Prevalence and correlates of prescription drug diversion and misuse among people living with HIV in the eThekwini district, KwaZulu-Natal, South Africa.

BACKGROUND: Prescription drug diversion, and misuse has increased over the past decade and is notably in high-income-countries and significantly contributes to the opioid epidemic. People living with HIV (PLWH) are particularly vulnerable to prescription drug diversion, and misuse as most experience chronic pain, mental health problems and HIV-related illnesses. The researchers investigated the prevalence and correlates of prescription drug diversion, and misuse among PLWH in the eThekwini district, KwaZulu-Natal.
METHODS: A cross-sectional study was conducted among 392 PLWH, conveniently recruited from the public healthcare facilities located in rural, semi-urban and urban areas of the eThekwini district. Participants answered questions about their background, prescription medications, substance use, and prescription drug diversion, and misuse. Descriptive analysis was performed to estimate the prevalence of prescription drug diversion, and misuse. Multivariable logistic regression was used to identify predictors of prescription drug diversion, and misuse.
FINDINGS: Overall, 13% of the participants reported lifetime prescription drug diversion. The most common type of diversion was using prescription medication not prescribed by a healthcare provider (11%), followed by sharing of prescription medication (9%) and buying prescription medication without a medical script (5%). Twenty-three per cent of the participants reported prescription drug misuse in the past 90 days, with using prescription medication without a healthcare providers' guidance (9%) and not following the scheduled time periods (8%) being the most common reported types of misuse. Self-medicating was identified as a risk factor for prescription drug misuse. There was no association between ART adherence and prescription drug diversion, and misuse.
CONCLUSION: The study findings contribute to improving the limited data available on prescription drug diversion, and misuse among PLWH in South Africa. The prevalence underscores a need for urgent interventions when prescribing medications with potential risks. Addressing the risk of self-medicating is imperative for HIV care outcomes and to avert death.

Introduction

For the purpose of this study, the following terms were defined as

Prescription drugs

Controlled medicines that would legally require a medical prescription from a healthcare provider in order to be dispensed due to their potential risks of being diverted, misused, abused and addiction.

Drug

Refers to any substance which has a physiological effect or the potential to prevent or cure a disease. In this study it will be used interchangeably with medicines or medications.

Drug diversion

Acquiring or distributing any legally prescribed controlled medicinal drugs to/from another person for any illicit use. Literature also reports drug diversion or medication diversion interchangeably alongside misuse, abuse, recreational use, non-medical use, illicit use, selling, sharing, giving-away, trading, stealing, missing/losing, doctor shopping and pharmacy shopping [1].

Misuse

The inappropriate use of medicine or not adhering to the information leaflet of the medicine. This includes the following: without a health care provider’s guidance or instructions, non-medically (e.g. used it recreationally), without following the dosage instructions, not following the correct way of administration of the medication, not following the scheduled time periods to further reduce the amount of pain you were feeling, to increase the ability to concentrate, to boost feelings and performance enhancing, to experience pleasure or excitement and intense feelings of well-being and happiness and to get high [2]. However, according to the National Institute on Drug Abuse, medication misuse is defined as use of medication in ways or amounts other than intended by a doctor, by someone other than for whom the medication is prescribed, or for the experience or feeling the medication causes. This term is used interchangeably with “non-medical” use, a term employed by many national drug use surveys [3].

Self-medication

The use of medicinal drugs without the guidance of a health care provider [3].

Global efforts are being made to end the AIDS epidemic, but countries have reported mixed results [4]. In 2018 about 37.9 million people were living with HIV worldwide, with 1.7 million new HIV infections and 770 000 deaths [5]. Approximately 23.3 million people living with HIV (PLWH) were receiving antiretroviral therapy, but only 53% were virally suppressed [5]. Since the universal test and treat, the 90-90-90 target, as well as the UNIADS fast track strategy, only about 57% AIDS event and death reduction has been achieved with early ART initiation [6]. South Africa (SA) accounts for 19% of the global HIV prevalence rate, [5] and the KwaZulu-Natal province has the highest HIV prevalence within the country [7]. Barriers to achieving and maintaining viral suppression have been documented, one of which is prescription drug diversion, and misuse [1,8]. One of the dangers of prescription drug diversion is transitioning into illicit drug use, worsening the barriers of HIV care and challenging the HIV/AIDS targets to put a stop to the HIV epidemic. A recent scoping review study revealed that prescription drug diversion among PLWH came into existence as early as 1995; however, there is a lack of evidence in resource-limited-settings with high HIV rates [9].

PLWH often experience illnesses that correlate with prescription drug diversion, and misuse. Evidence suggests that chronic pain is the common comorbidity among PLWH that correlates with prescription drug misuse [10,11]. Literature has also shown substance use disorders or dependence as risk factors for prescription drug diversion among PLWH [8,10,12-21]. Prescription drug diversion, and misuse have been associated with mental health problems such as depression symptoms, psychiatric disorder and anxiety as risk factors among PLWH [16,18,20,22-24]. Other factors associated with prescription drug diversion, and misuse among PLWH included being male [16,20], young in age [14,22], HIV-related stigma [16], poor in health [18], lower in the level of education [19,24], unemployed or having no income [25], having an experience of diverting, and misusing medicines [15,26], owning addictive prescriptions [10,24,26], and living in high-level neighbourhood disorder [19]. In the United States, these factors were reported as being associated with new HIV infections [27]. A recent scoping review study showed that there is limited research on prescription drug diversion in low-income-countries among PLWH [28].

Research investigating medication misuse among the general population can reveal and guide how prescription drug diversion, and misuse might unfold among PLWH in our local context. Our local context is surrounded by several challenges related to consumer, health care providers and drug treatment centres. Challenges related to the user includes: codeine misusers did not know of their dependency and regarded their use as appropriate [29]. Codeine misusers were also unaware of the addictive potential of codeine, including the risk of dependence and harm [29]. Another challenge related to the user was the ability to read and understand the medicine information leaflet [30]. About 29% of people accessing medicines from the SA retail pharmacies reported not reading the medicine information leaflet [30]. Most people perceived codeine-containing products as safe medications, while only a few people acknowledged the harmfulness and addictiveness of these medicines [30]. Health care providers also have their share of challenges. A study conducted in SA showed that about 95% of health care providers prescribing codeine products raised a concern that patients do not understand the risks of medicine dependence since they perceived it as safe [31].

Furthermore, health care providers perceive that patients were inadequately given information on medicine usage [31]. Another challenge faced by health care providers was the difficulty to identify problematic medicine use without the patient disclosing of their addiction problem [31]. A study by Carney et al. (2018), reported that pharmacy staff were challenged by lack of specialist training in substance use issues, therefore, making it challenging to curb medicine misuse [32]. Dada et al. (2015) study among the South African Community Epidemiology Network on Drug Use (SACENDU) identified a gap that codeine might be used as a substitute for heroin in SA when the heroin is unavailable, therefore worsening the problem of drug diversion [33]. On the other hand, drug treatment centres in SA are mostly located in urban areas which lack tracking of codeine misuse and dependence [34]. Additionally, a qualitative study in SA revealed that ARVs were crushed and added to illicit drugs and smoked for their psychoactive effects [35].

Data on the extent of prescription drug diversion, and misuse among PLWH are very limited, despite the emerging public health problems and the high HIV prevalence in the KwaZulu-Natal province. This study aims to investigate the prevalence and correlates of prescription drug diversion, and misuse and among PLWH in the eThekwini district, KwaZulu-Natal, as well as the association between ART adherence and prescription drug diversion, and misuse. Knowing the extent and correlates associated with prescription drug diversion and misuse could inform healthcare providers and policymakers to incorporate appropriate and relevant interventions to address problems when managing HIV.

Methods

Study design

This study comprised of a cross-sectional survey conducted in the eThekwini district, KwaZulu-Natal province, South Africa, and forms part of a sequential, explanatory mixed-method approach involving a cross-sectional survey and in-depth interviews. The researchers surveyed to investigate the extent of prescription drug diversion, and misuse among PLWH in the eThekwini district from August 2018 to May 2019. An interviewer-administered questionnaire was used to collect data from a convenient sample of PLWH who, at the time of the survey, were at the recruitment site.

Study setting

South Africa is situated on the African continent and made up of nine provinces. The entire study was conducted in the eThekwini district, KwaZulu-Natal province, located in the southeast on the country. eThekwini is one of eleven districts in KwaZulu-Natal province. According to Statistics SA, the population size was estimated at 3.4 million in 2011 [28]. The majority of its people speak isiZulu (62%), and 26% are English language speaking. About 73% of the residents are Black African, 16% Indian/Asian, 6% White and 2.5% Coloured. The Fifth South African National HIV Prevalence, Incidence, Behaviour and Communication Survey (SABSSM V) reported that the HIV prevalence in KwaZulu-Natal was the highest of all provinces at 27.0% (95% CI 23.9–30.4) in 2017. Two of the key challenges to eThekwini district’s health care service provision is the high prevalence of HIV/AIDS and also the abuse of drugs and alcohol [36]. Until now, the extent of prescription drug diversion, and misuse in the eThekwini district is, however unknown.

Sampling method

The researchers employed a stratified sampling method for sampling public healthcare clinics as recruitment sites for PLWH. The researchers grouped all eThekwini district public healthcare clinics providing HIV care programme into strata based on geographic settings, for example, rural, urban, and semi-urban. Overall, the eThekwini district has 118 public healthcare facilities, of which 6% are in rural areas, 53% in urban areas and 41% in semi-urban areas. Stratified randomisation of public healthcare clinics in the three strata was ensured that the proportions of the sample matched those from all three geographical settings. Permission to conduct the survey at each clinic was granted through relevant authorities and clinics that were unable to give permission to participate in the study were excluded. The researchers conveniently approached and invited people who were waiting for healthcare worker consultations at the clinics to participate in the study. To reach people who were drug users and living with HIV, we employed an outreach and recruitment approach through a Non-Governmental Organisation (NGO). The NGO had an existing and on-going outreach programme which supported drug users around eThekwini. They had programmes in place such as HIV testing, linkage to care and harm reduction interventions (including the provision of clean syringes and needles and opioid substitution therapy), and permission was obtained to join their outreach team in order to recruit drug users who might be interested in participating in our study.

Participants

Three hundred and ninety-two people who verbally confirmed to be living with HIV and were aged 18 years or older, gave written consent. Some participants were assisted in their preferred language during consenting and conducting the survey. All participants were in the eThekwini district at the time of the survey.

Data collection and instruments

The researchers adapted the cross-sectional survey tool from a study conducted by Tsuyuki and Surratt (2015) [37] to fit our population, setting and study aim. We used hard copies of the survey which consisted of ten categories: background, prescription medications (PM), substance use, prescription drug diversion, non-diversion, prescription drug misuse, impressions of PM black market, antiretroviral therapy adherence, and stress, mental and emotional health and stigma. Data were collected by the researcher and a trained research assistant. A pilot test study, to ensure the validity of the tool among PLWH, was conducted before administering the questionnaire and helped us to improve precision, reliability, validity of data, identify problems/omissions, and assess time spent when completing the survey. The survey was conducted in English; however, questions were clarified to the participants in their preferred language. The survey lasted 60 minutes or less.

Ethical considerations

Full ethical clearance from the University of KwaZulu-Natal’s Biomedical Research Ethics Committee (Reference No. BE666/17), the KwaZulu-Natal Provincial Department of Health (Reference No. HRKM055/18, KZ_201802_026) and recruitment site clearances were obtained before commencing our data collection. Questionnaires were identified by study ID (identity) numbers and not linked to consent forms. The dynamics of each clinic was obtained from the facility manager and the nurses responsible for HIV management before conducting the survey to minimise harm and stigma when inviting potential participants to participate in the study. A private space was allocated to maintain confidentiality during data collection, data collection date was set, a tour around the facility and introduction to other healthcare providers were all done in preparation for the data collection. On the day, the facility manager or the nurse in charge made a brief introduction to the patients in the waiting area before data collection commenced. During the consenting process, researchers explained extensively to each participant that their participation was voluntary, that the study was not associated with the clinic, and participation would not affect their healthcare. Participants were assured that their responses would be identified with study ID numbers, not their names.

Measures

Outcomes

This study focused on investigating the prevalence of and factors influencing prescription drug diversion, and misuse among PLWH. The researchers used the Gelberg–Andersen Behavioural Model to guide the development of the survey questions and operationalization in this study. This conceptual framework has been fully studied for healthcare utilisation for vulnerable populations [38].

Our analysis examined three outcome measures: The first outcome was prescription drug diversion which stemmed from the question: “Have you ever diverted (sold, borrowed, shared, gave-away, visited multiple doctors/pharmacies for the same medications or traded or stole) prescription medication (yes/no)?” Prescription drug diversion was then recoded to “have you ever diverted prescription drugs in your lifetime (yes/no)?”.

The second outcome was prescription drug misuse which stemmed from several questions investigating different types of drug misuse: “In the past 90 days, have you ever used prescription medication (PM): without a healthcare provider guidance or instructions, not following the scheduled time periods, non-medically, to further reduce pain, without following the dosage instructions, not following the correct way of administration, to enhance performance, for pleasure or to get high?” Prescription drug misuse was recoded to “in the past 90 days, have you ever misused prescription drugs (yes/no)?”.

Additionally, our questionnaire incorporated the AIDS Clinical Trials Group (ACTG) adherence questions to measure per cent ART adherence. Per cent ART adherence was measured using the self-reported data on ART adherence in the past seven days, which stemmed from:

How many pills (ARVs) are you prescribed to take each time?

◦ Number of pills per day

◦ Number of pills per time

How many doses did you miss in the past seven days?

The per cent ART adherence was calculated by the total weekly ARV doses taken divided by the total ARV doses that were prescribed, converted to a percentage

Demographics assessed whether age, gender, marital status, level of education, employment, self-rated health, neighbourhood one resides at and homelessness had an influence on prescription drug diversion or misuse. Disorderly neighbourhoods and homelessness were hypothesized to add vulnerability to individuals which then influences prescription drug diversion or misuse.

Self-medicating for conditions that one was diagnosed with (coded as yes/no) assessed whether self-medicating had any influence or facilitated prescription drug diversion or misuse.

Owing prescription drugs or being in possession of prescription drugs (coded as yes/no) assessed whether owing or being in possession with prescription drugs had any influence or facilitated prescription drug diversion or misuse.

Being educated or trained on how to use or take any prescription medication (coded as yes/no) assessed whether education or training acted as a protective factor from prescription drug diversion or misuse.

Being knowledgeable that prescription drugs are harmful to one’s health if not used as prescribed assessed whether knowing the risk of prescription drugs would protect a person from diverting or misusing the medication.

Substance use (included alcohol, marijuana, tobacco and illegal drugs) which stemmed from a question “when was the last time that you used alcohol/marijuana/tobacco/illegal drugs?” This was then recoded to, “have you ever used alcohol/marijuana/tobacco/illegal drugs in your lifetime (yes/no)?”. Substance used is hypothesized to add vulnerability to individuals which then influenced prescription drug diversion or misuse.

Data management and analysis

Data were captured into a password protected excel spreadsheet with study ID numbers. The accuracy of source data was ensured through double capturing, with the researcher as the second data capturer. After data cleaning and verification, data were imported into The R Project for Statistical Computing R version 3.5.1 (2018-07-02) [39] for analysis. All source documents were stored in a secured place as back up.

Outcome variables were coded as follows: 1 = ever diverted prescription drugs vs 0 = never diverted prescription drugs; 1 = misused prescription drugs in the last 90 days vs 0 = never misused prescription drugs in the past 90 days.

Continuous variables were summarised using means with standard deviation and compared across the two outcomes variables (diversion and misuse), using either independent samples t-test or Wilcoxon rank-sum test. Categorical variables were reported as frequencies with percentages and compared using Fisher’s exact test. The prevalence with exact 95% confidence interval was calculated for prescription drug diversion, and misuse among PLWH in the eThekwini district.

Two types of multivariable logistic regression models were performed to identify variables that were associated with diversion and misuse. The association of each variable on the outcome (diversion or misuse) controlling for the geographical setting was assessed for model A. Model A represent the odds ratios showing the association of each variable on diversion or misuse among PLWH in the eThekwini district and detailed information presented in the S2 Table. All variables were fitted in the model, including geographical setting, to assess their collective association on the outcome for model B. Model B represents adjusted odds ratios showing the association of a group of variables on diversion or misuse among PLWH in the eThekwini district and detailed information presented in the S3 Table. Since the study had too many variables to use in model B, age and gender were selected as these variables are known to be confounders of many exposure-outcome relationships [40,41]. All variables that were plausible enough to be associated with each of the outcome variables were then included in model B. We regarded this strategy as well thought out and scientifically rigorous, as opposed to using p-values to reduce the number of variables.

The following variables were included in model B for prescription drug diversion outcome: age, gender, education, income, substance use (tobacco and illegal drugs), and self-medicating. The variables included in model B for prescription drug misuse were age, gender, homelessness, substance use (tobacco and illegal drugs), self-medicating and awareness of the black market.

The percentage of ARV adherence was calculated by the total weekly ARV doses taken divided by the total ARV doses prescribed, converted to a percentage. The analysis was conducted to assess the association between the percentage of ARV adherence with prescription drug diversion or misuse.

All analyses were performed using R studio, and Microsoft Excel was used to plot graphs and forest plots.

Justification for not performing ad hoc corrections to correct for multiple comparisons

Our analyses were exploratory, and there was no a priori hypothesis being tested. We were of the view that either committing type 1 error (false positive) or type 2 error (false negative) in our analyses is not critical. Our reasoning was also based on the premise that we are not testing any treatment nor performing any analyses that can potentially change guidelines, so that is why we did not correct for multiple comparisons.

Results

Characteristics of study participants

The study received a 100% response rate since all persons approached to participate agreed and completed the survey. Completed responses were received from 392 HIV positive participants, recruited from three geographic settings: rural setting 6% (n = 25), urban setting 58% (n = 226) and semi-urban setting 36% (n = 141). The demographics of the participants were females 67% (n = 263), males 33% (n = 129), average age 36.78 (± 10.0) years, African 99% (n = 388), single 86% (n = 335) and reported being homeless in the past 30 days 4% (n = 17) (Tables 1 and S1 in the supplementary material file). Eighty-two per cent of participants were on antiretroviral treatment, with an average of 96% of ART adherence in the past seven days, 61% (n = 238) have not completed high school, and 53% (n = 209) were employed.

Table 1

Participants characteristics stratified by prescription drug diversion and misuse.

Characteristics	Total N = 392	Ever diverted prescription drugs, N = 51	Did not divert prescription drugs, N = 341	p value	Ever misused prescription drugs, N = 89	Did not misuse prescription drugs, N = 303	p value
Age in years, mean ±SD	36.8 ± 10.0	37.1 ± 8.5	36.7 ± 10.2	0.807	35.4 ± 8.6	37.2 ± 10.4	0.103
Age groups, n (%)
18–24 years	27 (7)	4 (8)	23 (7)	0.2487	7 (8)	20 (7)	0.8149
25–29 years	79 (20)	6 (12)	73 (21)		19 (21)	60 (20)
30 + years	286 (73)	41 (80)	245 (72)		63 (71)	223 (73)
Gender, n (%)
Male	129 (33)	24 (47)	105 (31)	0.032	29 (33)	100 (33)	1.000
Female	263 (67)	27 (53)	236 (69)		60 (67)	203 (67)
Education groups, n (%)
Not completed high school	238 (61)	34 (67)	204 (60)	0.6305	55 (62)	183 (60)	0.694
Completed only high school	114 (29)	13 (25)	101 (30)		25 (28)	89 (29)
Completed both high school and tertiary studies	40 (10)	4 (8)	36 (10)		9 (10)	31 (10)
Employment status, n (%)
Employed	209 (53)	25 (49)	184 (54)	0.549	44 (49)	139 (46)	0.629
Unemployed	183 (47)	26 (51)	157 (46)		45 (51)	164 (54)
Participants who received income past month, n (%) a
No	90 (23)	18 (35)	72 (21)	0.041	20 (22)	70 (23)	0.991
Yes	300 (77)	33 (65)	267 (78)		69 (78)	231 (76)
Homeless, n (%)
No	374 (95)	48 (94)	326 (96)	0.716	82 (92)	292 (96)	0.144
Yes	18 (5)	3 (6)	15 (4)		7 (8)	11 (4)
Substance use, n (%)
Ever used alcohol b
No	205 (52)	23 (45)	182 (53)	0.410	49 (55)	156 (51)	0.567
Yes	186 (48)	27 (53)	159 (47)		39 (44)	147 (49)
Ever used tobacco c
No	285 (73)	27 (53)	258 (76)	0.001	61 (68)	224 (74)	0.360
Yes	106 (27)	24 (47)	83 (24)		28 (32)	78 (26)
Ever used marijuana d
No	335 (86)	39 (76)	296 (87)	0.063	72 (81)	263 (87)	0.171
Yes	55 (14)	12 (24)	43 (13)		17 (19)	38 (13)
Ever used illegal drugs
No	371 (95)	45 (88)	326	0.042	79 (89)	292 (96)	0.012
Yes	21 (5)	6 (12)	15 (4)		10 (11)	11 (4)
Perceived misusing or abusing of prescription medication not harmful their health, n (%) e	66 (17)	8 (16)	58 (17)	0.495	13 (15)	53 (18)	0.725
% ART adherence, mean ±SD f	96.4 ± 12.9	94.5 ± 16.0	96.7 ± 12.	0.366	96.0 ± 11	96.5 ± 13.3	0.705
Diagnosed with other conditions, n (%)
No	274 (70)	36 (71)	238 (70)	1.000	64 (72)	210 (69)	0.7344
Yes	118 (30)	15 (29)	103 (30)		25 (28)	93 (31)
Self-medicated for diagnosed conditions, n (%)	61 (16)	10 (20)	51 (15)	0.517	25 (28)	36 (12)	< 0.001
Awareness of black market, n (%)	94 (24)	13 (25)	81 (24)	0.933	26 (29)	68 (22)	0.247
Geographical setting, n (%)
Rural	25 (6)	4 (8)	21 (6)	0.132	8 (9)	17 (6)	0.218
Urban	226 (58)	35 (69)	191 (56)		55 (62)	171 (56)
Semi-urban	141 (36)	12 (24)	129 (38)		26 (29)	115 (38)

a n = 2: Missing data,

b n = 1:Missing data,

c n = 1: Missing data,

d n = 2: Missing data,

e n = 3: Not sure,

f n = 12: Missing data.

Rates of prescription drug diversion and its predictors

Overall, 13% (n = 51) of the sample reported ever diverting prescription drugs (Fig 1). About 11% (n = 42), 9% (n = 36), and 5% (n = 20) of the participants reported using PMs not prescribed by a healthcare provider, sharing PMs, and to have bought PM from someone else or from a pharmacy without a medical script for their own use in the past 90 days, respectively. The least type of diversions reported, at 2% (n = 9) or less, were giving away, borrowing, and trading of PM.

Fig 1

The prevalence of diversion and types of prescription drug diversion, with a 95% confidence interval reported by people living with HIV in the eThekwini district.

Correlations between characteristics of participants who have diverted versus those who have never diverted prescription drugs are shown in Table 1. In the multivariable model A (Fig 2) gender, income and tobacco use were associated with prescription drug diversion. Males had 2-fold increased odds of diverting PM compared with females [OR = 2.01 (95% CI: 1.10–3.65), p = 0.022]. Participants who had not received an income in the past month had 2-fold increased odds of diverting when compared to those who have received income in the past month [OR = 2.02 (95% CI: 1.08–3.80), p = 0.029]. Participants who used tobacco had an almost 3-fold higher chance of diverting PM compared to non-substance users [OR = 2.74 (95% CI: 1.50–5.02), p = 0.001]. Participants who used marijuana and illegal drugs had 2-fold higher chance for diverting PM, [OR = 2.05 (95% CI: 0.99–4.23), p = 0.052] and [OR = 2.69 (95% CI: 0.98–7.38), p = 0.054].

Fig 2

Odds ratios showing the association of each variable on diversion among PLWH in the eThekwini district.

Model A represent the odds ratios showing the association of each variable on diversion among PLWH in the eThekwini district. All variables were fitted in the model, including geographical setting, to assess their collective association on the outcome for Model B. Model B represents adjusted odds ratios showing the association of a group of variables on diversion among PLWH in the eThekwini district.

In the multivariable model B, there was no association in participants with no income, have used tobacco, used illegal drugs, and have self-medicated (S2 Table).

Rates of prescription drug misuse and its correlates

Overall, 23% (n = 89) of the sample reported having misused prescription drugs in the past 90 days (Fig 3). Among all the participants in the study, 9% (n = 35) and 8% (n = 33), reported to have used PM without a healthcare provider’s guidance and not following the scheduled time periods in the past 90 days, respectively. In addition, among the participants who misused drugs, 6% (n = 25), 5% (n = 18), 4% (n = 14), reported to have used prescription drugs non-medically, to further reduce the amount of pain, and without following the dosage instructions in the past 90 days. The least reported types of misuse, each reported at 1% (n = 5), were not following the correct ways of administering the medication to increase the ability to concentrate and boost feelings; and performance-enhancing to experience pleasure and to get high.

Fig 3

The prevalence of misuse and the types of prescription drug misuse with a 95% confidence interval reported by people living with HIV in the eThekwini district.

The characteristics of participants who have misused versus those who have never misused prescription drugs were compared (Table 1). Participants who used illegal drugs had more than 3-fold higher odds of misusing PM [OR = 3.37 (95% CI: 1.37–8.30), p = 0.008] when compared to those who do not use illegal drugs. Participants who were self-medicating for diagnosed conditions had almost 3-fold higher odds [OR = 2.91 (95% CI: 1.62–5.21), p <0.001] of misusing PM (Fig 4).

Fig 4

Odds ratios showing the effect of each variable on misuse among PLWH in the eThekwini district.

Model A represent the odds ratios showing the association of each variable on misuse among PLWH in the eThekwini district. All variables were fitted in the model, including geographical setting, to assess their collective association on the outcome for Model B. Model B represents adjusted odds ratios showing the association of a group of variables on misuse among PLWH in the eThekwini district.

Results from model B showed that self-medicating increased the odds of misusing PM [OR = 2.63 (95% CI: 1.44–4.82), p = 0.002]. Participants with the following characteristics: being homeless, tobacco use, illegal drug use and being aware of the black market, had no association with PM misuse (S3 Table).

Association between ART adherence with prescription drug diversion and misuse

The reported ART adherence levels were greater than 95%, which is required to obtain a successful ART outcome. This was, however, not statistically significant as participants who were diverting PM had lower ART adherence (96%) when compared with participants who never diverted PM (97%) in the past seven days. Similarly, participants who were misusing PM had lower ART adherence (96%) when compared with participants who never misused PM (97%) in the past seven days. No regression analyses were performed since the ART adherence of all groups were relatively similar.

Discussion

The researchers investigated the prevalence and correlates of prescription drug diversion, and misuse in a cohort of HIV-infected adults in the eThekwini district, KwaZulu-Natal province. The prevalence of prescription drug diversion among PLWH was 13% during their lifetime. Twenty three per cent of participants reported prescription drug misuse in the past 90 days. Self-medicating was the only statistically significant predictor for prescription drug misuse among PLWH.

Although there is limited evidence on prescription drug diversion in resource-limited settings, numerous studies have been conducted in the United States where higher rates of prescription drug diversion have been reported among PLWH when compared to our findings [9]. A clinical trial conducted in the United States among men who have sex with men living with HIV on treatment documented that 27% of the participants reported to have ever sold their medication and 19% had diverted their treatment in the past year [17]. Prescription drug misuse is a silently growing problem globally, and literature has documented a wide prevalence spectrum, but our prevalence of prescription drug misuse is within the reported spectrum. Two longitudinal studies conducted in the United States among PLWH reported the prevalence of 21% opioid misuse in the past 90 days [42], and 37% abnormal use of prescription opioids in the past 90 days [43]. Other studies in the United States, however, reported lower prevalence when compared to our study: prevalence of 19% with a history of prescription drug misuse was reported among HIV-infected individuals in the US military [44] and a cross-sectional study in the United States among PLWH reported a prevalence of 11% of non-medical use of prescription opioids in the past year [45].

The prevalence of prescription drug diversion and misuse among PLWH is higher when compared with the general population. In support of this, according to the 2016 South African Demographic and Health Survey about 14% (women) and 15% (men) aged 15 years and older reported the use of codeine-containing medicines in the previous year however among these group, 2% of the women and 2% of the men reported non-medical use of codeine-containing products [46]. A recent study in South Africa reported 6% non-medical use of codeine-containing mixtures among the general population [30]. Results from the 2013 National Survey on Drug Use and Health documented 7% prescription drug misuse among adults [47] while current trend analysis reports a lower prevalence in 2018 when compared with rates reported in 2015–2017 among people aged 12 years or older [48]. Furthermore, in South Africa, codeine misuse was raised as a public health issue by health care providers since codeine-containing products were readily available in retail pharmacies and through the internet [31,32]. On the other hand, drug misusers in South African reported having exerted manipulative behaviour to influence healthcare workers in prescribing codeine-containing products [49].

Although these studies were conducted in upper-middle-income countries, an example in Jordan [50-52], China [53,54], Iran [55] and Turkey [56] among the general population instead of on PLWH, similar patterns of prescription drug misuse were observed, i.e. individuals stopped taking medications earlier than instructed [50,53], accessing prescription medicines without a doctor’s medical script [51-55], using the medication without a medical indication [56], and taking higher dosages in the event of missed doses [50]. Increases in prescription drug overdose deaths have coincided with an increased rate of drug diversion, drug misuse and prescription drug-related illnesses [57-59]. Despite that, there was no association between ART adherence and prescription drug diversion and misuse in the past seven days in our study. Earlier research in the United States found that prescription opioid misuse was linked with less than 90% ART adherence [60], while a study conducted in 2017 in the United States reported that opioid use was significantly associated with a failure to achieve viral load suppression [61].

In South Africa, the Fifth South African National HIV Prevalence, Incidence, Behaviour and Communication Survey, (SABSSM V) of 2017, reported the highest HIV prevalence at 27% (95% CI 23.9–30.4) in KwaZulu-Natal province among people aged 15–49 years [62]. This shows that the population of PLWH is growing disproportionally at the provincial level. Our findings predict their vulnerability to prescription drug diversion, and misuse. This reflect a need for tailor-made interventions to protect PLWH when they seek health care.

Though self-medicating is a commonly practised behaviour, inappropriate self-medication practices have potential risks, not limited to the incorrect manner of drug administration, incorrect dosage consumption, incorrect choice of medicines, adverse drug reactions, risky drug-drug interactions, and risk of developing drug dependence and abuse [63]. This strongly supports the association between misuse and self-medicating. Inappropriate self-medicating might hinder progress to achieving and maintaining viral suppression outcomes among PLWH.

Our participants showed no engagement with selling their prescribed medications, but prescription drug diversion was commonly reported. Although the reason is unclear in our local context, men have been reported to delay seeking healthcare professional for any illness due to competing forces such as masculinity and stigma, however during the course of the illness men do obtain healing through other avenues which include getting medications from their partners, family, friends, traditional healers or from the black market place. Individuals aged 30 years and older had experienced challenges, barriers, stigma and discrimination during the early years of the HIV epidemic when there were conflicting information about the HIV epidemic and also the lack of political support which facilitated more stigma, slow development of HIV programmes, implementation of HIV management programmes and support programmes for affected persons. This has played a critical role in accessibility and utilization of healthcare systems, especially for males, due to inadequate or lack of male-tailored services.

In our local context, the household income does not play a major role in health care since most of our participants accessed health care from the public health sector where all services were free. Furthermore, substance use and dependency are known barriers to ART adherence since it disrupts healthcare engagement and retention in care, which later translates into incomplete HIV care. The implication of incomplete HIV care is unsuppressed viral loads; therefore, greater risk of transmitting the HIV virus resulting in increased HIV incidence rates and possibly AIDS-related deaths, due to opportunities infections. Due to the diverse dynamics of people living with HIV, one size fits all interventions to protect vulnerable individuals from prescription drug diversion, and misuse would not work.

The researchers recommend additional awareness interventions to supplement the medicine information pamphlet that is read and comprehended by only a few consumers. PLWH often present with known risk factors associated with prescription drug diversion, and misuse which include, but are not limited to, substance use, mental health problems and chronic pain. They frequently get prescribed controlled medications to alleviate experienced health problems. Therefore consideration of the correlates of prescription drug diversion, and misuse among PLWH is critical.

Strengths and limitations

This study is among the first to investigate prescription drug diversion, and misuse among PLWH in the eThekwini district. Although the study consisted of sensitive questions, it provided valuable information on prescription drug diversion, and misuse. Behaviour practice-related questions were based on self-reports, and these may be impacted by recall and social desirability bias. The exclusion of a mental health variable in the analysis could be a limitation since mental health problems are one on the risk factors. Gaps were identified in the study questionnaire, and the questionnaire did not adequately cater for resource-limited settings, like the eThekwini district, since most people had low knowledge and awareness of different types of prescription drugs.

Due to limited human resource and funding to the study survey was only available in English, although these were the interviewer-administered questionnaires which allowed the researcher to assist participants who needed clarity in their preferred language. To this study’s advantage, the researcher was able to speak and understand most South African languages and all-dominating languages spoken in the KwaZulu-Natal province. The minimum sample size of 385 was determined using the formula developed by Daniel (1983) for a finite population with an assumption of 50% prevalence [64]. Due to limited funding for data collection, the study was able to enrol 392 PLWH, the small sample size to be representative PLWH in eThekwini district and attending healthcare services. A random sampling of public health clinics was employed, but there were inherent shortcomings due to convenient sampling. This approach is highly vulnerable to selection bias and influences beyond the control of the researcher has a great level of sampling error, and reliability can be compromised. Since some clinics refused to give permission to recruit participants at their facilities, the missed individuals might have different characteristics. The low variability in adherence scores (96% is impossible to really look at adherence in sub-analyses.)is a limitation, and future research should perhaps select PLWH with more diverse adherence profile. The type of prescription drugs diverted, and misused was excluded in the analysis since we only had data for the type of prescription drugs diverted. This posed as a limitation to our study analysis to compare the types of prescription drugs diverted, and misused. Comparisons were limited since the prevalence of prescription drug diversion, and misuse were investigated in different time periods in this study than in literature. Laboratory HIV viral load test results could be used to support the self-reported ART adherence. These findings may or may not be generalised beyond the eThekwini district among PLWH from resource-limited settings accessing care in public healthcare facilities, whereas prescription drug diversion, and misuse among PLWH accessing care in private facilities might be different.

Conclusion

The high prevalence of prescription drug diversion and misuse among PLWH underscores the need for integrated surveillance tools at all levels in the healthcare system. Monitoring the dispensing of medicine and the evaluation of medicine usage by healthcare providers is highly recommended for routine prescriptions and for patients presenting with risk factors. The researchers recommend that education and awareness campaigns should be investigated. These campaigns need to be directed at patients or public knowledge of the potential risks and promoting greater awareness among healthcare providers and to the public. Caution should be taken when prescribing medicines with potential risks for diversion, and misuse especially when the patient is presenting with risk factors. This study estimates the prevalence of prescription drug diversion, and misuse among PLWH in the eThekwini district. Prescription drug diversion, and misuse practices need urgent attention since there is a risk of it transitioning into illegal drug use.

Characteristics participants for the study.

(DOCX)

Click here for additional data file.

Correlates of prescription drug diversion adjusted for geographical setting among PLWH in eThekwini district, participants who ever diverted prescription drugs, N = 51.

(DOCX)

Click here for additional data file.

Correlates of prescription drug misuse adjusted for geographical setting among PLWH in eThekwini district, participants who ever misused prescription drugs, N = 89.

(DOCX)

Click here for additional data file.

Prescription drug diversion, misuse and abuse among people living with HIV in eThekwini District, KwaZulu-Natal.

(PDF)

Click here for additional data file.

(CSV)

Click here for additional data file.

10 Feb 2020

PONE-D-19-31426

Prevalence and Predictors of Prescription Drug Diversion, Misuse and Abuse among People Living with HIV in eThekwini District, KwaZulu-Natal, South Africa

PLOS ONE

Dear Ms. Buyisile,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Overall, reviewers thought this was an important topic and study (“really important and shows another angle in abuse and misuse of prescription medication”) but requires major revisions.  These reviews suggest two issues that should be the primary focus of the revision:

Contextualization in the literature, especially from South Africa

Doing this will also requiring clarifying key points raised by the reviewers

Make sure the terminology describing substance use is accurate and up to date for this field.

More clear and accurate presentation of the data, and in some places consider re-running statistics (see comments on logistic regression model)

Please consider carefully the comments on presentations of data and interpretation.

Mare sure that the narrative carefully reflects the results

Please address sample size considerations

Generally, there are many small errors that need to be corrected.

There is a growing literature that examines the multiple challenges faced by people living with HIV which can meaningfully be used to inform policy and intervention, so I encourage the authors to use these thoughtful and specific reviews to improve this paper.

We would appreciate receiving your revised manuscript by Wednesday, March 25 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Jennifer Zelnick

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please refer to any post-hoc corrections to correct for multiple comparisons during your statistical analyses. If these were not performed please justify the reasons. Please refer to our statistical reporting guidelines for assistance (https://journals.plos.org/plosone/s/submission-guidelines.#loc-statistical-reporting). In addition, please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information. This should also include further details of the pilot testing of the questionnaire, i.e. how many participants were involved and where they were recruited from.

3. We note that Figure 1 in your submission contain map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

You may seek permission from the original copyright holder of Figure 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

6. Please upload a copy of Supporting Information Table 1 which you refer to in your text on page 7.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Partly

Reviewer #4: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Data presented in the results section does not match the narrative (comments attached). It is also not clear if the authors are discussing diversion and misuse of ARVs, or opioids used for pain management.

Reviewer #2: I think the study is really important and shows another angle in abuse and misuse of prescription medication; but I have some concerns about it:

*I dont believe that most of the research on this topic is mostly conducted in the USA only

*Reference provided about studies done in other countries (China, Jordan) show findings of prescription drug misuse in the general population and not im PWLH; therefore not similar to this study's findings

*There are several studies conducted in SA that looked at use and abuse of prescription medications; but these studies are not mentioned; and the statement mentioned about the study on misuse of codeine containig medicines use is not correct

*Description of 'misuse and abuse' is not clearly stated and that is confusing

*I appreciate the acknowledgement of excluding mental health variable as a limitation; i think the authors should have looked at it as is almost impossible to talk about HIV and exclude mental health (depression, anxiety, etc)

*Looks like figue 2 and figure 3 is the same graph and the y-axis heading is the same, confusing

*There are spelling mistakes, so the authors need to re-read the manuscript to avoid this

*I would have liked to see the names of commonly misused/diverted/abused medications; to see if these medications are mostly associated with mental health illnesses (to self-medicate)

Data analysis was conducted rigourously and that is appreciated. So i think the authors should be given another chance to look at corrections and resubmit the manuscript

Reviewer #3: The reviewer feels that it would be extremely valuable for the authors to engage further with literature around prescription and over the counter medication in low and middle income countries, but in South Africa particularly to get a sense of what some of the concerns and issues are in this context. Examples are publications on codeine data (authors to look for are Parry, C., Dada, S. and Carney, T.) and the SACENDU data which is widely available. There are also definitely at least a few studies on opioid use and HIV, which is increasingly a problem in KZN. Definitions of misuse and abuse and diversion are also not clearly provided. Please look at the latest terminology used in substance use literature.

It would have been very interesting to look at the relationship between HIV outcomes and problem prescription use among PLWH-could these types of analyses be run? There are some studies conducted in South Africa (one qualitative one actually conducted in KZN) that mention people may be misusing ART as well. Was this included in the current study?'

The results section could be presented in a more succinct way and it is difficult to interpret Figure 3 and 4. A suggestion would be to present the univariate and multiple regression models in one table-perhaps look at other literature for examples.

A study limitation is that this is a small sample size to be representative of eThekweni PLWH and attending healthcare services. Are there any additional comments to make around this, such as do the authors think that this is a representative sample?

Reviewer #4: The authors present results of a cross-sectional study on the prevalence and predictors of prescription drug diversion, misuse and abuse among individuals with HIV in a district within South Africa. The authors find that there is more misuse than drug diversion among their sample, and very little prescription drug abuse. They also identify a few risk factors for drug diversion and misuse. The manuscript will be strengthened if the authors consider the following points.

1. Figure 2: it is unclear why the confidence intervals presented in Figure 2 are not symmetric, since typical confidence intervals for proportions are symmetric. Authors should clarify how confidence intervals were computed and verify that what is presented in Figure 2 is correct.

2. lines 212-216 (also in the Abstract lines 36-40) and lines 248-250: The authors are reporting percentages out of all 392 participants, which isn't necessarily wrong, but the start of the sentences ("Majority" or "More than one-third") suggest that the reader should be focused on just those with drug diversion or with drug misuse. The authors are encouraged to think about what point they really want to make and whether it is most appropriate to report percentages out of the entire sample or out of just those that had the action of interest.

3. The authors have a tendency to overstate the findings, even when qualifying a statement by saying something wasn't statistically significant. In most of the situations, findings are not even close to being significant. Examples include lines 233-239, 241-244, 263-266, 272-275). Authors already include the ORs and 95% confidence intervals for the terms in the model in supplemental table 3, so readers can see these results. Some of these end up being discussed in the discussion (age and aware of the black market) and highlighted in the Abstract (homeless and awareness of the black market) as findings, which is inappropriate. In the discussion, they further highlight differences between males and females, but this finding does not hold up in the adjusted model.

4. In the adjusted models for drug diversion (model B), the authors include more variables than are generally recommended given the number of people who had the action of interest.Standard rules of thumb for logistic regression would suggest that no more than 5 predictors should be included in a model, yet the authors include 7.

Minor points:

1. line 27: "as mostly" should be "as most"

2. line 42: "diversion and misuse" should be "diversion or misuses" since not all factors were associated with both outcomes.

3. line 62: "South Africa account" should be "South Africa accounts" and consider changing "owning the" to "having the"

4. line 63: "has been" should be "have been"

5. line 70: "that correlates" should be "that correlate"

6. line 71: "suggest" should be "suggests"

7. line 73: "and abuse has been" should be "and abuse have been"

8. line 83: what do the authors mean by saying trends are close to non-existence (or non-existent) - do they mean there is little information?

9. lines 86-87: the sentence starting with "Furthermore" is an incomplete sentence

10. line 106: remove "until"

11: line 130: "Data was" should be "Data were"

12. line 144: "development the" should be "development of the"

13.lines 145-146: the sentence "The emphasises the importance..." is unclear and authors are encouraged to reword it for clarity.

14. line 148: "diversion stem" should be "diversion which stems"

15. line 150: "is prescription drug misuse was" should be " is prescription drug misuse and was"

16. line 151: "drug misuse stem" should be "drug misuse stems"

17. line 185: "rigour strategy" should be "rigorous strategy"

18. line 193: "Analysis were" should be "Analyses were"

19: line 203: the authors report a response rate of 102%, but it is unclear how a response rate can be larger than 100%.

20. Figure 2 and 4 appear to be identical, so authors should verify that the correct figure was uploaded for Figure 4.

21. lines 287-291: these lines should be reworded given that the percentages are so similar.

22. lines 298-299 - authors should be clear this result is from the adjusted model

23. line 328: "with an increase rates" should be "with increased rates"

24: line 349: "This strongly support" should be "This strongly supports"

25. line 361: "metal" should be "mental"

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Reviewer #4: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PONE-D-19-31426_reviewer.pdf

Click here for additional data file.

9 Apr 2020

Rigorous editing was done to the entire paper.

Definition section was added.

The narrative of the results section was revised to match the data.

The type of prescription drugs diverted, misused and abused were excluded in the analysis since we only had data for type of prescription drugs diverted and abused. This posed as a limitation to our study analysis to list and compare the types of prescription drugs diverted, misused or abused.

The authors had taken note that the p-values were >0.05 (no association). However, our participants with the following characteristics (homelessness, used tobacco, used marijuana and are of the black had higher chances of misusing prescription medication.

Studies conducted in South Africa were included.

The study was self-report and most participants did not know their latest viral load. We were unable to evaluate the HIV outcomes.

Regression models A and B were presented in one table for prescription drug diversion and the same was done for prescription drug misuse.

Mode B for prescription drug diversion and misuse with only five variables was redone. However, we found that there was no significant difference between the original model and revised model looking at the Odds ratios. Therefore both Model B figures were unchanged.

Due to limited funding for data collection the study was only able to enroll 392 PLWH. However The minimum sample size of 385 was determined using the formula developed by the Daniel (1982) for a finite population with an assumption of 50% prevalence.

Submitted filename: Response to reviewer comments 3 April 2020.docx

Click here for additional data file.

9 Jul 2020

PONE-D-19-31426R1

Prevalence and Predictors of Prescription Drug Diversion, Misuse and Abuse among People Living with HIV in eThekwini District, KwaZulu-Natal, South Africa

PLOS ONE

Dear Dr. Chibi,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Aug 23 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Becky L. Genberg

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

While some of the concerns from the previous review have been addressed, there remain concerns by one reviewer.

Additionally, I have concerns regarding the definitions of prescription drug abuse, self-medication, and diversion. Firstly, what drugs specifically are you inquiring about? the potential for misuse or abuse and the implications for HIV prevention and treatment would apply only to specific classes of prescription drugs. Were these assessed? Secondly, the definition of abuse is not in line with the substance use literature and should be reconsidered. Please clarify that 100% response rate means that all persons approached to participate agreed and completed the survey. Finally I also share concerns of the reviewer about the need for careful editing, as some numbers appear incorrect and there are several typographical and grammatical errors throughout the manuscript that remain in this revision. In this revision I would recommend adhering to commonly accepted terms when referring to people who use drugs.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

Reviewer #4: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

Reviewer #4: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

Reviewer #4: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #4: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: (No Response)

Reviewer #4: The authors have addressed the majority of my earlier concerns. There are a few remaining issues that should be addressed.

1. line 103: "might unfolds" should be "might unfold"

2. lines 122-123: "areas with lack training" should possibly be "areas which lack tracking"

3. line 207: "would their participation affect their healthcare" should be "and participation would not affect their healthcare"

4. line 280: mention these are exact 95% confidence intervals

5. Table 1: authors should carefully check numbers and p-values as I found a number of incorrect calculations. Examples include the percentage for 18-24 years in the Total sample (which should be 10% not 7%), the percentage of completed high school and tertiary studies in the did not misuse prescription drugs group (which should be 10% not 26%), the numbers of homeless add to 18, not 17 when looking at the yes/no columns for both divert and misuse prescriptions, the p-value for ever used alcohol comparing those that diverted and those that did not seems wrong (based on the numbers provided), I get a much smaller p-value for ever used tobacco in those that diverted vs not, the numbers of ever used marijuana in the diverted and did not divert columns add to 56 not 55.

6. It is recommended that in the text (lines 339, 360, 390) and S2 and S3 tables, the authors remind the readers what model A and model B mean. At a minimum, there should be a note for S2 and S3 tables that defines these models.

7. lines 350-356 are all results that have extremely high p-values, so reporting a 20% or 50% increase in odds is misleading. There is no evidence in the data to suggest that there is any increased chance of diverting PM based on these characteristics. I am fine with keeping the marijuana and illegal drugs findings, since with the 95% confidence interval, the readers can see that the lower limit for the CI is just under 1.

8. lines 365 and 367, remove "among the participants who misused drugs", because the authors are reporting percentages out of the entire sample not just those who misused drugs.

9. lines 321-385: it is misleading to say that these have higher odds of misusing PM, since the results all have p-values >0.1.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Reviewer #4: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

25 Aug 2020

All reviewers suggestions were considered.

All analysis were redone to double check any errors. Table 1, supplementary information and Fig 1-4 were update.

Definitions were updated according to the National Institute on Drug Abuse.

Prescription drugs inquired about in the study were sedatives, stimulants, analgesics, antibiotics, antiretroviral drugs and any emerging prescription drugs with potential of being diverted, misused or abused.

For example:

Any sedatives (tranquilizers) such as Xanax, Valium, Ativan, Klonopin, Soma, Barbiturates or others

• For euphoric effects such as excitement and happiness

Any stimulants such as Adderall, Ritalin, Adipex or any other stimulants

• For increasing the ability to concentrate, boost feelings and performance enhancing

Any analgesics (painkillers) such as Opioids, Codeine, Tramadol, percocet, Vicodin, OxyContin, , Demerol, Darvon, Demerol, Morphine, Methadone, Fentanyl, Pentazocine

• For reducing pain or feeling high

Any antibiotics such as amoxicillin, ampicillin, cloxacillin, metronidazole, co- trimoxazole, ciprofloxacin, or others

Any ARVS

Were the prescription drugs assessed?

There was a limitation in the study since most participants had low knowledge and awareness of different types of prescription drugs. Participants were unable to specify the exact name of the prescription medication.

Grammarly was used to improve the manuscript.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

16 Sep 2020

PONE-D-19-31426R2

Prevalence and Predictors of Prescription Drug Diversion, Misuse and Abuse among People Living with HIV in eThekwini District, KwaZulu-Natal, South Africa

PLOS ONE

Dear Dr. Chibi,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

ACADEMIC EDITOR:

Thank you for addressing most of the concerns from the previous review. There are some remaining concerns:

1) The legends (or footnotes) for Figures 2 and 4 should specify the difference between Model A and Model B.

2) In the analysis section, the statements about the selection of factors included in the model should be appropriately referenced. For example, where in lines 329-30 commenting on "known confounders" of the exposure-outcome relationship. Also, with respect to the language in this paragraph; these are associations, not effects, between the variables and the outcomes. These factors are also more appropriately termed correlates, not predictors.

4) The assessment of "drug abuse" is not an accepted method of measurement, and the terminology is outdated. The idea that one could assess this type of use by directly asking about use to intentionally harm oneself lacks the understanding that addiction is a behavioral pattern in which one uses substances despite the negative consequences. I highly recommend removing this from the paper and focusing on diversion and misuse and the types of misuse examined.

==============================

Please submit your revised manuscript by Oct 31 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Becky L. Genberg

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

26 Sep 2020

Reviewer comment 1: The legends (or footnotes) for Figures 2 and 4 should specify the difference between Model A and Model B.

Author response 1: Description of model A and B were explained in the legends for bot drug diversion and misuse.Page 25, line 382-387; Page 26, line 415-421

Reviewer comment 2: 2) In the analysis section, the statements about the selection of factors included in the model should be appropriately referenced. For example, where in lines 329-30 commenting on "known confounders" of the exposure-outcome relationship. Also, with respect to the language in this paragraph; these are associations, not effects, between the variables and the outcomes. These factors are also more appropriately termed correlates, not predictors.

Author response 2: Reference 40 and 41 were added; Page 14, line 318. The word predictors was replaced with correlates in the manuscript title; Page 1, line 1. The word effect was replaced with associations. Page 13, line 308-317

Reviewer comment 3: 4) The assessment of "drug abuse" is not an accepted method of measurement, and the terminology is outdated. The idea that one could assess this type of use by directly asking about use to intentionally harm oneself lacks the understanding that addiction is a behavioral pattern in which one uses substances despite the negative consequences. I highly recommend removing this from the paper and focusing on diversion and misuse and the types of misuse examined.

Author response 3: Suggestion taken into consideration. All information related to drug abuse was removed from the paper.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

30 Nov 2020

Prevalence and correlates of prescription drug diversion and misuse among people living with HIV in the eThekwini district, KwaZulu-Natal, South Africa

PONE-D-19-31426R3

Dear Dr. Chibi,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Dawn K. Smith

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Thank you for the revisions in response to reviewer comments.

7 Dec 2020

PONE-D-19-31426R3

Prevalence and correlates of prescription drug diversion and misuse among people living with HIV in the eThekwini district, KwaZulu-Natal, South Africa

Dear Dr. Chibi:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Dawn K. Smith

Academic Editor

PLOS ONE