Chronic pain and associated factors in Maputo, Mozambique: a pilot study

Abstract

Introduction:

Studies estimate that 20% of adults suffer from chronic pain. A meta-analysis in low- and middle-income countries (LMICs) found 34% had chronic pain. There are few studies on pain prevalence gathered in Africa. This study surveyed the capital city of Mozambique.

Methods:

This was a cross-sectional study employed in a community setting. The Vanderbilt Global Pain Survey comprised questions on the behaviour and attitudes of respondents regarding pain, including previously validated metrics: the Pain Catastrophizing Scale, the World Health Organization Disability Assessment Schedule, the Brief Pain Inventory, Widespread Pain Index and Symptom Severity Score, and the Michigan Body Map.

Results:

Ninety-seven surveys were completed out of 100. Pain every day lasting for more than 6 months in their lifetime was reported as 39.2% (CI: 29.4–49.6), and 52% of respondents had pain the day of the interview. However, the pain resulted in little difficulty with activities of daily living and maintaining relationships (61%–89%). Although none reported mental health disorders, 53.6% had experienced a traumatic event in their life, with 45.2% having related nightmares, anxiety, or fear. Most respondents (99%) would take oral medication if it helped their pain, with a large proportion willing to spend significant money for these (49% would pay >US$40) and willing to travel long distances to get help (55.2% would travel >40 kilometer).

Conclusion:

The prevalence of chronic pain in Maputo, Mozambique is similar to the average for LMICs. Trends in high-income countries suggest that multimodal pain management and multidisciplinary treatments may improve optimal pain control in LMICs.

Introduction

Studies in high-income countries (HICs) estimate that one in five adults suffer from chronic moderate to severe pain.^1–4 A meta-analysis of the prevalence of chronic pain in low- and middle-income countries (LMICs) found 34% of the general population and 62% of the elderly had chronic pain.⁵ A survey of 17 countries reported pain prevalence to be 37.3% in HICs and 41.1% in LMICs.⁶ Half or two-thirds of these patients have decreased or no ability to participate in normal daily activities.⁴ Disability associated with pain includes physical limitations as well as psychological burden, with increased rates of anxiety, depression, abuse and post-traumatic stress disorder in this population.^7,8 The economic burden of chronic pain is high and can be significant on the family of a chronic pain patient.⁹ Limited studies have examined the prevalence and burden of pain in low resource settings, with even fewer data gathered in Africa.^10,11

Pain diagnosis and treatment, along with access to essential medicines are considered a basic human right as stated by the International Association for the Study of Pain and the World Health Organization.⁸ Despite global consensus, 80% of the world population experiences insufficient access to pain medicines and treatments.^12,13 Furthermore, pain treatment medications are unevenly distributed in the world, with 89% of morphine consumption in North America and Europe and only 6% in LMICs.^14–16 In addition, a significant number of healthcare providers in Africa report inadequate pain treatment training.^16,17

Mozambique is located in Southeast Africa and has an estimated population of 26 million persons.^18,19 The United Nations Human Development Index (HDI) ranked Mozambique 181 out of 188 countries in 2016.²⁰ There is paucity of pain studies in Mozambique, with pain prevalence only reported in two populations with one paediatric study and one study of pain clinic patients. Children aged 6–7 years had a low back pain rate of 13.5%. Fifty-eight percent of the children reported having low back pain in their lifetime with 28% having multiple episodes.²¹ The Maputo Central Hospital established a pain unit in 2007. Most patients reported osteomuscular (33.9%) or neuropathic pain (33.9%), with 33.1% also reporting depression and 17% anxiety.²² Almost half of the patients in the pain unit (49.2%) also received treatment from traditional healers.^23,24 Paracetamol was the most commonly used drug (33.9%) with morphine used in 10.2%. About 40.7% of patients had nerve blocks to treat pain. A significant number of patients (78%) reported pain interfering with activities.^25,26

The primary objective of this study was to determine pain prevalence among a pilot population in Maputo, the capital of Mozambique, with the goal of expanding this data set to the whole country in the future.

Methods

Following IRB approval at Vanderbilt University and within Mozambique, this cross-sectional structured interview was conducted in a community setting using a household survey. The surveys contained questions on pain levels, related behaviours and attitudes on pain, demographics, medical or psychological conditions and disability. The Vanderbilt Global Pain Survey (VGPS)²⁷ included validated pain prevalence metrics: the four-item short form of the Pain Catastrophizing Scale (PCS), the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0), the Brief Pain Inventory (BPI), Widespread Pain Index and Symptom Severity Score from the American College of Rheumatology Diagnosis Criteria for Fibromyalgia (WPI) and the Michigan Body Map (MBP).^28–32 The surveys also included a visual analogue scale for determination of pain levels and were piloted in Nepal and India, allowing for evaluation in two different cultures.²⁷

Stratified samples of respondents were collected with sizes proportional to the population demographics information obtained from 2007 census data.³³ The city of Maputo is divided into urban, sub-urban and peri-urban areas, indicating a grade between urban and neighbourhoods with rural elements.^34,35 Neighbourhoods were randomly selected, with one neighbourhood out of each of the three areas. The urban neighbourhood selected was Bairro Central B, sub-urban Bairro Chamanculo D and peri-urban Bairro Mahotas. Households within neighbourhoods were systematically chosen: once the interviewer arrived in the neighbourhood, they visited every third house starting at the right of the administration building. If no respondent was home, the surveyor proceeded to the next third household. Surveys were conducted on weekdays between 8 a.m. and 6 p.m. and on Saturdays from 8 a.m. to 2 p.m. Interviewers were Mozambican medical students trained to maintain neutral expressions and language throughout the interview. Surveys were conducted in Portuguese, translated from English by a native Portuguese speaker physician who is fluent in English and checked for local Portuguese nuances by Mozambican physicians.

The survey sample size was powered to achieve a 6% margin of error with 95% confidence, assuming 20% pain prevalence; to achieve this level of precision, 100 household respondents were targeted.² Subjects were adults in the household who consented to participate. Data analysis was done using the REDCap (Research Electronic Data Capture) application, Microsoft Excel and Stata.³⁶ Categorical variables were analysed using the Fisher exact test, and confidence intervals (CIs) were obtained using binomial exact calculations.

Results

Demographics and medical history

One hundred subjects agreed to participate in the survey, a 94% response rate. Of the 100 respondents, three surveys were not used for analysis due to being incomplete. Completed surveys corresponded to 39% at Central B (urban), 27% Bairro Chamanculo (sub-urban) and 33% Bairro Mahotas (peri-urban). Most respondents were female (64.9%) aged 18–30 years (54.6%) or 31–50 years (30.9%). The nearest clinic or hospital was <10 kilometre away for 81.3% of people. Mozambique has a nationalized system which is typically free, with a parallel private clinic system. In this survey, medical service was predominantly paid in cash (88.7%) (Appendix 1), with only 10.3% paid by the government, suggesting a large number of respondents used the private system. Most respondents had comorbidities (87%), with intestinal problems as the most common medical history (13.4%), followed by cardiac (9.3%) and gynaecological (9.3%). None reported having a mental health disorder. Twenty people (54.1%) were receiving treatment for the underlying medical condition, although this question had a poor response rate (34 out of 97 or 35%). Thirty-four percent of pain was due to a medical problem (Table 1).

Table 1.

Past medical history.

Characteristic	N/total responses	%
Medical issues	84/97	87
Intestinal problem	13	13.4
Heart problem	9	9.3
Gynaecologic problem	9	9.3
HIV, TB and other infection	7	7.2
Osteoarthritis	6	6.2
Rheumatologic disease	5	5.2
Diabetes	3	3.1
Congenital deformity	1	1
Neurologic disease	1	1
Cancer	0	0
Mental health issue	0	0
Other	30	31
Hypertension	7	7.2
Anaemia	6	6.2
Asthma	3	3.1
Sinusitis	3	3.1
Receiving treatment for underlying medical condition	20/37	54.1
Pain due to medical problem	33/97	34
If female
Natural births (live + stillbirth)	34/63	54
Caesarean sections	34/63	54
Pain started as a result of an injury or trauma	8/97	8.2
Vehicle accident	3	3.1
Work injury	1	1
Injury giving childbirth	1	1
Other	3	3.1
Experienced traumatic event in life	52/97	53.6

Chronic pain prevalence

In total, 32% (CI: 22.9–42.2) reported having daily pain currently, and 39.2% (CI: 29.4–49.6) reported having had daily pain that lasted more than 6 months in their lifetime (Figure 1 and Table 2). Fifty respondents stated they had pain the day of the interview (52% (CI: 41.2–61.8)) (Table 2). Out of these, respondents reporting mild (0–2 on a 10 scale), moderate (3–5) and severe (6–10) pain were 21 (42%), 24 (48%) and 5 (10%), respectively. Most respondents had intermittent pain (81%) rather than constant pain. Using the MBP, the majority reported low back pain (34.4%), with 16.7% reporting neck pain and 14.6% abdominal pain (Figure 2). On the BPI, most respondents reported pain in addition to minor every day pains (84.4% (CI: 75.5–91)). On a scale of 0–10, pain was rated 4.62 ± 3.55 at its worst in the past 24 hours, 1.79 ± 1.96 at its best, with an average of 2.89 ± 2.39 (Table 3).

Figure 1.

Reported pain.

Table 2.

Pain prevalence.

Characteristic	N/total responses	% (95% CI)
Lifetime every day pain that lasted more than 6 months	38/97	39.2 (29.4–49.6)
Current daily pain	31/97	32 (22.9–42.2)
How long have you had this pain
0–6 months	9/31	29
6 months–1 year	9/31	29
>1 year	13/31	41.9
Does this daily pain come and go?
Always there	6/31	6.2
Comes and goes	25/31	81
Pain today	50/97	52 (41.2–61.8)
Pain rating today	x/50
0 – No pain	1	2
1	2	4
2 – Mild, annoying pain	18	36
3	9	18
4 – Nagging, uncomfortable troublesome	6	12
5	9	18
6 – Distressing, miserable	2	4
7	1	2
8 – Intense, dreadful, horrible	2	4
9	0	0
10 – Worse possible, unbearable, excruciating	0	0
Pain Catastrophizing Scale (PCS)
When I feel pain I think
It’s terrible and I think it’s never going to get any better	17	17.5
I become afraid the pain will get worse	53	54.6
I can’t seem to keep it out of my mind	17	17.5
I keep thinking about how badly I want the pain to stop	28	28.9

CI: confidence interval.

Figure 2.

Michigan Body Map.

Table 3.

Brief Pain Inventory by Charles S Cleeland, PhD, Pain Research Group.

Characteristic	Average/total responses	% (95% CI) or SD
Throughout our lives, most of us have had pain from time to time (such as minor headaches, sprains and toothaches). Have you had pain other than these kinds of every day pain today?	81/96	84.4 (75.5–91)
Please rate your pain by circling the one number that best describes your pain at its worst in the last 24 hours (0 = no pain, 10 = worst pain)	4.62	±3.55
Rate your pain by circling the one number that best describes your pain at its least in the last 24 hours.	1.79	±1.96
Rate your pain by circling the one number that best describes your pain on the average.	2.89	±2.39
Rate your pain by circling the one number that tells how much pain you have right now.	1.18	±1.75
In the last 24 hours, how much relief have pain treatments or medications provided? Circle the percentage that most shows how much relief you have received.	62.46	±34.23
Number on a scale of 0–10 describing how, during the last 24 hours, pain has interfered with: (0 = no interference, 10 is complete interference)
General activity	0.56	±1.34
Mood	0.63	±1.42
Walking ability	0.63	±1.56
Normal work	0.39	±1.20
Relations with other people	0.20	±0.96
Sleep	0.42	±1.02
Enjoyment of life	0.33	±1.13

CI: confidence interval; SD: standard deviation.

Disability and central sensitization

Eight respondents had trauma as the initiating insult that caused pain (8.2%), with three of these from motor vehicle injuries, one due to work and one injury from childbirth (Table 1). Fifty two participants (53.6%) experienced a traumatic event in their life, with 45.2% having nightmares, anxiety, or fear related to this trauma. Using the PCS, a majority (54.6%) of respondents were afraid the pain will worsen, and 28.9% kept thinking about how they wanted the pain to stop (Table 1).

The WHODAS 2.0 indicated that most respondents (59.7%–89%) had no difficulty with activities of daily living and maintaining relationships (Table 4). Twelve (12%) felt at least moderately emotionally affected by health problems, and 22 (22.6%) had at least moderate trouble standing for longer than 30 minutes. Four responses indicated extreme disability. Respondents reported minimal effect of pain in the last 24 hours on their daily activities (Tables 3 and 4). In the WPI, 21% of respondents reported moderate-severe fatigue in the past week, 7% reported trouble thinking or remembering and 12% woke up tired. The majority (82.5%) reported a headache, 33% depression and 49.5% abdominal pain in the past 6 months (Table 5).

Table 4.

World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0).

In the past 30 days, how much difficulty did you have in:	None	Mild	Moderate	Severe	Extreme or cannot do
Standing for long periods such as 30 minutes?	58 (59.7%)	16 (16.5%)	18 (18.6%)	3 (3.1%)	1 (1%)
Taking care of your household responsibilities?	72 (74.2%)	17 (17.5%)	4 (4.1%)	4 (4.1%)	0
Learning a new task, for example learning how to get to a new place?	86 (88.7%)	8 (8.2%)	2 (2.1%)	1 (1%)	0
How much of a problem did you have joining in community activities in the same way as anyone else can?	77 (79.4%)	16 (16.5%)	2 (2.1%)	0	2 (2.1%)
How much have you been emotionally affected by your health problems?	69 (71.1%)	16 (16.5%)	8 (8.2%)	3 (3.1%)	1 (1%)
In the past 30 days, how much difficulty did you have in:	None	Mild	Moderate	Severe	Extreme or cannot do
Concentrating on doing something for ten minutes?	80 (82.5%)	11 (11.3%)	5 (5.2%)	1 (1%)	0
Walking a long distance such as a kilometer (or equivalent)?	69 (71.1%)	15 (15.5%)	11 (11.3%)	2 (2.1%)	0
Washing your whole body?	96 (99%)	1 (1%)	0	0	0
Getting dressed?	97 (100%)	0	0	0	0
Dealing with people you do not know?	90 (92.8%)	6 (6.2%)	1 (1%)	0	0
Maintaining a friendship?	87 (89.7%)	5 (5.2%)	4 (4.1%)	1 (1%)	0
Your day-to-day work/school?	86 (88.7%)	6 (6.2%)	3 (3.1%)	2 (2.1%)	0
Overall, in the past 30 days, how many days were these difficulties present?	3.698 ± 7.16
In the past 30 days, for how many days were you totally unable to carry out your usual activities or work because of any health condition?	0.521 ± 1.23
In the past 30 days, not counting the days that you were totally unable, for how many days did you cut back or reduce your usual activities or work because of any health condition?	1.19 ± 3.57

Table 5.

Widespread Pain Index and Symptom Severity Score from the American College of Rheumatology Diagnosis Criteria for Fibromyalgia.

Indicate your severity over the last week	No problem – 0	Mild problem – 1	Moderate – 2	Severe – 3
Fatigue	47 (48.5%)	29 (29.9%)	16 (16.5%)	5 (5.2%)
Trouble thinking or remembering	80 (82.5%)	10 (10.3%)	2 (2.1%)	5 (5.2%)
Waking up tired	56 (57.7%)	29 (29.9%)	8 (8.2%)	4 (4.1%)
Characteristic		N	%
In the past 6 months, have you had any of the following?
Pain or cramps in lower abdomen		48	49.5
Depression		32	33
Headache		80	82.5

Pain attitudes and behaviours

In total, 88% of respondents have sought treatment for pain, with the majority of therapy received from physicians (66%) and nurses (11.6%). The majority of treatment received was in the form of oral medication (93.1%). Most did not travel far for treatment (70.9% travelled <10 kilometre), with only 33.7% finding the treatment to be very effective. Respondents (99%) would take a pill if it would help their pain, with a large proportion willing to spend a significant amount of money for medication (49% would pay >US$40) and willing to travel long distances for help (55.2% would travel >40 kilometers). Most feel comfortable talking about their pain in the community (60.8%). Respondents would participate in group therapy (89.7%) (Table 6).

Table 6.

Access to pain treatment.

Characteristic	N	%
If there was a pill to help your pain, would you take it?	96	99
How much would you pay per month (or single course)?
<$US5	18	18.8
$US5–15	16	16.7
$US15–30	10	10.4
$US30–40	5	5.2
>$US40	47	49
How far would you travel to get it?
<10 km	23	24
10–20 km	11	11.5
20–30 km	4	4.2
30–40 k	5	5.2
>40 km	53	55.2
Do you (or would you) feel comfortable talking about your pain with others in your community?	59	60.8
Do you think the treatment of people’s pain is important?	95	97.9
Would you participate in group treatment to teach you how to move and cope/live with pain?	87	89.7
Have you ever sought treatment for your pain?	86	88.7
Who gave the treatment? (may choose more than one)
Physician	64	66
Nurse	10	11.6
Friend or family member	6	7
Local healer	8	9.3
You gave to yourself	7	8.1
Other (all responded ‘pharmacy’)	3	3.5
What was the treatment? (may choose more than one)
Nothing	1	1.1
Pill	81	93.1
Herbal therapy (medicine from a plant)	9	10.3
Mind- based therapy (meditation, breathing and counselling)	2	2.3
Procedure (injection and surgery)	6	6.9
Other (physiotherapy and glasses)	2	2.3

Discussion

The prevalence of chronic pain in this sample was 39.2% (CI: 29.4–49.6), similar to the previously described chronic pain prevalence of 34% in LMICs.⁵ A high percentage of respondents (52%) reported pain the day of the interview, suggesting chronic pain may be even higher than reported. Most pain was lower back pain, consistent with previous studies.²⁵ Our sample’s pain prevalence rates mimic previously published surveys of pain point prevalence in Nepal.³⁷ Both countries share a high rate of exposure to traumatic events,^38–40 which are linked to chronic pain.⁴¹ Mozambique has a low healthy life expectancy of 55.5 years, with most mortality due to communicable disease, especially malaria (28.8%) and HIV (26.9%).^42,43 HIV is associated with a high incidence of chronic pain in LMICs, which can be significantly improved with palliative care intervention.⁴⁴ Non-communicable diseases also contribute significantly to morbidity and mortality in Mozambique, with injuries accounting for 16.4% of disability-adjusted life years (DALYs), neuropsychiatric disorders 7.1% and malignant neoplasms 4.2%.^42,43 One-third of women and one-fourth of men between 15 and 49 years of age are victims of physical violence.⁴⁵ Coexisting post-traumatic stress disorder (PTSD) has been linked to more severe pain and associated disability.^46,47 Trauma and injury, HIV, mental illness and cancer may contribute to the prevalence of pain in Mozambique, exacerbated by limited options for diagnosis and treatment of pain.

Although 39.2% of respondents reported chronic pain, most respondents reported little difficulty with household responsibilities (74.2%), community activities (79.4%) or friendships (89.7%). This differs from the 78% of patients at the Maputo Pain Clinic who report pain interfering with daily activities and from previously reported studies where pain sufferers in 15 countries had activity limitations.^7,25,26,48 However, 22.6% of respondents reported trouble standing for more than 30 minutes, which is significant interference with activities. Furthermore, the 88% of respondents who sought treatment for pain suggests a higher degree of disability than reported. This may indicate variations in culture and that how questions are phrased may affect how patients respond, highlighting the need for continued standardization of validated metrics.

A significant portion of respondents had nightmares, fear or anxiety related to a previous trauma, with most reporting fears that the pain will get worse, and 29% feeling emotionally affected by health problems. A previous study in Maputo indicated the presence of depression and anxiety to be 33% and 17% in chronic pain patients, respectively.²² Although our study did not assess for depression and no respondent reported a mental health diagnosis, these numbers suggest that there may be a large proportion of undiagnosed mental health comorbidities. Human resource capacity to diagnose mental illness remains scarce in Mozambique with 10 psychiatrists in 2014, but increasing through task shifting.⁴⁹

Chronic pain is a complex disorder with physiological, psychological and cultural components. Comprehensive pain management through a multidisciplinary pain team includes multimodal treatment affecting multiple pain receptors, as well as mental and physical rehabilitation. Nerve blocks, physician therapy and movement strategies such as yoga, meditation and other adjunctive therapies may decrease the prevalence of chronic pain. In this study, although most pain sufferers were receiving some form of treatment, only 33.7% found the treatment to be very effective, suggesting that more comprehensive treatment may be warranted. Most respondents were receiving oral medication for pain management (93.1%), but few (12%) received multimodal pain management. Respondents indicated willingness to receive treatment, to travel long distances for treatment and to pay large sums for these, indicating the surveyed population would value further pain treatment services. Comprehensive pain treatment is warranted and wanted in Maputo. In addition, a focus on prevention of chronic pain may be very valuable in these LMICs where resources are limited.

Chronic pain should be treated by personnel trained in multimodal treatment and palliative care.⁵⁰ There is paucity of hospice and palliative care availability in Africa.⁵¹ As of 2016, Mozambique had limited resources available for the treatment of pain, with only two physicians and two nurses specifically trained in pain management. There were four pain clinics with nine rotating anaesthesiologists, one paediatrician and five rotating residents, with only three of these outside of the capital. The Maputo pain unit had two full-time physicians, three nurses, one psychologist and five support staff. In 2014, the pain unit saw 200–300 new patients. Drug shortages were common and complementary medicines such as amitriptyline and gabapentin were not available.⁵² Epidural and nerve blocks were offered at the clinic. Recent efforts to improve access to pain management include a national palliative care policy, training curriculum and increased government funding of palliative programmes.⁵²

Limitations of this study include small sample size that may not be generalizable to the whole country. Interviews were conducted during the day, which may bias results towards respondents who cannot work. Males may have been underrepresented. Given the personal nature of questions, some respondents may not have answered appropriately. For example, the HIV prevalence in Mozambique is 26.9%, whereas our sample’s prevalence was 7%, possibly showing respondents did not want to admit they carry the disease in a nonclinical setting. Some questions may have been misunderstood due to cultural interpretation. For example, 38 patients reported having ‘every day pain lasting more than 6 months’, but only 31 responded to the question separating the pain into 0–6 months, 6 months–1 year and >1 year (Table 2). Similar questions were repeated because previously validated surveys were used and the investigators did not want to alter these metrics. However, the investigators’ intention is to repeat this study nationally, and repetitive questions will be deleted to avoid misunderstandings.

Conclusions

The prevalence of chronic pain in the surveyed population in Mozambique is similar to the average for LMICs. Contributors may include communicable diseases, trauma, cancer, mental illness and limited access to pain management resources. Most respondents were prescribed oral medication by a physician for treatment. Trends in HICs suggest that multimodal pain management and multidisciplinary or holistic treatments may improve optimal pain control in LMICs. The international community is just beginning to measure and recognize the global burden of pain. Further studies on prevalence of pain and treatment modalities in Mozambique and other LMICs are needed.

Appendix

Appendix 1.

Demographics characteristics of survey participants.

Characteristic (N = 97)	N	%
Age (years)
18–30	53	54.6
31–50	30	30.9
>50	14	14.4
Gender
Male	34	35.1
Occupation
Work at home	22	22.7
Industrial labour	3	3.1
Service (cook, clean and repairs)	7	7.2
Merchant	8	8.2
Driver	2	2.1
Construction	4	4.1
Other	51	52.6
Student	32	33
Government	4	4
Distance travelled to nearest clinic/hospital
<10 kilometre	78	81.3
10–20 kilometre	14	14.6
20–30 kilometre	4	4.2
>30 kilometre
Method of payment for medical service
Cash	86	88.7
Credit or debit card	1	1
Paid by government	10	10.3
Barter/trade	0	0
Private insurance	2	2.1