Some Conclusions about Addiction from a Look at Some Numbers

Substance use disorders

Conclusion: Even when exposed to substances that can be used in ways that cause problems, most people don’t become addicted to them. Most people who develop addiction have problems already.

Note: For an individual, dependence, i.e. experiencing withdrawal symptoms without the substance, is not addiction, i.e. feeling compelled to persist in a behavior despite negative consequences.

Opioid use and opioid use disorder

  • 1 in 100 Americans, 12 and older, met the diagnostic criteria for opioid use disorder in 2015 (0.2 percent for heroin use disorder, and 0.8 for pain reliever use disorder). (Source)
  • 75% of people with prescription opioid addiction obtained the substance from a family member or a friend, not from a doctor. (Source)
  • 92% of people exposed to prescription opioids do not become addicted to them. (Source)
  • 3.6% of people who misuse prescription opioids try heroin. (Source)
  • According to a recent study, over 70% of pain patients who developed opioid use disorder had a prior history of psychoactive drug use. (Source reported here)
  • Heroin addiction is more than three times as common in people making less than $20,000 per year compared to those who make $50,000 or more. (Source)
  • From a 2017 study for approximately 70% of the patients sampled: “Our results suggested that self-treatment of co-morbid psychiatric disturbances is a powerful motivating force to initiate and sustain abuse of opioids and that the initial source of drugs – a prescription or experimentation – is largely irrelevant in the progression to a SUD [substance use disorder].” (Source)
  • Since 2011, overdose deaths from prescription opioids alone are down. Deaths from synthetic opioids, like fentanyl, are up.

Conclusion: Even when opioids – created by pharmaceutical companies – are prescribed by doctors, most people don’t become addicted to them. And socioeconomics matter.

Numbers from the New River Valley of Virginia

  • 16,000+ people in the New River Valley have alcohol and other drug problems. (Source)
  • According to local sources*, in 2015, the New River Valley had 34 cases of opioid overdose. Among those were 3 cases of heroin overdose, and 5 cases of fentanyl overdose.
  • According to local sources, in 2015, 11.4% of New River Valley high school students had misused prescription drugs in the past 30 days to “get high.” The national average is 3.2%. In 2015, 5.6% New River Valley high school students had used heroin at least once in their lifetimes. The national average is 3.2%.

Yearly death total statistics that make – and don’t make – the news in the U.S.

  • Marijuana overdose deaths: 0 (Source)
  • Terrorism-related deaths, U.S. citizens, overseas and domestic: 32 (2014: Source)
  • Opioid-related deaths: 35,000 (2015: Source)
  • Gun-related deaths: 35,000 (2014: Source)
  • Alcohol-related deaths: 88,000 (2015: Source)
  • Obesity-related deaths: 300,000 (Source)
  • Tobacco-related deaths: 480,000 (Source)

Conclusion: The selection of subjects covered by the media may not be due to high death rates.

*Grateful acknowledgement is made to New River Valley Community Services for synthesizing data from multiple sources for this post.

Laurel Sindewald contributed to the research for this post.

This post was prepared as part of a packet of handouts for a talk on the opioid epidemic by Anne Giles for the Montgomery County, Virginia Democratic Party on 8/17/17.

Last updated 10/26/17

This content is for informational purposes only and is not a substitute for medical or professional advice. Consult a qualified health care professional for personalized medical and professional advice.

Medication-Assisted Treatment for Opioid Use Disorder – Infographic

Medication-assisted treatment (MAT) with methadone and buprenorphine is the only known treatment – not abstinence, not counseling, not 12-step approaches – to cut death rates from opioid use disorder by 50-70% or more.

“The principle behind MAT is this: Because opioid addiction permanently alters the brain receptors, taking the drug completely out of someone’s system can leave them less able to naturally cope with physical or emotional stress…”

U.S. federal officials decry opioid misuse as a public health crisis, yet federal rules limit access to the only known effective treatment. Due to federal restrictions, few health care professionals are approved to dispense methadone or prescribe buprenorphine. Further, laws dictate how many patients those few can prescribe and, increasingly to whom, in what form, and how much medication can be prescribed. People suffer, even die, on wait lists to receive medication. In contrast, countries that effectively address their overdose crises, loosen, not tighten restrictions. Indeed, Stefan G. Kertesz, M.D. states, “The dominant priority should be the assurance of subsidized access to evidence-based medication-assisted treatment for opioid use disorder.”

In the event that lack of understanding of MAT may be contributing to restricting access to it, we offer this simple infographic explaining medication-assisted treatment (MAT) for opioid use disorder (OUD).

Medication-Assisted Treatment for Opioid Use Disorder

Medication-assisted treatment (MAT) does not replace one addiction with another. It creates stability by treating the medical condition of addiction. Currently, effective medications for addiction create dependence – negative symptoms from withdrawal if doses are discontinued – but not addiction, which, per its definition, involves continued use despite negative consequences.

Tolerance occurs when people’s bodies adapt to a drug over time, responding less and less to the same dose. In order to have an effect, doses must keep increasing for people to continue to get high, or to keep from getting sick. When people use opioids regularly, therefore, they tend to use more and more of the drug over time. Even if they stop using, then return to use, they are at risk for overdose because illegal drugs are not monitored, and so they may be cut with much more powerful drugs, such as heroin with fentanyl.

(Recent increased rates in drug overdose are due to heroin and illicit fentanyl, not prescription pain medications. Reported deaths in 2015 from opioids by prescription account for under 15%. Of those addicted to prescription pain medications, 75% received them from a family member, friend or dealer, not through medication prescribed to them.)

When people with OUDs enter abstinence-based treatment, or otherwise stop using, their tolerance drops. They may not know their tolerance is diminished, or they may not know how much it has decreased, and if they take an opioid at the high dose they were once used to, they are likely to overdose and die.

People who are given MAT for OUD take an opioid (buprenorphine or methadone) at a consistent dose, which effectively stabilizes them. Once stabilized on an effective dose, they do not experience withdrawals, cravings, or highs. They can provide child or dependent care, hold a job, adhere to treatment, and comply with the law.

In contrast, people who are not on MAT will experience withdrawal symptoms and strong cravings, especially when under stress. People with OUDs permitted only abstinence-based treatment are at high risk for all of the same problems people with untreated addiction are at risk for: recidivism and crime, unemployment, contracting and transmitting diseases, overdose and hospitalization, and fatal overdose. Up to 90% of people with opioid use disorder relapse when not on medication-assisted treatment.

Maintenance may need to be long-term, or even life-long, because while addiction lasts, people who terminate maintenance treatments are at elevated risk for fatal overdose.

The US opioid epidemic has changed profoundly in the last 3 years, in ways that require substantial recalibration of the US policy response…Heroin and fentanyl have come to dominate an escalating epidemic of lethal opioid overdose, whereas opioids commonly obtained by prescription play a minor role, accounting for no more than 15% of reported deaths in 2015…The observed changes in the opioid epidemic are particularly remarkable because they have emerged despite sustained reductions in opioid prescribing and sustained reductions in prescription opioid misuse. Among US adults, past-year prescription opioid misuse is at its lowest level since 2002. Among 12th graders it is at its lowest level in 20 years. A credible epidemiologic account of the opioid epidemic is as follows: although opioid prescribing by physicians appears to have unleashed the epidemic prior to 2012, physician prescribing no longer plays a major role in sustaining it. The accelerating pace of the opioid epidemic in 2015–2016 requires a serious reconsideration of governmental policy initiatives that continue to focus on reductions in opioid prescribing. The dominant priority should be the assurance of subsidized access to evidence-based medication-assisted treatment for opioid use disorder. Such treatment is lacking across much of the United States at this time. Further aggressive focus on prescription reduction is likely to obtain diminishing returns while creating significant risks for patients.
– Stefan G. Kertesz, M.D.

People with opioid addiction can live full lives as family members and citizens. MAT benefits the general public health, employers, law enforcement, taxpayers, and the human beings who need our help.

Infographic by Laurel Sindewald. A printable .pdf version is here.

Laurel Sindewald contributed to this article.

Related reports on addictions treatment, addictions recovery, and addictions policy from Handshake Media, Incorporated:

This post was last updated 5/4/17.

After the Shootings: Community Violence, Collective Trauma and Addiction

It’s Fourth of July in a small town in America, and people of all ages and races, classes and creeds are gathered in a park for the fireworks display. People in lawn chairs are chatting over drinks. Here and there people are stretched out on old blankets or beach towels, reading or scrolling through Facebook. Kids are tearing up and down the playground equipment, playing with glowsticks or laughing at gravity on the swings. Everywhere the people breath deep and smell wafting barbecue above fresh grass, spiced with smoke from exhausted sparklers.

The community seems happy with the warmth of summer, and nothing around could show you the difference between this town and so many other small towns in America. But the people of this town have been changed. One April ten years ago, someone walked onto the local college campus, entered a building and locked the doors, killed 32 people and wounded 17 before committing suicide.

About 500 people are gathered to see the fireworks. Because of that violent event, research suggests that 75 of those gathered are suffering symptoms of trauma severe enough to be diagnosed with post-traumatic stress disorder (PTSD).

Statue on Virginia Tech Campus

Photo: “The Garden Sprite” statue on Virginia Tech Campus in the Hahn Horticulture Garden, a rendition of Frank Lloyd Wright’s “Maid in the Mud”

Mass shootings are one type of community violence, a term that includes terrorist attacks, riots, gang wars, workplace assaults, torture, bombings, war, and many other acts of violence. Community violence has far-reaching effects very different from some other forms of trauma, and impacts people who may not have been directly present at the violent event.

“Several aspects of community violence make it different from other types of trauma. Although there are warnings for some traumas, community violence usually happens without warning and comes as a sudden and terrifying shock. Because of this, communities that suffer from violence often experience increased fear and a feeling that the world is unsafe and that harm could come at any time. Although some traumas only affect one individual or a small group of people, community violence can permanently destroy entire neighborhoods. Finally, although some types of trauma are accidental, community violence is intentional, which can lead survivors to feel an extreme sense of betrayal and distrust toward other people.”
– Hamblen and Goguen, US Department of Veterans Affairs, 2016

An estimated 1015% of people who experience community violence report severe PTSD symptoms afterward. (Researchers estimate that 15-30% of people in the Blacksburg community experienced PTSD symptoms after the Virginia Tech shootings.)

Risk factors include female gender, proximity to the violence, knowing victims of the violence, pre-existing psychological conditions, emotion regulation difficulties, anxiety sensitivity, and low social support (Lowe and Galea, 2017; Bardeen et al., 2013; Grills-Taquechel et al., 2011; Stephenson et al., 2009; Scarpa et al., 2006Norris et al., 2002; North et al., 1994). The mental health effects include “psychological distress and clinically significant elevations in posttraumatic stress, depression, and anxiety symptoms in relation to the degree of physical exposure and social proximity to the shooting incident” (Schultz et al., 2014).

Researchers report that people in the surrounding community may also experience these effects, even if they did not witness the violence first-hand. For example, people who experience stronger emotional reactions, regardless of proximity, are at higher risk for later PTSD symptoms.

“For some people it’s water off a duck’s back. Some people are drowning.”
– Anne Giles, private interview

People who were unable to contact or locate loved ones during the Virginia Tech shootings were subject to trauma symptoms, even years later, as Victoria Sagstetter discovered.

“As an English teacher, she found herself re-reading student poems that seemed unusually dark, looking for the kind of clues Cho left behind in his writings before his killing spree.”
Jacob Demmitt of The Roanoke Times

After the Virginia Tech shootings, 4.5 years went by before some people sought treatment for trauma. Among the mental health effects people experience after trauma are substance use disorders (SUDs) and other addictions. The connection between trauma and addiction is well-documented, and current research indicates that trauma is a causal factor leading to addiction.

“Consistent with the self-medication hypothesis, the theory that people use substances to cope with psychological distress, PTSD tends to precede and predict SUD.”
Trauma and Addiction: Common Origins and Integrated Treatment

About 34% of people diagnosed with PTSD also suffer from addiction, and about two-thirds of people with addiction have experienced past trauma. Therefore, of the revelers at the Fourth of July celebration, an estimated 25 are at risk for developing addiction due to the event of community violence alone.

If people have experienced other traumatic events in their lives, as 1 in 4 American children have (CDC 2014; Felitti et al., 1998), they will be at even higher risk for addiction. For example, women who had experienced sexual trauma prior to the Virginia Tech shootings reported significantly more depressive symptoms, shooting-related PTSD, and lower belief in benevolence and family support.

“Maia Szalavitz, in her book Unbroken Brain, reports that, ‘Even just one extreme adversity – like losing a parent or witnessing domestic violence – before age 15 doubles the odds of substance use disorders, according to a study of the entire Swedish population’ (Unbroken Brain, 65).”
– Trauma and Addiction: Common Origins and Integrated Treatment

The estimated 75 people with trauma and 25 people with addiction watching the fireworks display are therefore a low estimate of the true risk for trauma and addiction in this small American town. Trauma and addiction are already very likely to occur, and an event of community violence such as the mass shootings in this community means trauma and addiction are almost certain to happen.

After community violence happens, everyone should be screened for trauma. (If resources are limited, those exhibiting risk factors should be prioritized, as should those with pre-existing disorders that put them at higher risk for addiction.) Community members must be aware that their friends and family and neighbors are at risk for developing addiction, and that if they do, they need treatment not tough love.

Some variables are known to reduce the risk of PTSD (and therefore addiction) after community violence. If a person has a belief in his or her ability to handle the trauma, i.e., self-efficacy, he or she will be less likely to experience PTSD symptoms. After community violence, in-person social support is known to mitigate resource loss (social or physical) and compensate for low levels of self-efficacy, reducing PTSD risk (Warner et al., 2015; Hawdon et al., 2012; Littleton et al., 2009). Specifically, sharing about thoughts and emotions with others may attenuate PTSD risk, but sharing bare facts will likely not help, and may increase the risk of developing PTSD.

“Thus, it is argued here that efforts to reduce risk and resource inequities, engage local people in mitigation, create organizational linkages, boost and protect social supports, cultivate trusted and responsible information resources, and enhance decision-making skills will augment more specific intervention efforts to promote safety, calming, efficacy, hope, and connectedness in the aftermath of mass trauma.”
Norris and Stevens, 2007

People in a community have a choice to make after community violence. What they choose determines how long and how badly the violence continues to affect their community. People who come together, in person, to share thoughts and feelings about the violence during and after it occurs are less likely to suffer from PTSD and addiction in the future. People who continue to come together, who recognize that some of their friends and neighbors will inevitably be suffering, and who affirm their own and others’ abilities to cope with the violence will be more likely to heal.

Let’s say you’re sitting on a towel on the 4th of July in that small town, surrounded by your friends and neighbors as the fireworks begin. You share, together, openly and shamelessly, about the trauma or addiction with which you may struggle.  In a special place created by “safety, calming, efficacy, hope, and connectedness,” you may start to feel a little bit better.

Author’s note: To estimate the number of people in my imagined 500 likely to develop PTSD, I reviewed the literature. One source estimated 10-15%, another estimated 15.4%, and a third estimated 15-30% of people who experience a mass shooting (directly or indirectly) develop PTSD. I chose 15% as a conservative estimate, because I do not have access to the data sets for each of these three sources. I multiplied 500 by 0.15, and arrived at my estimate of 75 people at high risk for PTSD in my imagined gathering of 500.

To estimate the number of people in my imagined 500 likely to develop addiction, I again reviewed the literature and found that 34% of people diagnosed with PTSD also have addiction. I multiplied 75 by 0.34 to arrive at 25 estimated people with addiction in the gathering of 500. One may also multiply 0.15 by 0.34 to obtain 0.05, or 5%, and multiply the full 500 by 0.05, again equaling 25 people at risk for addiction.

In a gathering of 500 people who experienced community violence, 15% are at risk for PTSD, or 75, and 5% are at risk for addiction, or 25 – due to the event of community violence alone.

PDF of Research Excerpts

Photo: Laurel Sindewald, statue, Hahn Horticulture Garden, Virginia Tech

A personal note from Anne Giles, added 4/5/17: I am one among a likely cohort of 300 who developed addiction in Blacksburg, Virginia after the Virginia Tech shootings. Of the 40,000+ people living in Blacksburg in 2007, research predicts 15% of them would develop PTSD. That would be 6,000. Of that 6,000, research predicts 5% would develop addiction. That’s 300.

Handshake Media maintains a list of addiction recovery resources for people living in the Blacksburg, Virginia area.

This content is for informational purposes only and is not a substitute for medical or professional advice. Consult a qualified health care professional for personalized medical and professional advice.

On Counseling and Medication-Assisted Treatment

Facing Addiction in America: The Surgeon General’s Report on Alcohol, Drugs, and Health categorizes substance use disorders, a.k.a. addictions, with “chronic and expensive medical illnesses” such as heart disease, diabetes, cancer, hypertension, asthma, arthritis, and chronic pain. The first order of treatment for these illnesses is medical care, primarily through medication. Patients with these illnesses are not required to receive multiple counseling sessions prior to receiving medication.

Challenging Myths about Medication Assisted Treatment for Opioid Use DisorderCounseling and support groups do exist for assistance with these illnesses. But counselors help patients comply with the course of treatment determined by their physicians. Support groups can help people live life while having the illness. But counseling and support aren’t replacements for medication and medical care. Except in addictions treatment.

Counseling is not medical care for a medical condition. We need to give people with the medical condition of substance use disorders (SUDs) medical care first – which may or may not include medication-assisted treatment (MAT) – but which may include stabilizing medical treatment for co-occurring mental and physical illnesses. Confusing medical care with counseling results in what the New England Journal of Medicine reported in December 2016: “Despite the demonstrated efficacy of maintaining abstinence by treating patients with opioid agonists, patients can remain on clinic waiting lists for months, during which time they are at risk of premature death.”

With regard to MAT and counseling, my conclusions from reading the research are that for most people most of the time, counseling has no significant effect on abstinence among those receiving MAT for SUDs, whether for opioid use disorder or other SUDs. Counseling may have a larger effect on those with co-occurring SUDs and mental disorders, but those effects would occur over time. People with the medical condition of SUDs do not have time for the possibility that counseling might be effective. They are at higher risk of dying as long as they are not on opioid maintenance medication. They need medical care stat.

From the Surgeon General’s report, page 4-21: “Nevertheless, multiple factors create barriers to widespread use of MAT. These include provider, public, and client attitudes and beliefs about MAT…”

“Buprenorphine is an effective medication in the maintenance treatment of heroin dependence, retaining people in treatment at any dose above 2 mg, and suppressing illicit opioid use (at doses 16 mg or greater) based on placebo-controlled trials…Methadone is superior to buprenorphine in retaining people in treatment, and methadone equally suppresses illicit opioid use.”
Cochrane Database of Systematic Reviews, 2014

We need to shift our policies from belief-based practices to what the science reports is best practices addictions treatment for people with substance use disorders: medical care first. Then, we can work on secondary, individualized assistance, perhaps counseling, case management, support groups, or support services according to each client’s case.

. . . . .

“In fact, no rigorous study has ever been able to show that the addition of psychosocial services to opioid agonist therapy alone improves outcomes in the treatment of opioid use disorder.”
Lean Forward, Harvard Medical School, May 2017

“Despite the demonstrated efficacy of maintaining abstinence by treating patients with opioid agonists, patients can remain on clinic waiting lists for months, during which time they are at risk of premature death. The use of interim treatment with buprenorphine without formal counseling while patients remain on waiting lists may mitigate this risk during delays in treatment.”
New England Journal of Medicine, December 2016

“Unfortunately, despite decades of research, it cannot be concluded that general group counseling is reliably effective in reducing substance use or related problems.”
Surgeon General’s Report, November 2016

“The commonly held belief that opioid agonist treatment alone is inferior treatment to such treatment combined with psychosocial treatment (which many will understand to mean counseling) is not supported by the research evidence and it results in limitations on the use of these effective medications.”
Journal of Addiction Medicine, July/August 2016

“[T]here is little empirical evidence suggesting which psychosocial treatments work best in conjunction with medication-assisted treatment as there are relatively few studies comparing the differential effectiveness of various psychosocial approaches (eg, CM [contingency management], MI [motivational interviewing]) for individuals receiving medications for the treatment of opioid addiction.”
Journal of Addiction Medicine, March/April 2016

“Patients who received only psychological support for opioid dependence in England appear to be at greater risk of fatal opioid poisoning than those who received opioid agonist pharmacotherapy.” – Addiction, November 2015

“Clinical trials show that opioid agonist therapy (OAT) with methadone or buprenorphine is more effective than behavioral treatments, but state policymakers remain ambivalent about covering OAT for long periods…OAT is associated with lower total healthcare expenditures compared to other forms of behavioral treatment for patients with opioid addiction.”
– Journal of Substance Abuse Treatment, October 2015

“Supplementing standard medical management with cognitive-behavioural therapy did not further reduce opioid use or promote abstinence among primary care patients being maintained on buprenorphine.”
Drug and Alcohol Findings, 2013

“For the considered outcomes [retention in treatment and use], it seems that adding any psychosocial support to standard maintenance treatments do not add additional benefits.”
Cochrane Database of Systematic Reviews, 2011

“Consistent with results from a previous study of predominantly heroin-dependent patients receiving buprenorphine-naloxone in a primary care setting, individual drug counseling did not improve opioid use outcomes when added to weekly medical management visits.”
Archives of General Psychiatry, 2011

“We conducted a 24-week randomized, controlled clinical trial with 166 patients assigned to one of three treatments: standard medical management and either once-weekly or thrice-weekly medication dispensing or enhanced medical management and thrice-weekly medication dispensing…All three treatments were associated with significant reductions from baseline in the frequency of illicit opioid use, but there were no significant differences among the treatments.”
New England Journal of Medicine, July 2006

The Comprehensive Addiction and Recovery Act (CARA) requires that an MAT treatment provider be able to provide or refer patients to counseling. Other than this wording in CARA – “appropriate counseling and behavioral therapies” – neither mentions the type, length or number of sessions of counseling required. Wisely, for an individual, the law does not require him or her to receive counseling in order to receive MAT. Unwisely, in Virginia, Medicaid patients are required to receive SUD counseling in order to be covered for either methadone or buprenorphine.

Image: Challenging the Myths about Medication Assisted Treatment (MAT) for Opioid Use Disorder (OUD) from The National Council for Behavioral Health

Laurel Sindewald contributed to the research for this report.

Last updated 7/30/2017.

Addiction Is Not a Choice

Facing Addiction in America: The Surgeon General’s Report on Alcohol, Drugs, and Health makes the current discussion we have about addiction, i.e. helping people with addiction “make better choices,” “make better decisions, or “understand the consequences of their behavior” – whether coerced through drug court or supported through counseling – well-meaning but, unfortunately, irrelevant. Addiction compromises all of those functions in the brain.

The brain no longer does what it needs to do in a person with addiction, just as a pancreas no longer does what it needs to do in a person with diabetes. Medical care for a brain illness – not tough lovearrest, commitment to a mental institution, or recovery support services – needs to be the first order of treatment, including assessment for suitability for medications. Medications need to be prescribed by qualified medical professionals, not by non-qualified court officials or lawmakers.

Surgeon General: I'll stand up for recovery with you

We would rush our neighbor with acute diabetes to the doctor. Why don’t we rush our citizens with an acute brain disease to the doctor as well? Because, at essence, contrary to scientific evidence, we still believe addiction is a choice. We believe that if people with addiction could just see the errors of their ways and would work hard on those errors (rather than be lazy, immoral, selfish, self-indulgent, or inadequately faithful or spiritual), addiction would go away.

Working hard on one’s ways may help one live a better life. Might working hard to be a better person alter the brain in targeted ways that reverse or  ameliorate addiction? Possibly. Neuroscience research may ultimately support that. Certainly on ways to live a better life, many people, both with and without academic or medical credentials, can offer helpful guidance. But, for now, what we know is that addiction is a medical condition, 1 in 7 Americans is expected to get it, a person dies of a drug overdose every 19 minutes in the U.S., and only a fraction of those who need help are receiving it. For this dire medical condition, insufficiently treated such that a public health crisis has occurred, medical care is an imperative.

Do no harm” is a principle of health care. By stubbornly holding onto the concept of “choice” – in spite of the data that says we’re simply wrong to do so – we’re harming, even killing, our own citizens when we require them to, at essence “be better and do better,” rather than provide them with medical care.

May the Surgeon General’s report inform and direct the treatment we provide our fellow citizens struggling with the grave and dangerous medical condition of addiction.

. . . . .

On 11/17/16, the U.S. Department of Health & Human Services issued Facing Addiction in America: The Surgeon General’s Report on Alcohol, Drugs, and Health.

Here’s a link to the YouTube video of the 3-hour presentation of the Surgeon General’s report.

Here’s Maia Szlavitz’s commentary on the Surgeon General’s report.

This post is an updated and expanded version of my letter to the editor about the Surgeon General’s report published by the Roanoke Times on 12/13/16.