Data collection was performed via a data extraction checklist to review patients’ records. The contents of the checklist included sociodemographic characteristics and clinical variables. The data extraction checklist used to assess predictors of BMI change was prepared by a principal investigator with modifications by reviewing different studies (4), (9) and by referring to intake forms A and B (intake forms are completed at enrollment, and they capture demographic data) and follow-up charts developed by the Ministry of Health of Ethiopia.

Data collectors

Two BSc nurses (ART providers) and one supervisor were recruited to collect the data.

Data collection procedures

The secondary data were collected from the ART unit for 45 days. Baseline and subsequent clinical and sociodemographic data were collected from the records of randomly selected patients. The principal investigator was under all supervision throughout the data collection period. Height data were collected once at baseline and were converted to m², and weight, CD4, WHO stage and functional status data were collected at baseline and then every six months over five years (ten repeated measurements). Body mass index was calculated by dividing weight by height in meters squared at the baseline and then every six months for the five year follow-up period. The viral load was measured annually for five years.

Data Quality Control

To maintain the quality of the data, one-day orientation was given for the supervisor and the data collectors on the objective, purpose, procedures of the study and details of the checklist. The principal investigator and the supervisor provided daily supervision during the data collection period. Pretesting was conducted on 5% (57 charts) of the sample size at Gelemso General Hospital.

Data Processing and Analysis

After data collection, the data were coded, cleaned and entered into Epi data statistical software version 4.6 and exported to SPSS version 25 for analysis. Descriptive statistical analyses such as simple frequency, mean and median, and measures of variability such us range and standard deviation were used to describe the characteristics of the patients. The information was presented as frequencies, summary measures, tables and figures. The model goodness of fit was assessed via LR/AIC test. The normality assumption was checked via a Q-Q plot. By using the likelihood ratio test, model comparisons were made between the random intercept model and the random intercept and slope model. The independent variable that was significantly associated with longitudinal BMI change (i.e., P value less than or equal to 0.25) in the bivariable analysis was included in the final multivariate model of the linear mixed-effects model. To identify predictors associated with BMI change, the β coefficient at the 95% CI was estimated. To declare a statistically significant association of a predictor in the final model of the LMM, a P value of less than 0.05 was used as a cutoff point.

Sociodemographic characteristics

A total of 1049 records of HIV positive clients on ART were retrieved with 100% card retrieval rate. The mean age of the study participants was 37.5 years (SD = 9.7) and more than half of them (65.5%) were females. More than three-fourths, 926 (88.3%) of the study participants were from urban areas. With respect to religion, 652 (62.2%) of them were Orthodox, whereas 294 (28%) and 101(9.6%) were Muslim and Protestants respectively. More than half of the study subjects attended primary (32.3%) or secondary school (35.2%). On the other hand, 12.4% of them did not attend any formal education, whereas 21.1% had attended preparatory and colleges (tertiary educations). Almost two-thirds of the clients (65.7%) were married, whereas 16.7%, 13.1% and 4.6% were single, divorced and widowed, respectively. With respect to occupational status, 29.6% were house wives and 23.2% and 16.4% were daily laborers and privately employed respectively (Table 1).

Clinical characteristics of ART patients

Among the participants, 51.9% had received cotrimoxazole preventive therapy (CPT), whereas 65.1% had received TB preventive therapy (TPT). Most of the clients were on TDF+3TC+DTG (89.4%) followed by AZT+3TC+ATV/r (3.4%) and TDF+3TC+ATV/r (3.1%). Almost one fifth, 19.4% of the ART clients had a history of an ART regimen shift at some point during follow-up. Among the 203 (19.4%) ART users who had a history of ART shifts, the majority (83.4%) were shifted from (the TDF+3TC+EFV) regimen to (the TDF+3TC+DTG) regimen during the course of treatment. Most, 95.7% of the clients reported good ART adherence, whereas 2.9% and 1.4% reported fair and poor ART adherence respectively. The prevalence rates of opportunistic infection (OI) and TB coinfection were 5.1% and 3.9% respectively. Comorbid chronic conditions were present in 2.4% of the ART clients (Table 2).

Comparison of BMI changes: ANOVA and chi-square tests

Since the distribution of the CD4 count was skewed, as evidenced by a skewed mean above the median and histogram, the median and IQR (quartile 3 minus quartile 1) are presented. For BMI, the mean and the corresponding standard deviation were reported, as the distribution was approximately normally distributed.

A statistically significant improvement in the clinical parameters of the ART patients was observed after they started ART. The percentage of ART clients with undernutrition has significantly declined compared with the baseline prevalence of undernourishment. For instance, the prevalence of under nutrition was 37% at six the month follow-up after starting ART and this value declined to 26.6% (at 18 months), 17.9% (at 30 months) and 10.8% (after five years of ART follow-up). The percentage of ART clients with bed ridden decreased by 76%, that is from 1.3% to 0.3%; and there was an almost 50% decline in the number of ambulatory patients; specifically, from 6.9% to 3.7%. Furthermore, meaningful improvement in the clinical staging of ART clients was also observed where the percentage of ART clients with advanced HIV disease (WHO stage III and Stage IV) declined from 13.6% to 3%; from 2.5% to 0.6% at baseline and after five years of ART follow up. The percentage of stable ART clients (WHO stage I) rose from 77.3% to 92.7% from baseline to five years of ART treatment. With respect to immunological recovery, the CD4 count significantly improved over the follow-up periods (P < 0.001), with an increase from a median CD4 count of 303 cells/mm³ at baseline, to 380 cells/mm³ (at 18 months), 418 cells/mm³ (at 30 months) and 470 cells/mm³ (at the fifth year after ART), indicating an average increase the CD4 count of 167 cells/mm³ over the five-year period (Table 3).

Using the Bonferroni technique and one-way ANOVA and post hoc comparisons, a statistically significant improvement in the mean BMI was observed during the early care and follow-up periods (P value <0.001). However, the differences during the late consecutive follow up periods were not statistically significant (p-value = 0.99). For instance, these differences were not significant between the early follow up periods. Moreover, this is clearly depicted in Figure 1, which shows a linear improvement in the mean BMI across the follow-up periods after starting ART.

The study also revealed a linear mean BMI increase from 19 kg/m² at the baseline to 21.2 kg/m² at the end of the follow-up period. The absolute mean BMI change was 2.2 kg/m² and the relative change was 22 kg/m². The mean change in BMI from baseline at each follow-up point was also computed as follows: at 6 months 0.1, at 12 months 0.4. at 18 months 0.6, at 24 months 1, at 30 months 1.2, at 36 months 1.4, at 42 months 1.6, at 48 months 1.8, at 54 months 2 and at 60 months 2.2 (Table 3).

Predictors of BMI Change Over Time: Linear Mixed Effect Model

The subject ID was considered a random factor in the linear mixed model, as it allows us to account for the within-subject correlation in BMI measurements over time to capture individual-specific variations in BMI trajectories. It also better elucidates the associations between individual level factors and outcomes. It allows for the estimation of subject-specific intercepts and slopes. This helps to capture the within-subject correlation structure and control for individual-level differences in BMI trajectories over time. A multivariable linear mixed effect model was fitted in SPSS version 25 via the Alkaiks information criterion under the maximum likelihood estimation method.

Compared with those of previous models, significant decreases in the AIC and BIC values were observed in the final model. Hence, an LMM was fitted that included eleven predictors of BMI change with an AIC value of 32426.4 and a BIC value of 32542.6. In addition, under the maximum likelihood test, the model fitness has increased significantly compared with that of previous models (log likelihood = -16197.2; X²= 6487.8; p <0.00001). Hence, the fitted model is presented in Table 4. Initially, a bivariable analysis was conducted and variables with p values ≤ 0.25 were included in the multivariable analysis. The results of a linear mixed effect model to identify predictors of BMI change among ART clients over the follow-up periods are presented in the table below. The final model examined various factors, both categorical and continuous, and their coefficients (β) were estimated to determine their influence on BMI change. Hence, age, residence, marital status, duration of ART, duration of follow-up, ART adherence, ART shift, CPT, WHO stage and CD4 count were found to be significant predictors of BMI change over time.

Compared with rural residents, urban residents presented a 0.767 kg/m² greater increase in average BMI (β = 0.767, 95% CI; 0.401, 1.132). With respect to marital status, the average BMI increment was 0.275 kg/m² lower among married participants than among single participants (β = -0.275, 95% CI; -0.457, -0.093).

ART patients without a history of ART shift from one regimen to another (from TDF+3TC+EFV to TDF+3TC+DTG was the common shift) were more likely to have a decrease in BMI than were those with a history of ART shift (β = -0.844, 95% CI; -1.135,-0.552). Additionally, the coefficients for ART adherence revealed that clients with good adherence had a 0.975 kg/m²increase in BMI compared with clients with poor adherence (β = 0.975, 95% CI: 0.302, 1.649). Compared with those receiving CPT, individuals not receiving cotrimoxazole preventive therapy (CPT) had greater changes in BMI (β = 0.591, 95% CI 0.365, 0.817). The coefficient for WHO clinical staging indicated that being in higher stages (stage III or IV) was associated with a decrease in BMI compared with being in lower stages (stage I or II) (β = -0.496, 95% CI:-0.548,-0.443).

With respect to the continuous predictors, the coefficient for age indicated that for each one-year increase in age, there was a 0.058 kg/m²increase in BMI (β = 0.058, 95% CI: 0.043,0.072). Moreover, each one-month increase in ART duration was positively associated with an increase in BMI (duration: β = 0.005, 95% CI: 0.001, 0.009). Similarly, a greater increase in BMI was observed in the early follow-up periods than in the late visits after initiating ART. The coefficient for the CD4 count revealed that as the CD4 cell count increased by one cell/mm³, the mean BMI significantly increased by 0.001 kg/m² (β = 0.001, 95% CI; 0.001,0.001) (Table 4).

In general, in this study patients presented a linear increase in the mean BMI and decrease in the rate of undernutrition (underweight) over time. The predictors found to have a statistically significant effect on BMI change over the follow-up period via a multivariable linear mixed effect model were:– age in years, residence, marital status, duration of ART, time of follow- up, ART adherence, ART shift, CPT, WHO clinical staging and CD4 count.

The prevalence of undernutrition (BMI < 18.5 kg/m²) was 37% at the six month follow-up and this value declined to 26.6% (at 18 months), 17.9% (at 30 months) and 10.8% (after five years of ART follow up). This finding is consistent with a study conducted in Greece; according to this study, the prevalence of obesity increased from 5.7% at ART initiation to 12.2% after four years of follow-up (10). This greater effect of ART on weight gain or reduction in the proportion of underweight clients may be related to the fact that current drugs are better tolerated and easier to use regimens, resulting in better ART adherence, increased appetite and higher caloric intake (11).

The results of the current study revealed that the mean BMI of ART clients significant improved after the baseline. For example, the mean BMI at baseline was 19 kg/m² and it reached 21.2 kg/m² five years after the initiation of ART. This finding is in line with a longitudinal retrospective study conducted in Nigeria, which revealed a significant increase in the median BMI after the beginning of ART (z = 21.864, p < 0.001), with a moderate effect size (r = 0.334). According to this study, the median BMI increased from 22.2 kg/m² (IQR: 19.8, 24.8) at baseline to 23.4 kg/m² (IQR: 21.0, 26.4) at 12 months (12). The possible reason for the increase in BMI gain can be attributed to different factors, such as; the advantage of ART in preventing HIV- related opportunistic infections such as wasting syndrome, chronic diarrhea and free access to medical care (13). Overall health improvement and a reduced illness burden can also be among the reasons.

The current study revealed that, being married was associated with a 0.275 kg/m² lower BMI than being single (β = -0.275, 95% CI; -0.457, − 0.093). This result is supported by a study conducted at Mekele, which reported that being married is associated with a lower BMI by 1.178 kg/m² than being single (β= -1.178, p < 0.00304) (14). However, a contrary result was reported in a study conducted in South Africa, which reported that, being married/cohabitating was associated with an increase in BMI (ARR = 0.58, P = 0.03) compared with being single (15). This contradictory result might be due to the type of study design used or sociocultural differences between nations, and further investigation is needed.

Place of residence was another predictor that had a positive effect on BMI variation. Compared with clients who reside in rural areas, those who reside in urban areas presented an average BMI increase of 0.767 kg/m² during the follow-up period (β = 0.767, 95% CI: 0.401, 1.132). This finding is in line with a study conducted in North Ethiopia (β = 0.45, p = 0.000) (16). This difference in weight gain between urban and rural areas could be explained by the different lifestyle factors in urban and rural areas.

This study also revealed that, clients with good self-reported ART adherence had a significant increase in BMI. Thus, the average BMI increased by 0.975 kg/m² compared with that of clients with poor ART adherence (β = 0.975, 95% CI 0.302, 1.649). This finding is consistent with the findings of a study conducted in Northwest Ethiopia, which reported that, compared with clients with poor ART adherence, those who took the prescribed treatment well or who had at least > 85% good adherence had a 3.988 kg greater mean weight increase (β = 3.988, p < 0.0001) (17). These consistent findings could be explained by a relatively better understanding of life-long care and treatment/chronic care among clients with good adherence, which further improved their weight and body mass index.

According to the results of the present study, a one-month increase in ART duration was associated with a 0.005 kg/m² increase in the mean BMI (β = 0.005, 95% CI; 0.001, 0.009). Similarly, a previous study in Ethiopia reported similar findings, indicating that duration of ART was significantly association with BMI change (duration β = 0.43, p < 0.001) (5). Another consistent result reported in a study conducted in the United States of America, in which a longer duration of ART, was associated with a greater increase in BMI (P < 0.05) (13).The main reason for the clinical improvement, including a positive change in BMI associated with a longer duration of ART, is thought to be a reduction in the viral load. Patients who stay longer on care and treatment are more likely to develop preventive health behaviour and to receive social support that improves their clinical outcome (18).

With respect to the continuous predictors, in the present study, the coefficient for age indicated that a one-year increase in age, was correlated with a 0.058 kg/m² increase in BMI (β = 0.058, 95% CI; 0.043,0.072). A study conducted in Switzerland also indicated that BMI change was associated with older age (β = 0.15, P = 0.002) (19). Similarly, another study from Ethiopia reported that the age of respondents was significantly associated with BMI change (β = 0.136, 95% CI; 0.044:0.227) (9). This might be due to changes in body composition and lean body mass in addition to the effects of ART.

According to the current study, those clients without a history of an ART shift from one regimen to another (TDF + 3TC + EFV to TDF + 3TC + DTG was the common shift) were associated with a 0.844 kg/m² decrease in the mean BMI compared with those with a history of an ART shift (β = -0.844, 95% CI: -1.135, -0.552). This might be because remaining on a suboptimal regimen may not cause clients to gain the expected weight. A similar finding was reported in a study conducted in Burkina Faso, which reported that taking PI-based combination therapy was associated with a decrease in BMI (β= -0.01, p = 0.001) (20). On the other hand, in addition to the previously mentioned studies, a retrospective cohort study conducted in Eswatini (Swaziland) reported that, the BMI rate of change increased by 1.2 kg/m² (P = 0.001) among clients who had ART shifted to the “DTG’’-based regimen” (21). This is because dolutegravir (DTG) causes significant weight gain, as evidenced by many studies (22). However, no statistically significant association was observed between weight gain and switching to the “DTG’’-based regimen” according to a retrospective study conducted in London (23).

The coefficient for the CD4 + T-cell count revealed that as the CD4 + T-cell count increased by one cell/mm³, the average BMI increased by 0.001 kg/m² (β = 0.001, 95% CI: 0.001, 0.001 ). A consistent result was reported in a study conducted in the USA, in which patients with a higher CD4 count had an increased BMI (P = 0.001) (13). In addition, this finding is also supported by another retrospective cohort study at Bahirdar, which reported that as the baseline CD4 cell count increased by one cell per mm³, the odds of having a normal BMI for HIV patients increased by 5.4% (AOR = 1.0538, 95% CI:(1.0032, 1.2489), P-value = 0.0231) (24). Another study that was conducted in Switzerland is also in line with the results of the present study; according to the present study, HIV patients with lower baseline CD4 (0–99 and 100–199 cells/µl) had the greatest effect on BMI (β = 1.64, P = 0.001and β = 0.46, P = 0.001 respectively) compared with patients with baseline CD4 level > 350 (19). These consistent findings can be explained by immune recovery or improved immune function, which can lead to better absorption and nutrient utilization, potentially resulting in weight and BMI gain.

The coefficient for WHO clinical staging indicated that being in a higher stage or advanced HIV/AIDS disease (stage III or IV) was associated with a 0.496 kg/m² decrease in BMI compared with being in a lower WHO clinical stage (stage I or II) (β= -0.496, 95% CI: − 0.548, -0.443). This might be due to the development of OI at advanced stages of diseases, which may lead to increased energy expenditure or negative energy balance. This finding is consistent with a retrospective longitudinal study conducted in Northwest Ethiopia, which reported that, the BMI of clients with advanced WHO stages was 0.6 kg/m² lower than that of clients with mild stages (β=-0.6, P = 0.002) (25) (24).

Moreover, the results of the current study revealed that, individuals who did not receive CPT had a 0.591 kg/m² higher BMI than those who received CPT did (β = 0.591, 95% CI 0.365,0.817). This might be because individuals not receiving CPT are stable clients such that their BMI can be higher than that of their counterparts. However, a contrary result was reported in a previously conducted study in Ethiopia, in this study, clients who received cotrimoxazole prophylaxis had an increased BMI compared with patients who did not receive prophylaxis (4). These contradictory findings might be due to individual variation and lifestyle factors and need further study.

The current study revealed that a greater increase in BMI score was observed in the early follow-up periods/visits than in the late visit time after initiating ART. Accordingly, the mean BMI increased by 0.205 kg/m² as the follow-up time increased by six months (β = 0.205, 95% CI: 0.198, 0.212). This finding is in line with a study performed in Hawasa, Ethiopia, which reported that weight increased by 1.44 kg as the treatment time increased by six months (β = 1.44; 95% CI:1.23–1.66, p < 0.0001) (26, 27). This could be explained; by the fact that with time, antiretroviral combination drugs have started to effectively control replication of the virus so that the number and severity of illnesses decreases which in turn contributes to weight gain.

In this study, a significant improvement in the mean BMI score during the five year follow-up period was observed. The percentage of underweight individuals declined during consecutive periods after starting ART. These changes were accompanied by improvements in clinical and immunological recovery indicators, which could strongly predict positive linear changes in BMI scores over the follow-up period. Furthermore, age in years, residence, marital status, duration of ART (in months), time of follow up, ART shift, adherence to ART, CPT, WHO clinical staging and CD4 count were found to be significant predictors of BMI change over time.

AIDS: Acquired immune deficiency syndrome, ART: Anti-retroviral therapy, BMI: Body mass index, CDC: Centers for Disease Control, CD4: Cluster of Differentiation, CPT: Cotimoxazole preventive therapy; CVD Cardio Vascular Disease, DHIS2; District Health Information System 2, HIV: Human Immune Virus, HMIS: Health Management Information System. ICAP: International Center for AIDS Care Program, IPT: Isoniazed Preventive Therapy, LMIC: Low- and Middle-Income Countries, LMM: Linear Mixed Effect Model, OI: Opportunistic Infection, PLHIV: People Living with HIV, TPT: Tuberculosis Preventive Therapy, WHO: World Health Organization

Ethics approval and consent to participate

Ethical clearance to conduct this study was obtained from Haramaya University (HrU), the College of Health and Medical Sciences (CHMS), and the Institutional Health Research Ethics Review Committee (IHRERC).The official letter of support obtained from the CHMS was submitted to Chiro General Hospital. The head/administer of the hospital was clearly informed about the purpose, procedures, risks and benefits of the study. Informed, voluntary, written and signed consent was subsequently obtained from the hospital head. Confidentiality was maintained throughout the research process by removing personal identifiers.

Consent for publication

Not applicable

Availability of data and materials

The materials and data that support the findings of this study are available upon request. Researchers interested in accessing the materials or data used in this research should contact Sintayehu Amsalu via [email protected] for further information.

We are committed for transparency and ensuring that the necessary resources are accessible for further research and validation of the results presented in this manuscript.

Competing interests

The authors declare that they have no competing conflicts of interest

Funding

Self-sponsored

Authors’ contribution

SA carried out the overall design and execution of the study, performed the data collection; and statistical analysis and drafted the manuscript. BG and TW conceived the study, critically revised the design of the study; data collection techniques and helped with the statistical analysis. SA and AO assisted in the statistical analysis of the data and the discussion. All the authors read this manuscript and approved it for submission.

Nygren-Krug H. The Joint United Nations Programme on HIV/AIDS: Oxford Scholarship Online Oxford, UK; 2018.
Stover J, Glaubius R, Kassanjee R, Dugdale CM. Updates to the Spectrum/AIM model for the UNAIDS 2020 HIV estimates. Journal of the International AIDS Society. 2021;24:e25778.
Oumer Abdu A, Abebaw Mekonnen B. Episodes of undernutrition and its predictors among clients on antiretroviral treatment in southwest ethiopia: A record review. HIV/AIDS-Research and Palliative Care. 2021:61-71.
Baraki AG, Gezie LD, Zeleke EG, Awoke T, Tsegaye AT. Body mass index variation over time and associated factors among HIV-positive adults on second-line art in north-west Ethiopia: a retrospective follow-up study. BMJ open. 2019;9(9):e033393.
Alebel A, Demant D, Petrucka PM, Sibbritt D. Weight change after antiretroviral therapy initiation among adults living with HIV in Northwest Ethiopia: a longitudinal data analysis. BMJ open. 2022;12(2):e055266.
Li X, Ding H, Geng W, Liu J, Jiang Y, Xu J, et al. Predictive effects of body mass index on immune reconstitution among HIV-infected HAART users in China. BMC infectious diseases. 2019;19(1):1-9.
Liu E, Makubi A, Drain P, Spiegelman D, Sando D, Li N, et al. Tuberculosis incidence rate and risk factors among HIV-infected adults with access to antiretroviral therapy. Aids. 2015;29(11):1391-9.
Waju B, Dube L, Ahmed M, Assefa SS. Unsuppressed Viral Load Level in Public Health Facilities: Nonvirological Predictors among Adult Antiretroviral Therapy Users in Southwestern Ethiopia. HIV/AIDS (Auckland, NZ). 2021;13:513.
Weldesenbet AB, Ayele TA, Sisay MM, Tusa BS, Kebede SA. Predictors of change in weight among people living with HIV on antiretroviral treatment in West Hararghe zone, Ethiopia: a retrospective longitudinal study. HIV/AIDS-Research and Palliative Care. 2020:373-80.
Pantazis N, Papastamopoulos V, Antoniadou A, Adamis G, Paparizos V, Metallidis S, et al. Changes in Body Mass Index after Initiation of Antiretroviral Treatment: Differences by Class of Core Drug. Viruses. 2022;14(8):1677.
Martínez‐Sanz J, Blanco JR, Muriel A, Pérez‐Elías MJ, Rubio‐Martín R, Berenguer J, et al. Weight changes after antiretroviral therapy initiation in CoRIS (Spain): a prospective multicentre cohort study. Journal of the International AIDS Society. 2021;24(5):e25732.
Olawepo JO, Pharr JR, Cross CL, Olakunde BO, Aniekwe C, Niyang M, et al. Changes in Body Mass Index Among Antiretroviral Therapy Naïve People Living with HIV in Southeastern Nigeria. Global Journal of Health Science. 2021;13(6):1-14.
Crum-Cianflone N, Roediger MP, Eberly L, Headd M, Marconi V, Ganesan A, et al. Increasing rates of obesity among HIV-infected persons during the HIV epidemic. Plos one. 2010;5(4):e10106.
MeleseChekole D, Gebrerufael G, Asfaw Z. Statistical Joint Modeling on Longitudinal Body Weight and CD4 Cell Progression with Survival Time-To-Death Predictors on HIV/AIDS Patients in Mekelle General Hospital, Ethiopia. 2020.
Huis't Veld D, Balestre E, Buyze J, Menten J, Jaquet A, Cooper DA, et al. Determinants of weight evolution among HIV-positive patients initiating antiretroviral treatment in low-resource settings. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2015;70(2):146-54.
Alema NM, Asgedom SW, Maru M, Berihun B, Gebrehiwet T, Atey TM, et al. Magnitude and predictors of first-line antiretroviral therapy regimen change among HIV infected adults: a retrospective cross sectional study. Annals of Medicine and Surgery. 2022;81:104303.
Birhan H, Seyoum A, Derebe K, Muche S, Wale M, Sisay S. Joint clinical and socio-demographic determinants of CD4 cell count and body weight in HIV/TB co-infected adult patients on HAART. Scientific African. 2022;18:e01396.
Reda AA, Biadgilign S, Deribew A, Gebre B, Deribe K. Predictors of change in CD4 lymphocyte count and weight among HIV infected patients on anti-retroviral treatment in Ethiopia: a retrospective longitudinal study. PLoS One. 2013;8(4):e58595.
Hasse B, Iff M, Ledergerber B, Calmy A, Schmid P, Hauser C, et al., editors. Obesity trends and body mass index changes after starting antiretroviral treatment: the Swiss HIV Cohort Study. Open forum infectious diseases; 2014: Oxford University Press.
Hema A, Poda A, Zoungrana J, Tougouma JB, Meda C, Kamoulé E, et al. Body mass index trends and overweight/obesity risk among patients initiating antiretroviral therapy in Bobo-Dioulasso Day Hospital, Burkina Faso: a cohort study from 2002 to 2019. 2022.
Thivalapill N, Simelane T, Mthethwa N, Dlamini S, Lukhele B, Okello V, et al. Transition to dolutegravir is associated with an increase in the rate of body mass index change in a cohort of virally suppressed adolescents. Clinical Infectious Diseases. 2021;73(3):e580-e6.
Esber AL, Chang D, Iroezindu M, Bahemana E, Kibuuka H, Owuoth J, et al. Weight gain during the dolutegravir transition in the African Cohort Study. Journal of the International AIDS Society. 2022;25(4):e25899.
Burns JE, Stirrup OT, Dunn D, Runcie-Unger I, Milinkovic A, Candfield S, et al. No overall change in the rate of weight gain after switching to an integrase-inhibitor in virologically suppressed adults with HIV. Aids. 2020;34(1):109-14.
Tegegne AS. Predictors associated with the variation of CD4 cell count and Body Mass Index (BMI) for HIV positive adults under ART. Scientific African. 2021;12:e00820.
Baraki AG, Gezie LD, Zeleke EG, Awoke T, Tsegaye AT. Body mass index variation over time and associated factors among HIV-positive adults on second-line art in north-west Ethiopia: a retrospective follow-up study. BMJ open. 2019;9(9).
Hirigo AT, Yilma D, Astatkie A, Debebe Z. Effect of dolutegravir-based first-line antiretroviral therapy on weight and body mass index among adult people living with HIV on follow up at health facilities in Hawassa city administration, Southern Ethiopia: a retrospective cohort study. Annals of Medicine. 2023;55(2):2242250.

Tables 1 to 4 are available in the Supplementary Files section

No competing interests reported.

Tables.docx

Download PDF

Editorial decision: Revision requested
11 Nov, 2024
Reviews received at journal
14 Oct, 2024
Reviewers agreed at journal
03 Oct, 2024
Reviews received at journal
23 Sep, 2024
Reviewers agreed at journal
09 Sep, 2024
Reviewers invited by journal
07 Sep, 2024
Editor invited by journal
01 Sep, 2024
Editor assigned by journal
20 Aug, 2024
Submission checks completed at journal
20 Aug, 2024
First submitted to journal
17 Aug, 2024

You are reading this latest preprint version

Body Mass Index Changes and Predictors among Adults Living with HIV/AIDS Who Are on Anti-Retroviral Therapy at Chiro General Hospital, Eastern Ethiopia: A facility-based retrospective cohort study

Status:

Version 1

Abstract

Figures

Background

Methods

Study Setting, Period and Design

Study population, inclusion criteria and exclusion criteria

Sample size determination and sampling technique

Data collection procedures

Data Quality Control

Data Processing and Analysis

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1