Healthcare avoidance by people who inject drugs in Bangkok, Thailand

Abstract

Background

Although people who inject drugs (IDU) often contend with various health-related harms, timely access to health care among this population remains low. We sought to identify specific individual, social and structural factors constraining healthcare access among IDU in Bangkok, Thailand.

Methods

Data were derived from a community-recruited sample of IDU participating in the Mitsampan Community Research Project between July and October 2011. We assessed the prevalence and correlates of healthcare avoidance due to one's drug use using multivariate logistic regression.

Results

Among 437 participants, 112 (25.6%) reported avoiding health care because they were IDU. In multivariate analyses, factors independently associated with avoiding health care included having ever been drug tested by police [adjusted odds ratio (AOR) = 1.80], experienced verbal abuse (AOR = 3.15), been discouraged from engaging in usual family activities (AOR = 3.27), been refused medical care (AOR = 10.90), experienced any barriers to health care (AOR = 4.87) and received healthcare information and support at a drop-in centre (AOR = 1.92) (all P < 0.05).

Conclusions

These findings highlight the need to address the broader policy environment, which perpetuates the criminalization and stigmatization of IDU, and to expand peer-based interventions to facilitate access to health care for IDU in this setting.

Background

People who inject drugs (IDU) often contend with a number of health-related harms that necessitate health care, including serious infectious diseases such as hepatitis B and C¹ and human immunodeficiency virus (HIV).^2,3 Accordingly, considerable attention has been put on the prevention and treatment of HIV infection among IDU.^3–7 For example, the World Health Organization, United Nations Office of Drugs and Crime and Joint United Nations Programme on HIV/AIDS (UNAIDS) have endorsed the recommendation of a ‘comprehensive package’ of harm reduction and healthcare services to address HIV/AIDS among IDU, which include HIV testing and counselling, antiretroviral therapy (ART), and prevention and treatment of viral hepatitis, tuberculosis and sexually transmitted infections.⁸

Thailand has long contended with dual epidemics of injection drug use and HIV infection,⁹ and has been criticized for failing to implement recommended programmes and services for IDU.¹⁰ Not surprisingly, HIV prevalence among IDU in Thailand continues to be between 30 and 50%.¹¹ As well, a high prevalence of injection-related nerve and vascular complications,¹² and hepatitis C virus (HCV) rates as high as 86%¹³ have been reported among this population. Yet IDU in Thailand are known to have disproportionately low access to ART,^9,14 and provision and utilization of services such as needle and syringe programmes^15,16 and HCV testing¹⁷ are low. However, little is known about more general healthcare access.

Following historically strict penalties for narcotics trafficking, production, possession and consumption,¹⁸ the 2003 Thai ‘War on Drugs’ campaign increased targeted policing and incarceration of IDU.¹⁹ This strategy, which has continued to be supported by successive governments, has likely created further harm, particularly given the known health risks associated with incarceration among IDU.^20–22 The criminalization of IDU and the associated stigmatization that often occurs within healthcare settings have been linked to negative HIV-related outcomes, including low rates of ART among IDU in other settings^23–25 and HIV test avoidance in Thailand.²⁶ Qualitative and anecdotal evidence suggests that the same may be true for access to other healthcare services.^19,27,28 In light of the ongoing problems with access to care among Thai IDU and repressive drug policies in Thailand, we sought to examine the social and structural factors that may be shaping healthcare avoidance among IDU in Bangkok, Thailand. By healthcare avoidance, we are referring to one avoiding accessing healthcare services specifically because of their status as a drug user.

Methods

Data for this study were derived from the Mitsampan Community Research Project (MSCRP), a collaborative research effort involving the Mitsampan Harm Reduction Center (MSHRC; a drug user-run drop-in centre in Bangkok, Thailand), Thai AIDS Treatment Action Group (Bangkok, Thailand), Chulalongkorn University (Bangkok, Thailand) and the British Columbia Centre for Excellence in HIV/AIDS/University of British Columbia (Vancouver, Canada). Between July and October 2011, the research partners undertook a cross-sectional survey involving 440 community-recruited IDU in Bangkok. Potential participants were recruited through peer-based outreach efforts and word of mouth and were invited to attend the MSHRC or O-Zone House (another local drop-in centre) to be part of the study. Adults residing in Bangkok or the adjacent provinces who had injected drug(s) in the previous 6 months were eligible for participation in the study. All participants provided oral informed consent and completed an interviewer-administrated questionnaire eliciting a range of information, including demographic characteristics, drug use patterns, experiences with the criminal justice system and access to healthcare services. The questionnaire was piloted with peer researchers (i.e. active or former IDU at the MSHRC) prior to data collection to ensure face validity and cultural appropriateness. While the total number of individuals contacted is unknown due to the peer-based method of recruitment (e.g. word of mouth), a total of 442 individuals provided informed consent and 440 completed the interview. Participants were given 350 Thai Baht (∼$12 USD) upon completion of the questionnaire. The study was approved by the research ethics boards at Chulalongkorn University and the University of British Columbia. A thorough description of the MSCRP and the development of the survey instruments used in this study has been previously published.²⁹

The sample was restricted to individuals who provided complete data for the present analyses. The primary outcome of interest for this study was healthcare avoidance, defined as answering ‘yes’ to the following question: ‘Do you sometimes avoid accessing healthcare services because you are a drug user?’ Based on the ‘Risk Environment’ framework³⁰ and previous studies on HIV treatment²⁵ and test avoidance²⁶ among IDU, we considered a range of individual, social and structural factors as well as previous experiences with accessing health care that were hypothesized to be associated with the outcome. Specifically, demographic and drug use behaviour variables included: gender (male versus female), age (below or above median age; <38 versus ≥38 years), education level (<secondary education versus ≥secondary education), HIV serostatus (positive versus negative or unknown) and injection frequency in the previous 6 months for each of heroin, midazolam and methamphetamine (methamphetamine pills, locally called yaba; or crystal methamphetamine, locally called ice) (>once per week versus ≤once per week). Social and structural variables included: exposures to drug law enforcement, including ever incarcerated (yes versus no); ever drug tested by the police (yes versus no); and ever beaten by the police (yes versus no). Drug use-related stigma experience variables included: ever experienced verbal abuse from anyone related to one's drug use (yes versus no); ever discouraged from engaging in usual family activities (yes versus no); and ever refused medical care by a healthcare worker (yes versus no). Table 1 presents the survey questions asking about stigma experiences. Finally, healthcare experience variables included: ever accessed voluntary treatment; ever received healthcare information and support at a drop-in centre (e.g. the MSHRC); and any barriers to accessing health care in the previous 6 months. The barriers to accessing healthcare variable included a range of potential barriers: individual barriers such as lack of identification and money, social barriers such as being poorly treated by healthcare workers and fear of sharing information with police, and structural barriers such as hours of operation and wait time.

Table 1.

Drug use-related stigma experience questions from the MSCRP questionnaire used in the present analysis of healthcare avoidance

Interviewer-administered survey question	Derived variable
Have you ever been verbally abused or teased because you are a drug user? (For example, has anyone told you that you are a bad person because you are a drug user?)  □ Yes  □ No	Ever experienced verbal abuse (yes versus no)
Have some of your family ever discouraged you from engaging in usual family activities because you are a drug user?  □ Yes  □ No	Ever discouraged from engaging in usual family activities (yes versus no)
Have you ever had a healthcare worker refuse to treat you or denied you access to medical treatment or care because you are a drug user?  □ Yes  □ No  □ N/A—never accessed health care	Ever refused medical care by a healthcare worker (yes versus no or N/A)

To examine bivariate associations between the hypothesized explanatory variables and the outcome, we used Pearson's χ² test. We then used an a priori-defined statistical protocol that examined factors associated with the outcome by fitting a multivariate logistic regression model that included all variables that were associated with the outcome at the P < 0.05 level in bivariate analyses. All P-values were two sided. All statistical analyses were performed with SPSS version 21.0 (IBM SPSS Statistics, IBM Corporation, 2012).

Results

In total, 437 individuals were included in the study, including 85 females (19.5%). The median age was 38 years (interquartile range: 34–48 years). In total, 112 (25.6%) participants reported healthcare avoidance due to their drug use. As indicated in Table 2, in bivariate analyses, healthcare avoidance was significantly and positively associated with a number of social and structural factors, including having ever been drug tested by police [odds ratio (OR) = 1.80; 95% confidence interval (CI): 1.10–2.93], ever experienced verbal abuse (OR = 3.15; 95% CI: 1.99–4.98) and ever discouraged from engaging in usual family activities (OR = 3.27; 95% CI: 2.04–5.24); as well as the following healthcare experiences: having ever received healthcare information and support at a drop-in centre (OR = 1.92; 95% CI: 1.16–3.17), reporting any barriers to accessing health care (OR = 4.87; 95% CI: 2.44–9.70) and having ever been refused medical care (OR = 10.90; 95% CI: 5.39–22.06).

Table 2.

Bivariate analyses of factors associated with avoiding health care among a community-recruited sample of IDU in Bangkok, Thailand (n = 437)

Characteristic	Avoiding health care		Odds ratio (95% CI)	P-value
	Yes 112 (25.6%)	No 325 (74.4%)
Gender
Male	93 (83.0)	259 (79.7)	1.25 (0.71–2.19)	0.441
Female	19 (17.0)	66 (20.3)
Age (years)
<38	45 (40.2)	160 (49.2)	0.69 (0.45–1.07)	0.098
≥38	67 (59.8)	165 (50.8)
Education level
<Secondary education	43 (38.4)	129 (39.7)	0.95 (0.61–1.47)	0.808
≥Secondary education	69 (61.6)	196 (60.3)
HIV serostatus
Positive	21 (18.8)	60 (18.5)	1.02 (0.59–1.77)	0.946
Negative or unknown	91 (81.2)	265 (81.5)
Heroin injection frequencya
>Once per week	31 (27.7)	63 (19.4)	1.59 (0.97–2.62)	0.065
≤Once per week	81 (72.3)	262 (80.6)
Midazolam injection frequencya
>Once per week	70 (62.5)	170 (52.3)	1.52 (0.98–2.36)	0.062
≤Once per week	42 (37.5)	155 (47.7)
Methamphetamine injection frequencya
>Once per week	25 (22.3)	74 (22.8)	0.98 (0.58–1.63)	0.922
≤Once per week	87 (77.7)	251 (77.2)
Incarcerated ever
Yes	86 (76.8)	240 (73.8)	1.17 (0.71–1.94)	0.538
No	26 (23.2)	85 (26.2)
Ever drug tested by police
Yes	85 (75.9)	207 (63.7)	1.80 (1.10–2.93)	0.018
No	27 (24.1)	118 (36.3)
Ever beaten by police
Yes	53 (47.3)	128 (39.4)	1.38 (0.90–2.13)	0.141
No	59 (52.7)	197 (60.6)
Ever experienced verbal abuse
Yes	78 (69.6)	137 (42.2)	3.15 (1.99–4.98)	<0.001
No	34 (30.4)	188 (57.8)
Ever discouraged from engaging in usual family activities
Yes	82 (73.2)	148 (45.5)	3.27 (2.04–5.24)	<0.001
No	30 (26.8)	177 (54.5)
Ever accessed voluntary drug treatment
Yes	91 (81.2)	247 (76.0)	1.37 (0.80–2.35)	0.252
No	21 (18.8)	78 (24.0)
Ever received healthcare information and support at a drop-in centre
Yes	32 (28.6)	56 (17.2)	1.92 (1.16–3.17)	0.010
No	80 (71.4)	269 (82.8)
Ever refused medical care
Yes	33 (29.5)	12 (3.7)	10.90 (5.39–22.06)	<0.001
No	79 (70.5)	313 (96.3)
Reporting any barriers to accessing health carea
Yes	102 (91.1)	221 (68.0)	4.87 (2.44–9.70)	<0.001
No	10 (8.9)	104 (32.0)

IDU, people who inject drugs; CI, confidence interval.

^aRefers to activities in the previous 6 months.

Table 3 shows the results of the multivariate logistic regression analysis. As indicated, having ever been drug tested by police [adjusted odds ratio (AOR): 1.89; 95% CI: 1.08–3.29], ever experienced verbal abuse (AOR: 1.86; 95% CI: 1.10–3.17) and ever discouraged from engaging in usual family activities (AOR: 1.78; 95% CI: 1.04–3.06) remained independently and positively associated with healthcare avoidance; as did the healthcare experience variables: having received healthcare information and support at a drop-in centre (AOR: 1.89; 95% CI: 1.07–3.35), having ever been refused medical care (AOR: 7.73; 95% CI: 3.61–16.55) and reporting any barriers to health care (AOR: 3.02; 95% CI: 1.46–6.24).

Table 3.

Multivariate logistic regression analysis of factors associated with avoiding health care among a community-recruited sample of IDU in Bangkok, Thailand (n = 437)

Variable	AOR	95% CI	P-value
Ever drug tested by police
(Yes versus no)	1.89	(1.08–3.29)	0.025
Ever experienced verbal abuse
(Yes versus no)	1.86	(1.10–3.17)	0.021
Ever discouraged from engaging in usual family activities
(Yes versus no)	1.78	(1.04–3.06)	0.037
Ever received healthcare information and support at a drop-in centre
(Yes versus no)	1.89	(1.07–3.35)	0.029
Ever refused medical care
(Yes versus no)	7.73	(3.61–16.55)	<0.001
Reporting any barriers to accessing health carea
(Yes versus no)	3.02	(1.46–6.24)	0.003

IDU, people who inject drugs; AOR, adjusted odds ratio; CI, confidence interval.

^aRefers to activities in the previous 6 months.

Discussion

Main findings of this study

We found that approximately one quarter of Thai IDU in this study reported avoiding health care due to their drug use. Social and structural factors and healthcare experiences positively and independently associated with healthcare avoidance included: experiencing drug testing by police, verbal abuse, discrimination from family members, denial of medical care, reporting any barriers to health care and having ever received healthcare information and support at a drop-in centre.

What is already known on this topic?

A high prevalence of healthcare avoidance among IDU in this setting is concerning given the known health-related harms associated with injection drug use. While this is one of the first studies to measure healthcare avoidance explicitly, a few have assessed healthcare-seeking behaviour: in Mexico, only 12% of IDU sought formal treatment for skin abscesses³¹; in Miami, many chronic drug users reported ‘attitudinal and behavioural barriers’ to health care, such as not wanting treatment or self-treating³²; and in India, 33% of IDU did not seek formal healthcare services.³³ The latter, however, was to our knowledge the only study to analyse social and structural factors and found that low utilization of healthcare services was associated with exposure to policing, social exclusion and negative healthcare experiences.

What this study adds?

This study adds important evidence that social and structural forces operating in the broader risk environment of IDU can affect healthcare-seeking behaviour among IDU.

We found that participants who experienced drug testing by police were more likely to avoid health care. Despite legislation protecting personal information, it has been reported that some healthcare workers disclose information on suspected IDU to police, or that police obtain information from healthcare records by other measures.^19,34 Further, police are known to conduct urine testing around methadone clinics, and there is qualitative evidence of checkpoints such as these discouraging IDU from accessing health care.³⁵ Accordingly, IDU may be avoiding hospital care out of fear of potential arrest. Prior police exposure may also increase real or perceived risk, as previous studies of IDU in Bangkok have found increased police presence to be associated with HIV test avoidance.²⁶ The findings presented herein add to the growing body of evidence that the Thai government's emphasis on aggressive drug law enforcement undermines public health by constraining healthcare access,^19,28 which is consistent with international literature suggesting the negative impact of aggressive policing tactics.^25,33,36,37 As urged by the World Bank and UNAIDS,^4,10,11,38 Thai policymakers should re-examine their ongoing punitive strategies and instead place emphasis on access to comprehensive harm reduction and healthcare services.

We also found that denial of medical care increased the likelihood of avoiding health care, as did experiencing any barriers to healthcare access. That negative experiences in healthcare settings would be discouraging to utilization is understandable and supported by the previously mentioned Indian study, which found that IDU with such experience were less likely to utilize healthcare services.³³ All Thai citizens have a constitutional right to standard care without discrimination,^34,39 and narcotics legislation specifically identifies drug users as ‘patients’ in need of treatment^40,41; therefore, policymakers and health officials should seek to address and reduce the stigmatizing attitudes towards IDU that persist in healthcare settings in Thailand^28,42 to ensure fair and universal access.

Our findings also suggest that verbal abuse and stigma from families increased the likelihood of healthcare avoidance. Feelings of social exclusion may compound the fear of stigma from healthcare workers. In Vietnam, stigma in health care and the community was found to hamper utilization of HIV treatment, while family support was vital to outcomes following HIV diagnosis.²⁴ Interventions that aim to improve familial support, which may provide encouragement to access health care, and to reduce the view of IDU as dangerous ‘social cheaters’⁴² by healthcare workers and the public are needed.

Finally, we found an independent association between avoiding healthcare and accessing healthcare information and support through a drop-in centre, suggesting that peer-based services are reaching a subpopulation of IDU who avoid conventional healthcare venues. Previous studies have shown that Thai drug users are willing⁴³ or prefer to access services at a ‘centre just for drug users’²⁸ and internationally, peer-run harm reduction efforts have been found to successfully reduce risky injecting behaviour and extend the reach and coverage of conventional public health programmes.^44–46 Therefore, our finding may suggest that these programmes in Thailand have an opportunity to improve healthcare access by reaching IDU who are underserved by conventional programmes and services. Harm reduction centres like the MSHRC and O-Zone House, for example, already provide access to health care by referral or, in the case of the MSHRC, accompaniment to appropriate healthcare providers. Expanding such peer-run harm reduction efforts and the services provided, including provision of some healthcare services by either trained peers or healthcare workers who are specially trained to work with IDU, may improve healthcare access when access elsewhere is constrained. Future research should examine whether access to peer-based services or increased healthcare services offered at these centres, or both, improves access to health care among IDU who actively avoid health care.

Limitations of this study

This study has several limitations. First, because of the cross-sectional nature of this study, we cannot determine a temporal relationship between the explanatory variables and outcome. Second, since the study sample was not randomly selected, the study findings may not be representative of all IDU in Bangkok. Third, the data collected were self-reported and may be subject to reporting biases, including socially desirable responding and recall bias. For example, experiences of sensitive events or behaviour, such as drug testing by the police, may be under-reported because of socially desirable responding, while experiences of healthcare access may be over- or under-reported by recall bias. We believe, however, that it is unlikely that these biases differentially affected data by the healthcare avoidance status. Fourth, the survey instrument used was not statistically or biologically validated, although it was piloted with peer researchers in the study setting to ensure face validity and cultural appropriateness. Fifth, this study did not measure healthcare access or adequacy. Given the known health risks of injection drug use, healthcare access is vitally important. The findings of this study provide justification for further, in-depth qualitative exploration of healthcare access and adequacy. Finally, because of the small sample size, there were wide intervals around some of the estimates reported.

Conclusions

In summary, we found a high prevalence of healthcare avoidance among IDU in Bangkok. Various social and structural factors, including drug testing by police, verbal abuse and exclusion from family activities, and negative experiences with health care, including being refused services and reporting barriers, were associated with healthcare avoidance. These findings add further evidence of the need to address punitive drug policies and to reduce stigma and discrimination against IDU among healthcare workers and the public. Accessing healthcare information and support at a peer-run drop-in centre was independently associated with healthcare avoidance, suggesting that this type of centre has potential to facilitate healthcare access among IDU through the provision of healthcare services or integration with the existing healthcare system.

Author's contributions

A.H., K.H. and T.K. designed the study. A.H. and K.H. conducted the statistical analyses. A.H. drafted the manuscript and incorporated all suggestions. All authors made significant contributions to the conception and design of the analyses, interpretation of the data and drafting of the manuscript, and all authors approved the final manuscript.

Funding

The study was supported by the Michael Smith Foundation for Health Research. This research was also undertaken, in part, thanks to funding from the Canada Research Chairs Program through a Tier 1 Canada Research Chair in Inner City Medicine that supports E.W. K.H. is supported by the Canadian Institutes of Health Research New Investigator Award (MSH-141971). L.T. is supported by the Canadian Institutes of Health Research Fellowship.