Body weight loss and associated factors among adults People living with HIV/AIDS on antiretroviral therapy in Shashemane referral Hospital, Oromia national regional state, Ethiopia: A facility based cross sectional Study

Background: HIV/AIDS and under nutrition commonly manifested as body weight loss are both highly prevalent in many parts of the world. Their effects are interrelated and act in a vicious cycle. Both HIV and under nutrition can independently cause progressive damage to the immune system and increased susceptibility to infection. Ethiopia is one of the countries affected by both epidemics, despite, study done on acute under nutrition among HIV/AIDS adults in particular were found inadequate. Thus the main aim of this study is to assess the magnitude of acute under nutrition (weight loss within 3 months) and associated factors among people living with HIV/AIDS adults on ART in Shashemane referral Hospital. Methods: Facility based cross-sectional study conducted in shashemane referral Hospital Oromia region, Ethiopia from August 3 to September 4.2016. Systematic random sampling technique employed and 402 study participants were included. Data was collected by interview patients using pre-tested structured questionnaire, review of patients register, weight and height measurements were taken. The data were entered into Epi Info version 3.5.1 and analyzed using version 20 SPSS statistical package and percentages of body weight loss were calculated. Logistic regression analyses were carried out to identify factors associated with body weight loss of > 5% within three months. Strength of association determined using P< 0.05 and odd ratio (95% CI). Result: The magnitude body weight loss > 5% were 15.9% (95% CI; 12.4-19.7). Low CD4 level of less than 200mm3/ml (AOR=8.41,95% CI:3.46-20.44), inability to consume high protein diets like meat, egg & per week (AOR=2.97, CI;1.39-6.35), low meal frequency of 2 or less per (AOR=3.09,95% CI: 1.25-7.68) and low average income of 1000 birr per month (AOR=9.07,95% CI: 3.71-22.14) significantly associated with Body weight loss of > 5% within 3 months with P<0.05. with HIV/AIDS adults on ART, specifically among advanced immune compromised, low income, consumes less meal frequently per day and fewer or no consumption of animal product.


Background of the study
More than 35.3 million people are living with human immunodeficiency virus/acquired immunodeficiency syndromes (HIV/AIDS) worldwide, out of which sub-Saharan Africa take the major share of 25 million. Among those HIV positive in the world, adult's accounts for 91% of the total [1], more over food shortages and under nutrition have combined with HIV/AIDS to bring some developing countries to the rim of crisis [2]. HIV/AIDS damage the immune system of an individual's over long period of time that lead to incidence of many opportunistic infections (OI). Thus, Opportunistic Infection place people living with HIV/AIDS (PLWHA) at a high risk of developing under nutrition [3]. In Ethiopia, 1.5 percent of the adults aged 15-49 are infected with HIV [4]. According to Antenatal care (ANC) sentinel surveillance, adjusted national HIV prevalence in 2005 was 3.5% with an estimated prevalence of 10.5% in urban areas and 1.9% in rural areas [5][6], Similarly the data of 2009 ANC based sentinel surveillance of adjusted national HIV prevalence also shows a point estimate of 2.3%; 5.3% in urban areas and 1.9% in rural areas [7]. Ethiopia is also one of the countries in Sub-Saharan Africa with the highest rates of under nutrition as a result of the interaction between poor diet and disease [8].
HIV/AIDS and under nutrition are both highly prevalent in many parts of the world, especially in sub-Saharan Africa and other developing countries. Their effects are interrelated and aggravate one another in a vicious cycle [9]. Both can independently cause progressive damage to the immune system of human being and increased susceptibility to infection. HIV/AIDS may lead to morbidity and mortality through OI that may manifest symptom like fever, diarrhea, loss of appetite, nutrient mal-absorption, and weight loss [2]. HIV specifically affects nutritional status by increasing energy requirements, reducing food intake, and adversely affecting nutrient absorption and metabolism. Asymptomatic and symptomatic adult HIV/AIDS clients have increased energy requirements by 15% and 30% respectively to maintain body weight and physical activity [10]. Under nutrition on the other hand, contributes to immune system impairment, making the body vulnerable to frequent illness and increasing its energy and nutrient demand, in this way it accelerate HIV/AIDS disease progression [2,10]. According to CDC the most severe form of under nutrition in HIV disease defined as HIV wasting syndrome (that include body weight loss) [11]. This complex interaction between HIV & under nutrition acts through the immune system functions of the body [12]. Optimal nutrition can boost the immune function, maximize the effectiveness of antiretroviral therapy (ART), reduce the risk of chronic illnesses, and contribute to a better overall quality of life [12][13].
Ethiopia is one of the countries hit hardest by the HIV epidemic in Africa [14] and a country with the highest incidences of under nutrition; the country has high levels of chronic food insecurity and is further prone to acute food insecurity, primarily during times of drought, environmental degradation, and insufficient access to and availability of food. According to the report of Ethiopia Demographic and Health Survey (EDHS) 2005 and the EDHS 2000, one in four women of reproductive age have chronic energy deficiency and 27% are anemic [15].
HIV/AIDS and under nutrition act synergistically to undermine the immunity of many Ethiopians [15] and has taken lives of millions even in the era of ART, thus HIV/AIDS could not be managed well since the dispute of under nutrition is unanswered [16].
Despite huge problem of HIV/AIDS and under nutrition in the country, there are few studies conducted at a country level in general and study done on acute under nutrition among people living with HIV/AIDS adults in particular were found inadequate. Therefore this study aims to assess magnitude of acute under nutrition (weight loss within 3 months) and associated factors among people living with HIV/AIDS adults on ART in Shashemane referral hospital, Oromia regional state, Ethiopia.

Study setting
The study conducted at Shashemane referral Hospital in Oromia Region. Shashemane is located 250kms south of Addis Ababa on the way to Hawassa and has an altitude of 1940 meters above sea level. The total population of the town in 2013 estimated to be 157,604 of which 78,014 and 79,590 were male and female respectively [Shashemane municipality 2014 report]. Staple food in are were teff with injera (injera with shiro wot) like most of Ethiopian towns. There are three health centers and two hospitals that are led by Shashemane town administration that provide HIV/AIDS care, of which Shashemane referral hospital has heavy case load and selected in this study by sample size connivance. The hospital at Kuyara serves around 2.

Study design, period and population
A facility based cross-sectional study design was employed from August 3 to September 4, 2016.
The source population was all adults >18 years who were on ART and the study population were those randomly selected adults on ART in Hospital during the study period that fulfils the inclusion criteria. Patients who were seriously/mentally ill, patients whose weight not taken three months prior to data collection and pregnant women were excluded during data collection.

Sample size determination
The required sample size was determined using single population proportion formula

= ( − )
Where n is the sample size, z is the standard normal score set at 1.96 (with assumption of 95% confidence interval), D is the desired degree of accuracy (maximum allowable error set at = 0.05) and p is the estimated proportion of the target population, P=61% (proportion of Undernourished people living HIV/AIDS with the use of BWL percentage of > 5%) [28] , the computed sample size were 366, and after 10 % non-response rate added the final total sample size became 403.

Sampling technique:
From the total 2199 HIV/AIDS clients who were currently on ART in Shashemane referral Hospital, 2008 were Adults > 18 years and 191 were children and adolescents < 18 years. In order to select 403 adults' participants from those currently on ART, the total adult clients currently on ART greater than or equal to 18 years was divided to the required sample size (2008/403=4.99) which is approximately 5.The first case was selected from the first five card of the first day of data collection by lottery methods by labeling the chart of the patients, and then every 5th adults by use of systematic random sampling selected on their coming order for follow up to ART clinic for interview, their registration and follow up card reviewed, height and weight measurement taken.

Data collection technique
Three month prior to data collection (in April 2016) training was given to all ART clinic staffs on how to appropriately measure weight and to take weight of all clients come to clinic for their routine follow up, so that this weight used as baseline to calculate Body Weight Loss percentage.
During this time weight scale has been checked for accuracy and has been followed for one month. Then three months later (in August 2016) training were given to two data collectors and supervisor on how to collect data using prepared pretested questionnaires and how to measure weight and height. Before measuring weight and height of the clients during data collection, standardization were done to eliminate the individual's variability in the following ways, one clients was selected, weight and height was measured by two of data collectors two times each, then principal investigator have taken the weight of the same person with the same adult height and weight scale, then first the variation of two data collectors were evaluated (precision).
Second their difference from principal investigator weight and height value was compared (accuracy). Then based on the result (deviation) discussion and reorientation were provided.
Then data were collected from August 3 to September 4 /2016 on working days for one month in hospital using pretested structured questionnaire together with weight and height measurements.
Two data collectors outside ART clinic staffs and one supervisor were recruited and two days training was given. The data collection process was closely supervised daily by the supervisor and principal investigator. Weight of the clients was measured using adult weighting scale calibrated using standard 2kg tool every morning and before and after each measurement.
Participant's weight measured after removing heavy clothes and recorded to the nearest 0.1kg.
Height of the clients measured after removing shoes and anything on head and recorded as 0.1cm

Data management and analysis
Data were entered into Epi Info version 3.5.1, and then exported to SPSS version 20 statistical packages for recoding and analysis. During analysis the frequency of different variable was determined and their association with outcome variables (weight loss >5%) was examined using bivariate and multivariable logistic regression analysis techniques. Only variables that have shown P < 0.25 in bivariate analysis were included in multivariate analysis.
For bivariate and multivariable analysis, the outcome variable, weight loss >5% was coded as "1" and no weight loss > 5% was coded as "0. Odd ratio along with 95% confidence interval and P-value 0.05 were used to measure the strength of association and level of statistical significance respectively. Model of fitness were checked by use of omnibus in that all value shows (p<0.05) and hosmer and lemeshow (p>0.05) indicating that the model is doing well.  [18,19] Acute under nutrition; in adults of 18 years and above defined by use of percentage of recent body weight loss (BWL) greater than 5%, with the period of two to four months [18 ,19 ].

HIV/AIDS clients on ART; Those HIV positive confirmed by antibody test & currently taking
ARV drugs in the facility.

Magnitude of body weight change among adult's people living with HIV/AIDS on ART in Shashemane Referral Hospital
The proportion of participants who lost any amounts of weight were 176 (43.6%), However the

Results of bivariate analyses on medical and antiretroviral related factors associated with Body weight loss > 5% within 3 months (Acute under nutrition).
Among those Body weight loss >5% (undernourished) study participants, 22(34.4%) had CD4 level of less than 200mm3/ml, 7(10.9 %) were in treatment stage other than stage I and 6(9.4%) had opportunistic infection in the last three months prior to data collection. In bivariate analysis those variables have shown significant (P <0.05) association.

Results of bivariate analyses on diet related factors associated with Body weight loss > 5 % (Acute under nutrition)
Among those body weight loss > 5 %( undernourished) study participants 50(78.1%) usually eat injera with shiro wot, 29 (45.3%) have no dietary counseling after HIV was confirmed, 52(81.2%) did not get meat / egg or fish per week, and 54 (84.4%) had eaten meals three or less times per day in the last three months. As shown in Table 6, among diet related factors meal frequency, dietary counseling and consumption of meat/ fish /egg per week in the last three months showed significant (P <0.05) association, while the other variable like most usually eaten food and change in feeding style showed weak association(p< 0.25).

Factors associated with body weight loss > 5 %( under nutrition) among people living with HIV/AIDS adults on ART
Bivariate logistic regression analyses of independent variables: average income per month, CD4 level, meal frequency, high protein diet (meat/egg/fish) intake per week, opportunistic infection in the last three months and treatment stage other than WHO treatment stage I(one) have shown significant (p< 0.05) association with body weight loss of greater than 5%. on the other hand previous diet counseling, occupation, most usually eaten food and change eating style after HIV results confirmed have shown weak (p< 0.25) association with BWL >5%(acute under nutrition).
However, during multivariable logistic regression analyses after controlling for all possible confounding, the variables that retained statistical significance (P<0.05) were CD4 level, consumption of high protein diet like meat/ egg /fish per week, ,average income and meal frequency per day in the last three months prior to data collection. The odds (95% CI) of becoming body weight loss > 5%(undernourished) were 8.41(3.46-20.44) times higher for the individuals who had CD4 level less than 200mm3/ml as compared to those who had CD4 level more than 500mm3/ml in the last three months. The odds (95% CI) of becoming body weight loss > 5%(undernourished) were 2.97(1.39-6.35) times higher for those who were unable to consume high protein diet like meat or egg or fish per week as compared to those who were able to consume those diets per weeks at least once. The odds (95% CI) of becoming body weight loss > 5%(undernourished) were 9.07 (3.71-22.14) higher for those who earn less than 1000birr as compared to those who earn more than 1000 birr per Month and the odds (95% CI) of becoming body weight loss > 5%(undernourished) were 3.09 (1.25-7.68) higher for those who reported to have less meal frequency (2 or less times) per day as compared to those who get high meal frequency (more than three times) per day in the last three months prior to data collection.  On the other hand the prevalence rate of acute under nutrition (body weight loss greater than 5%) observed in this study is lower than the study conducted among HIV seropositive subjects in one of the AIDS clinic in Paris (37.9%) [23]. This may be due to Paris study include pre-ART patients where the complication like eating problems that contribute to body weight loss are more prevalent. Moreover the discrepancy could also be due to the longer period of time the patients followed (six months) and smaller sample size (n=124) in Paris study. The prevalence of acute under nutrition reported (BWLP >5%) were 60.9% in the study conducted at University of Gondar referral Hospital is much higher than the results observed in this study [28]. This could be due to methodological difference. The Gondar study used usual body weight as baseline (usually body weight before HIV status was known/ reported by clients), while in this study body weight taken/recorded by health care worker in ART clinic three months ago was used as a baseline. Furthermore most of the study participants were on ART for less than six months in the study done at University of Gondar referral Hospital, whereas this study includes over 80% of participants who are on ART for more than two years. The finding related to the strong association of meal frequency to body weight loss observed in this study is supported by the finding reported in the study conducted on under nutrition based on Body Mass Index and associated factors in health facilities at Hosanna Town [34].
This finding were also similar with study conducted in Nigeria on nutritional status of HIVpositive individuals on free HAART treatment in developing nation indicated that low socioeconomic status was associated with under nutrition [24]. The issue of low income could be related with some other points which may include no access to good quality and quantity of food that has direct relation with weight loss (under nutrition).No consumption of meat/egg/fish at least once per week has shown significant (p<0.05) association with acute under nutrition within three months. As there are no studies found regarding the association of consumption of these protein rich foods with body weight loss comparisons have been difficult. However, these food contain quality protein that are required for maintenance and growth of the body tissue, it is apparent that lack of these foods in the diet of PLWHA on ART significantly contribute to body weight loss. The other variable like Opportunistic infection 16 (4%), eating problem 23 (5.7 %), nutritional support 23 (5.7 %) & HIV related symptom 18 (4.5) which were significant in other study were not obtain enough sample size to show significant association with body weight loss greater than 5% in the three months period (acute under nutrition) in this study. This may be due to that most of the clients participated in this study (> 80%) were on antiretroviral therapy (ART) for more than 2 years, whereas most of the previous study conducted include early ART initiation and those not on ART.

Limitation of the study:
This study did not consider the seasonal variation of food access in area and may not represent the whole year nutritional status of the population. Lack of adequate similar studies in our country to make comparative discussion difficulty and the study didn't use qualitative method for detail investigation.

Conclusion
Body weight loss (Under nutrition) is still huge problem among adults people living with HIV/AIDS on ART particularly among those advanced immune compromised(CD4 < 200), low income family per month, consume less frequently meal per day and no animal products diet per week.